Proefschrift Biesheuvel

Keeping the clouds away

Prevention of relapse and recurrencein major depressive disorder

Karolien Biesheuvel - Leliefeld

Uitnodigingvoor het bijwonen van de openbare verdediging van

mijn proefschrift getiteld

Keeping the cloUds awayPrevention of relapse and

recurrence in major depressive disorder

Vrijdag 15 april 2016, 11.45 uur Auditorium

Vrije Universiteit De Boelelaan 1105, Amsterdam

Na afloop van de verdediging bent u van harte welkom in

café The Basket op de campus van de VU

Karolien BiesheuvelMeester Sixlaan 261181 PK Amstelveen

06 - [email protected]

ParanimfenPaul Leliefeld

[email protected] - 24884706

Karianne van der Weijden [email protected]

06 - 24594099

Karolien Biesheuvel - LeliefeldKeeping the clouds away

Depression is among the most disabling disorders and negatively affects many aspects of life. It is associated with a high risk of recurrence. Of all people with a first episode, more than half experiences such a recurrence. Treating depression in the Netherlands costs almost a billion euro (€966 million) per year. These substantial economic consequences of depression are mainly due to its recurrent nature. An important potential area of improvement in care for people with depression is the prevention of recurrence. The most commonly used strategy is the continuation of antidepressant medication. However, due to possible side effects and non-adherence issues, continuing antidepressant medication may not always be the preferred option. Psychological interventions could be a valuable alternative. In order to improve clinical outcomes it is highly relevant to study the prevention of recurrent depression using psychological interventions in primary care. Therefore, Karolien Biesheuvel and her colleagues carried out several studies, including a randomised controlled trial in The Netherlands. The research questions were:

1) What is the burden of disease of recurrent depression compared to single episode depression?

2) What is the effectiveness of existing psychological interventions compared both to usual care and the continuation of antidepressant medication, to prevent relapse and recurrence in recurrent depression?

3) What is the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care?

4) What is the (cost-)effectiveness of a psychological self-help intervention in primary care, for the prevention of relapse and recurrence in recurrent depression?


Prevention of relapse and recurrence in major depressive disorder

Karolien Biesheuvel-Leliefeld

Keeping the clouds awayPrevention of relapse and recurrence in major depressive disorder

The studies presented in this thesis were performed at the EMGO+ Institute for Health and Care Research and the General Practice of the VU University Medical Centre, Amsterdam, The Netherlands. The studies in this thesis were financially supported by the Netherlands Organisations for Health Research and Development (ZonMW, No 80-82310-97-11087), The Hague. Financial support for this thesis was kindly provided by the EMGO+ Institue for Health Care and Research and the General Practice of the VU University Medical Centre.

COLOFONAuthor: Karolien BiesheuvelCover design: Karolien & GildeprintLayout and printed by: Gildeprint - Enschede ISBN: 978-94-6233-251-5

©2016, Karolien Biesheuvel-Leliefeld, Amsterdam, The Netherlands All rights reserved. No part of this thesis may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise without prior permission of the author.

VRIJE UNIVERSITEIT


Prevention of relapse and recurrence in major depressive disorder

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aande Vrije Universiteit Amsterdam,

op gezag van de rector magnificusprof.dr. V. Subramaniam,

in het openbaar te verdedigenten overstaan van de promotiecommissie

van de Faculteit der Geneeskundeop vrijdag 15 april 2016 om 11.45 uurin het auditorium van de universiteit,

De Boelelaan 1105

doorKarolina Elisabeth Margaretha Biesheuvel-Leliefeld

geboren te Heerlen

promotoren: prof.dr. H.E. van der Horst prof.dr. H.W.J. van Marwijk prof.dr. H.F.E. Smitcopromotor: dr. D.J.F. van Schaik

What if I fall?Oh but my darling, what if you fly?

Erin Hanson

Voor mijn lieve ouders,zoveel dank!

CONTENTS

Chapter 1 General introduction 11 Chapter 2 Non-fatal disease burden for subtypes of depressive disorder: 27 population-based epidemiological study

Chapter 3 Effectiveness of psychological interventions in preventing 45 recurrence of depressive disorder: meta-analysis and meta-regression

Chapter 4 Cost-effectiveness of preventing depressive recurrences by psycho- 75 logical interventions; a population health economic modelling study

Chapter 5 Cost-effectiveness of nurse-led self-help for recurrent depression in the 89 primary care setting: design of a pragmatic randomised controlled trial

Chapter 6 Effectiveness of supported self-help for recurrent depression: 107 a randomised controlled trial in primary care

Chapter 7 Cost-effectiveness of supported self-help for recurrent depression 127 in primary care

Chapter 8 General discussion 147

Summaries in English / Dutch

Acknowledgement 189

List of publications 199

About the author 205

173/181

1General introduction

R1R2R3R4R5R6R7R8R9

R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

Chapter 1

12

R1R2R3R4R5R6R7R8R9R10R11R12R13R14R15R16R17R18R19R20R21R22R23R24R25R26R27R28R29R30R31R32R33R34R35R36R37R38R39

General introduction

13

1Major depressive disorder (MDD) is a prevalent mental disorder1, associated with a high risk of relapse and recurrence2, and with frequently incomplete remission between episodes3–5. It is considered to be among the most disabling illnesses6, and negatively affects many aspects of life7–9. MDD was the second leading cause of the total burden of disease in 201010, accounting for 8.2% of the global burden of disease, behind low back pain. Within the group of mental and substance use disorders, 41% of the global burden of disease was caused by MDD11. Treating MDD in the Netherlands costs almost a billion euro (€966 million) per year12. Decision makers seem to have become increasingly aware of the disease burden and healthcare costs associated with MDD due to the current economic down-turn and an expenditure for mental health care that is unlikely to be sustainable13. However, decision makers in health care are confronting a host of challenges because the healthcare environment, both in the Netherlands and globally, is changing rapidly. There are many catalysts for change, including government regulations, increased utilization of health care, patients’ expectations, competition, declining reimbursement and technological developments. Therefore, the question decision makers have to ask themselves is how healthcare for major contributors of disease burden and healthcare costs, like MDD, is to be organised; how it is to be channelled to the right people, and how the right services can be delivered at the right time, at the right place and at bearable costs.

An important part of the healthcare for MDD is the prevention of relapse and recurrence. The proactive management that is currently mostly used is the continuation of antidepressant medication (ADM). This may not be most optimal strategy14–19. Another approach to the prevention of relapse and recurrence in MDD, is the use of preventive psychological interventions. Research demonstrates that these interventions can be effective in reducing the risk of relapse and recurrence20–23.

There are several knowledge gaps regarding recurrent MDD and the prevention of relapse and recurrence by psychological interventions in people with a history of depression. First of all, only little attention is paid in literature to what part of the burden of disease of MDD is attributable to single episode depression compared to recurrent depression. Secondly, the effectiveness of all types of currently available, preventive psychological interventions compared to treatment-as-usual (TAU) and ADM is unknown. Thirdly, as the majority of the psychological interventions is offered in secondary care, often relying on intensive use of therapist’s time, these interventions are costly. Also, therapists are scarce. A minimally supported self-help intervention may help to overcome these problems and has already proved as effective as face-to-face treatment in acute depressed patients24. The integration of such an intervention into current longitudinal primary care systems, would fit best with the recurrent character of MDD. Besides, the prevalence of patients with MDD or depressive feelings in primary practice is high, around 21%25. So far, it is unknown if supported self-help for recurrent MDD, offered in primary care, is more effective than usual care. Finally,

R1R2R3R4R5R6R7R8R9


Chapter 1

14

it is the combination of data on effects and costs of interventions that will inform decision makers, who are forced to allocate limited resources to interventions that maximise cost-effectiveness. Data on both effects and costs is often lacking.

In this thesis we try to close these knowledge gaps by answering following research questions:

1) What is the burden of disease of recurrent depression compared to single episode depression?2) What is the effectiveness of existing psychological interventions compared both to usual care and the continuation of ADM, to prevent relapse and recurrence in recurrent depression?3) What is the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care?4) What is the (cost-)effectiveness of supported self-help in primary care, for the prevention of relapse and recurrence in recurrent depression?

In this chapter we present background information relevant to recurrent MDD. Furthermore, we present the main concepts and assumptions that motivate the aims and the design of our studies.

Major depressive disorder; definition and epidemiologyAccording to the most recent Diagnostic and Statistical Manual of Mental Disorders (DSM-5)26, for a diagnosis of MDD, at least five of the symptoms, mentioned in Table 1 have to be present during the same 2-week period and represent a change from previous functioning. At least one of the symptoms is either depressed mood or loss of interest or pleasure. In other words, MDD is not just having a bad day or a bad mood. It is a lasting and overwhelming sad feeling that does not resolve after a few weeks and, more importantly, interferes with daily life. It is a serious illness that is common, unfortunately, affecting no less than 18,7% of the population on a lifetime basis and 5,2% of the population over 12 months1. MDD has a detrimental impact on social, family and professional role functioning and is a leading cause of burden of disease10,11,27,28.



15

1Table 1. DSM-5 criteria for major depressive episode1

Depressed moodLoss of interest or pleasure

Significant weight loss or gainInsomnia or hypersomniaPsychomotor agitation or retardationFatigue or loss of energyFeelings of worthlessness or excessive or inappropriate guiltDiminished ability to think or concentrate, or indecisiveness andRecurrent thoughts of death or suicidal ideation

1 American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders (DSM-V)

The course of depressionMDD is characterized by a dynamic course29. To describe the course of depression, the operational criteria of Frank et al. (1991)30 are often cited (Figure 1). According to these criteria the course of depression is described as a series of disease stages in which a person can move from a symptom-free stage, to a stage characterized by some symptoms but not meeting the diagnostic criteria, to a stage with the full-blown disorder, after which the person can go into remission. When a person stays in remission for a minimum of six months, he or she is considered to be recovered. Subsequently, a relapse is defined as a depressive episode that occurs during remission and before recovery while a recurrence is defined as a depressive episode that occurs after recovery. Unfortunately, definitions of relapse, recurrence, remission and recovery are generally not being used in a consistent manner in the literature. Therefore, caution should be paid when interpreting results of studies on recurrent MDD.

Figure 1. Overview of criteria for remission, relapse, recurrence and recovery modified after Tohen et al.31

R1R2R3R4R5R6R7R8R9


Chapter 1

16

Recurrent depression; a chronic disorder?The risk of relapse and recurrence of MDD is high2. A long term follow-up study in the general population demonstrated that 38% of all people experienced a recurrence over a ten-year period33. The Netherlands Mental Health Survey and Incidence Study (NEMESIS), conducted in the general population, showed a similar percentage34. The National Institute of Mental Health Collaborative Study of the Psychobiology of Depression (NIMH) demonstrated the probability of recurrence to be 67% after 10 years35,36 and 85% after 15 years37. Other studies present recurrence rates ranging from 26.8%-33.5% over two years and 40%-60% over five years38–41. While the course of depression is portrayed somewhat inconsistently in the literature, the general picture is as follows: of all people with a first episode of MDD, at least 45% experience recurrences42, typically with seven to eight depressive episodes over the course of their life42 and spending as much as 21% of their lifetime in a depressed condition43. The risk of relapse and recurrence increases with every other episode36,44,45. Based on these findings, MDD should perhaps no longer be considered as an episodic-, but as a chronic and lifelong disorder46 for many people, with much of its disease burden stemming from its recurrent nature47.

The disease burden associated with subtypes of depression (i.e. a single episode and a recurrent episode48) has received remarkably little attention in research. Therefore, research question 1 of this thesis addresses the burden of disease per subtype of MDD (Chapter 2).

Risk factors for relapsePossible risk factors for relapse and recurrence according to the American Psychiatric Association26, are summarized in Table 2. Evidence for these predictors is inconsistent49 although the number of previous episodes and the level of residual symptoms have been found to consistently increase the risk of relapse and recurrence29,49–56. There is growing recognition that recurrently depressed people often experience residual symptoms during remission; approximately one third of these people fails to achieve full remission29,51,57, defined as an absence of symptoms for at least two months26. ‘Partial remission’ is defined in numerous ways in the literature. Most commonly used is a 7-13 score on the Hamilton Rating Scale for Depression58 (HAM-D)51 but it is also frequently defined as the remitted state before full remission is achieved26,59. Residual symptoms of depression probably reflect persistence of the original disorder in a milder form in partially remitted persons and cause significant functional impairment60,61.



17

1Table 2. Risk factors for relapse and recurrence in MDD1

Prior history of multiple episodes of major depressive disorder Persistence of sub threshold depressive symptomsSeverity of initial and any subsequent episodesEarlier age at onsetPresence of an additional non-affective psychiatric diagnosisPresence of a chronic somatic medical disorderFamily history of psychiatric illness, particularly mood disorderOngoing psychosocial stressors or impairmentNegative cognitive stylePersistent sleep disturbances

1 American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV)

The information on course of depression and risk factors for relapse and recurrence can be used for ‘profiling and staging’, which is a diagnostic strategy that takes the heterogeneity of depression into account. By profiling (what risk factors for relapse or recurrence does a person have, e.g. prior history of multiple episodes of major depressive disorder or early age at onset) and staging (in which phase is a person: acute, remission or recovery), more targeted interventions can be offered, in an early stage, to depressed people.

What treatment works best for people with recurrent depression As mentioned, the course of depression can be described as a series of disease stages. Treatment stages can be defined accordingly: acute-phase treatment is given to patients meeting the diagnostic criteria of MDD and its aim is to restore function and shorten episode duration, or equivalently, to promote remission. During remission, continuation treatment can be offered, which has the aim to sustain remission and to prevent relapse. Finally, during recovery, maintenance treatment can be offered, which has the aim to reduce the risk of recurrence (Figure 1). Similar to relapse, recurrence, remission and recovery, the terms continuation and maintenance are often used interchangeably in the literature.

Most people in the Western world who seek help for their acute depression, receive ADM62. Not surprisingly, the most commonly used strategy to prevent relapse during remission is continuation of ADM15,16,63,64. This is in line with guidelines of the National Institute of Health and Clinical Excellence (NICE)65, which recommend to encourage a person who has benefited from taking ADM in the acute phase to continue ADM for at least 6 months during remission on the dose that achieved remission. However, continuing ADM may not always be the preferred option. First, ADM may have side effects such as weight gain and loss of libido14. Second, the optimal duration of the use of ADM has not been studied well enough15,16. Third, there is divergent information on the relation between tapering ADM and relapse or recurrence17. As a consequence, reported levels of non-adherence have been consistently high17–19. Psychosocial interventions may therefore be a valuable alternative to ADM.

R1R2R3R4R5R6R7R8R9


Chapter 1

18

In various meta-analyses acute phase psychological interventions proved to have a sustained ‘prophylactic’ effect, that reduces the risk of relapse or recurrence21,66,67. Still, there is a high proportion of people who experience a relapse or recurrence. Therefore, longer-term strategies, integrated into current longitudinal primary care systems, would fit better with the recurrent and chronic character of depression. Research demonstrates that psychological interventions, specifically aimed at the prevention of relapse and recurrence offered during the continuation- or maintenance phase, are effective in reducing the risk of relapse and recurrence20–23. These interventions are mostly based on CBT, but add strategies such as modifying dysfunctional meta-cognitions in PCT68–70 and meditation in MBCT71–73. The psychological interventions target issues untouched by pharmacological treatments such as awareness and understanding of the disorder, early identification of prodromal symptoms and coping skills. Therefore, they may enable individuals to take a more active role in the management of their disorder. NICE guidelines65 recommend that people with recurrent MDD who are considered to be at significant risk of relapse or recurrence should be offered one of the following psychological interventions: 1) individual CBT for people who have relapsed despite antidepressant medication and for people with a at least 3 depressive episodes and residual symptoms despite treatment or 2) MBCT for people who are currently in remission or recovery, but have experienced three or more previous episodes of depression.

The question we asked ourselves is what preventive psychological interventions currently exist for the prevention of relapse and recurrence, thereby including all types of interventions and all types of delivery modes (e.g. booster sessions and over the Internet). Furthermore, we are interested in what interventions works best. We addressed this research question, by conducting a meta-analysis on the effectiveness of existing preventive psychological interventions offered in the continuation or maintenance phase aimed at the prevention of relapse or recurrence in depressive disorder (Research question 2, Chapter 3). Economic consequences of recurrent depressionMDD is one of the leading causes of health care expenditure. These substantial economic consequences of MDD are mainly due to its recurrent nature74–77. The total annual cost of depression in Europe was estimated at €118 billion in 2004, which corresponds to a cost of €253 per inhabitant and 1% of the total economy of Europe78. Treating MDD in the Netherlands costs almost a billion euro (€966 million) per year12. Research has shown that depressed patients are more often absent from work (absenteeism) and work less effectively while being present at work (presenteeism) than non-depressed people27,79,80. The costs of both absenteeism and presenteeism due to MDD are over €1,8 billion annually, in the Netherlands. The general tendency is that the costs due to depression are largely caused by costs due to lost productivity and that treatment costs comprise a smaller portion.



19

1Besides a need for clinical evidence for the prevention of relapse, there is an additional

need for economic evidence due to the current economic down-turn and an expenditure for mental health care that is unlikely to be sustainable13. It is the combination of data on effects and costs that will inform decision makers, who are forced to allocate limited resources to interventions that maximise cost-effectiveness. Therefore, in Chapter 4, we assess the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care (Research question 3).

The need for a low-cost preventive intervention in primary care The emphasis on cost-effectiveness puts pressure on primary and secondary mental health care, to deliver high quality care at affordable cost. As a consequence, decision makers in the mental health care system are well aware of the need for organizational change. In the Netherlands, the government supports a shift of patients with mild mental problems from secondary care towards primary care by introducing a primary care mental health nurse and a ‘generalistic basic mental health care’ (basis GGZ). Technological developments, such as Internet and mobile apps, and patient-empowering may also play an important role here. In this dynamic environment, a cost-effective, preventive intervention for recurrent depression is sorely needed. Existing psychological interventions often rely on intensive use of therapist’s time in secondary care and are therefore costly. Primary care might be a better setting to offer preventive interventions81. In the Netherlands, as in most western countries, it is the primary care professional who has regular contact with the vast majority of the population, knows about the patients’ social situation and provides low access, continuous care82. Besides, the prevalence of patients with MDD or depressive feelings in primary practice is high, around 21%25. Therefore, a previously evaluated face-to-face PCT 23,68 and mobile PCT70, developed by Bockting et al, was redeveloped into a supported self-help PCT (S-PCT) offered in primary care to persons with a history of depression. Even over 10 years, face-to-face PCT in remitted people with multiple prior episodes, has shown preventive effects on time to recurrence, compared to usual care. Mobile PCT70 has shown a more favourable course over 3 months in the mobile CT group compared to usual care with regard to residual symptoms during remission. A barrier might be that primary care practitioners claim they do not have enough time to perform preventive services. Therefore, the self-help is supported by para-professionals like mental health nurses or social workers in primary care. Growing evidence shows that these para-professionals can effectively deliver self-help treatment protocols for depression, particularly in chronic care models83,84.

First, we add S-PCT as hypothetical intervention to our health economic model in Chapter 4. We ask ourselves how effective this intervention needs to be, to become competitive in terms of its cost-effectiveness relative to existing psychological interventions.

R1R2R3R4R5R6R7R8R9


Chapter 1

20

Second, in order to evaluate the real-life effectiveness and cost-effectiveness of S-PCT, we conducted a pragmatic randomised controlled trial (RCT)- the Parade study- to evaluate this intervention. Details of this study are outlined in the next paragraph.

The PARADE-study; an economic evaluation alongside an effectiveness trialIn the PARADE-study, we made a comparison between usual care augmented with S-PCT, and usual care alone, in terms of terms of both their costs and effects85. (Research question 3, Chapters 5,6,7).

S-PCT is a manualised PCT-based bibliotherapy consisting of a printed self-help book with eight modules, with minimal guidance by a counsellor in primary care86. It is an adapted type of cognitive therapy for acute depression87 and aims to prevent relapse and recurrence in remitted people with a history of depressive episodes. Like regular CT, PCT follows a fixed structure, with agenda setting, review of homework, explanation of the rationale of each session, and the assignment of homework. Participants complete one module per week. Each module includes reading homework plus assignments to be completed in approximately 60 minutes. In the first meeting (by phone or face-to-face), the counsellor explained the rationale of PCT and coming week’s planning. Each week, the counsellor contacted the participant by phone to evaluate progress and understanding. This call was strictly protocolled and was designed to last no longer than 15 minutes. The nature of the contact was solely to support and counsellor the participant and not to actively engage in a therapeutic relationship. Each week, the counsellor completed a checklist with 4 items; (1) the number of that week’s module (1-8), (2) did the participants read the literature of that week (yes/no plus reason), (3) did the participant do the assignments (yes/no plus reason) and (4) time spent on the call (minutes).

The PARADE-study is carried out among 248 persons with a history of depression, currently in remission or recovery, in the Netherlands. Primary outcome is the incidence of relapse or recurrence of depression over a 12-months follow-up period.

Outline of this thesisIn Chapter 2, data from the first wave of the second Netherlands-Mental-Health-Survey-and-Incidence-Study88 (NEMESIS-2) are used to estimate the non-fatal disease burden for both single episode and recurrent depressions. The estimates are assessed from an individual and a population perspective. The estimates are presented as unadjusted, raw estimates and as estimates adjusted for comorbidity. Chapter 3 presents a review with the aim to meta-analytically examine the effectiveness of psychological interventions aimed at the prevention of relapse or recurrence. The health-economic modelling study presented in Chapter 4 assesses how offering preventive psychological interventions would improve the cost-effectiveness of the Dutch health care system for depressive disorders.



21

1Also, we addressed the question how effective a self-help Preventive Cognitive Therapy (S-PCT) needs to be, to become competitive in terms of its cost-effectiveness relative to existing psychological interventions. In Chapter 5 the study protocol of the PARADE-study is presented. This protocol includes a description of S-PCT and the procedure for recruiting people with a history of recurrent depressions, currently in remission. Chapter 6 evaluates the effectiveness of augmenting usual care with S-PCT compared with usual care alone at 12 months post baseline. Chapter 7 describes the cost-effectiveness and cost-utility of S-PCT compared with usual care.

In Chapter 8, we summarise the main findings, thereby answering the four research questions. Further, the main findings are described in light of previous research and we comment on some methodological considerations, associated with the studies in this thesis. Finally, we reflect on the implications for clinical practice and conclude with recommendations for further research.

R1R2R3R4R5R6R7R8R9


Chapter 1

22

REFERENCES

1. De Graaf R, ten Have M, van Gool C, van Dorsselaer S. Prevalence of mental disorders and trends from 1996 to 2009. Results from the Netherlands Mental Health Survey and Incidence Study-2. Social psychiatry and psychiatric epidemiology 2012;47(2):203-13. doi:10.1007/s00127-010-0334-8.

2. Burcusa SL, Iacono WG. Risk for recurrence in depression. Clin Psychol.Rev 2007;27 (8):959-985.

3. Nierenberg AA, Husain MM, Trivedi MH, et al. Residual symptoms after remission of major depressive disorder with citalopram and risk of relapse: a STAR*D report. Psychol.Med 2010;40 (1):41-50.

4. Fava M, Graves LM, Benazzi F, et al. A cross-sectional study of the prevalence of cognitive and physical symptoms during long-term antidepressant treatment. The Journal of clinical psychiatry 2006;67(11):1754-9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17196056. Accessed July 24, 2015.

5. Conradi HJ, Ormel J, de Jonge P. Presence of individual (residual) symptoms during depressive episodes and periods of remission: a 3-year prospective study. Psychological medicine 2011;41(6):1165-74. doi:10.1017/S0033291710001911.

6. Ustun TB, Ayuso-Mateos JL, Chatterji S, Mathers C, Murray CJ. Global burden of depressive disorders in the year 2000. Br.J Psychiatry 2004;184 (0007-1250 (Print)):386-392.

7. Ormel J, VonKorff M, Ustun TB, Pini S, Korten A, Oldehinkel T. Common mental disorders and disability across cultures. Results from the WHO Collaborative Study on Psychological Problems in General Health Care. JAMA 1994;272(22):1741-8. Available at: http://www.ncbi.nlm.nih.gov/pubmed/7966922. Accessed July 23, 2015.

8. Ormel J, Oldehinkel AJ, Nolen WA, Vollebergh W. Psychosocial disability before, during, and after a major depressive episode: a 3-wave population-based study of state, scar, and trait effects. Archives of general psychiatry 2004;61(4):387-92. doi:10.1001/archpsyc.61.4.387.

9. Wells KB, Stewart A, Hays RD, et al. The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA 1989;262(7):914-9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/2754791. Accessed July 23, 2015.

10. Ferrari AJ, Charlson FJ, Norman RE, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS medicine 2013;10(11):e1001547. doi:10.1371/journal.pmed.1001547.

11. Whiteford HA, Degenhardt L, Rehm J, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 2013;(1474-547X (Electronic)).

12. Slobbe L, Smit J, Groen J, Poos M, Kommer G. Kosten van Ziekten in Nederland 2007.; 2011.

13. Forti A, Nas C, van Geldrop A, et al. Mental Health Analysis Profiles (MhAPs) Netherlands, Health Working Papers.(2014). Available at: http://www.ggznederland.nl/uploads/assets/OECD 2014 MhAP The Netherlands.pdf.

14. LAREB. Bijwerkingen Centrum. Available at: http://www.lareb.nl/bijwerkingen.

15. Geddes JR, Carney SM, Davies C, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet 2003;361 (0140-6736 (Print)):653-661.

16. Kaymaz N, van OJ, Loonen AJ, Nolen WA. Evidence that patients with single versus recurrent depressive episodes are differentially sensitive to treatment discontinuation: a meta-analysis of placebo-controlled randomized trials. Journal of Clinical Psychiatry 2008;69 (1555-2101 (Electronic)):1423-1436.

17. Bockting CL, ten Doesschate MC, Spijker J, Spinhoven P, Koeter MW, Schene AH. Continuation and maintenance use of antidepressants in recurrent depression. Psychotherapy and Psychosomatics 2008;77 (1423-0348 (Electronic)):17-26.

18. Meijer WE, Heerdink ER, Leufkens HG, Herings RM, Egberts AC, Nolen WA. Incidence and determinants of long-term use of antidepressants. European Journal of Clinical Pharmacology 2004;60 (0031-6970 (Print)):57-61.

19. Ten Doesschate MC, Bockting CL, Schene AH. Adherence to continuation and maintenance antidepressant use in recurrent depression. J.Affect.Disord. 2009;115 (1573-2517 (Electronic)):167-170.



23

120. Guidi J, Fava GA, Fava M, Papakostas GI. Efficacy of the sequential integration of psychotherapy and

pharmacotherapy in major depressive disorder: a preliminary meta-analysis. Psychol.Med. 2011;41 (2):321-331.

21. Vittengl JR, Clark LA, Dunn TW, Jarrett RB. Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitive-behavioral therapy’s effects. J.Consult Clin.Psychol. 2007;75 (0022-006X (Print)):475-488.

22. Piet J, Hougaard E. The effect of mindfulness-based cognitive therapy for prevention of relapse in recurrent major depressive disorder: A systematic review and meta-analysis. Clinical Psychology Review 2011;31 (1873-7811 (Electronic)):1032-1040.

23. Bockting CLH, Spinhoven P, Wouters LF, Koeter MWJ, Schene AH. Long-term effects of preventive cognitive therapy in recurrent depression: a 5.5-year follow-up study. J Clin.Psychiatry 2009;70 (12):1621-1628.

24. Cuijpers P, Donker T, van SA, Li J, Andersson G. Is guided self-help as effective as face-to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychological Medicine 2010;40 (1469-8978 (Electronic)):1943-1957.

25. van’ t Land H, Grolleman J, Mutsaers K, Smits C. Trendrapportage GGZ 2008: Deel 2: Toegang En Zorggebruik: Basisanalyse. [Trendreport Mental Health Care 2008: Part 2: Admission and Care Utilization]. Utrecht: Trimbos Instituut; 2008.

26. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-V). Washington D.C.: American Psychiatric Association; 2013.

27. Broadhead WE, Blazer DG, George LK, Tse CK. Depression, disability days, and days lost from work in a prospective epidemiologic survey. JAMA 1990;264 (0098-7484 (Print)):2524-2528.

28. Kessler RC, Anthony JC, Blazer DG, et al. The US National Comorbidity Survey: overview and future directions. Epidemiologia e Psichiatria Sociale 1997;6 (1121-189X (Print)):4-16.

29. Judd LL, Akiskal HS, Maser JD, et al. Major depressive disorder: a prospective study of residual subthreshold depressive symptoms as predictor of rapid relapse. J Affect.Disord. 1998;50 (0165-0327 (Print)):97-108.

30. Frank E, Prien RF, Jarrett RB, et al. Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Archives of General Psychiatry 1991;48 (0003-990X (Print)):851-855.

31. Tohen M, Frank E, Bowden CL, et al. The International Society for Bipolar Disorders (ISBD) Task Force report on the nomenclature of course and outcome in bipolar disorders. Bipolar.Disord. 2009;11 (1399-5618 (Electronic)):453-473.

32. Tohen M, Frank E, Bowden CL, et al. The International Society for Bipolar Disorders (ISBD) Task Force report on the nomenclature of course and outcome in bipolar disorders. Bipolar disorders 2009;11(5):453-73. doi:10.1111/j.1399-5618.2009.00726.x.

33. Eaton WW, Shao H, Nestadt G, et al. Population-based study of first onset and chronicity in major depressive disorder. Archives of general psychiatry 2008;65(5):513-20. doi:10.1001/archpsyc.65.5.513.

34. Spijker J, Graaf R, Bijl R V, Beekman AT, Ormel J, Nolen WA. Functional disability and depression in the general population. Results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Acta Psychiatr.Scand. 2004;110 (0001-690X (Print)):208-214.

35. Keller MB, Lavori PW. Double depression, major depression, and dysthymia: distinct entities or different phases of a single disorder? Psychopharmacology bulletin 1984;20(3):399-402. Available at: http://www.ncbi.nlm.nih.gov/pubmed/6473640. Accessed October 28, 2014.

36. Solomon DA, Keller MB, Leon AC, et al. Multiple recurrences of major depressive disorder. American Journal of Psychiatry 2000;157 (0002-953X (Print)):229-233.

37. Mueller TI, Leon AC, Keller MB, et al. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am.J.Psychiatry 1999;156 (0002-953X (Print)):1000-1006.

38. Kanai T, Takeuchi H, Furukawa TA, et al. Time to recurrence after recovery from major depressive episodes and its predictors. Psychol.Med 2003;33 (5):839-845.

39. Mattisson C, Bogren M, Horstmann V, Munk-Jörgensen P, Nettelbladt P. The long-term course of depressive disorders in the Lundby Study. Psychological medicine 2007;37(6):883-91. doi:10.1017/S0033291707000074.

R1R2R3R4R5R6R7R8R9


Chapter 1

24

40. Surtees PG, Barkley C. Future imperfect: the long-term outcome of depression. The British journal of psychiatry : the journal of mental science 1994;164(3):327-41. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8199786. Accessed October 28, 2014.

41. Kiloh LG, Andrews G, Neilson M. The long-term outcome of depressive illness. The British journal of psychiatry : the journal of mental science 1988;153:752-7. Available at: http://www.ncbi.nlm.nih.gov/pubmed/3256374. Accessed October 28, 2014.

42. Kruijshaar ME, Barendregt J, Vos T, de GR, Spijker J, Andrews G. Lifetime prevalence estimates of major depression: an indirect estimation method and a quantification of recall bias. European Journal of Epidemiology 2005;20 (0393-2990 (Print)):103-111.

43. Vos T, Haby MM, Barendregt JJ, Kruijshaar M, Corry J, Andrews G. The burden of major depression avoidable by longer-term treatment strategies. Archives of General Psychiatry 2004;61 (0003-990X (Print)):1097-1103.

44. APA. Diagnostic and Statistical Manual of Mental Disorders : DSM-IV-TR.. Washington DC; 2000.

45. Keller MB, Boland RJ. Implications of failing to achieve successful long-term maintenance treatment of recurrent unipolar major depression. Biological Psychiatry 1998;44 (5):348-360. Available at: http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L28432646.

46. Richards D. Prevalence and clinical course of depression: a review. Clinical psychology review 2011;31(7):1117-25. doi:10.1016/j.cpr.2011.07.004.

47. Judd LL. The clinical course of unipolar major depressive disorders. Archives of General Psychiatry 1997;54 (0003-990X (Print)):989-991.

48. WHO. World Health Organization. The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Description and Diagnostic Guidelines. Geneva: World Health Organization; 1992.

49. Hardeveld F, Spijker J, De Graaf R, Nolen WA, Beekman ATF. Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta psychiatrica Scandinavica 2010;122(3):184-91. doi:10.1111/j.1600-0447.2009.01519.x.

50. Paykel ES. Partial remission, residual symptoms, and relapse in depression. Dialogues.Clin.Neurosci. 2008;10 (1294-8322 (Print)):431-437.

51. Paykel ES, Ramana R, Cooper Z, Hayhurst H, Kerr J, Barocka A. Residual symptoms after partial remission: an important outcome in depression. Psychological medicine 1995;25(6):1171-80. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8637947. Accessed March 21, 2015.

52. Van Londen L, Molenaar RP, Goekoop JG, Zwinderman AH, Rooijmans HG. Three- to 5-year prospective follow-up of outcome in major depression. Psychological medicine 1998;28(3):731-5. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9626729. Accessed March 9, 2015.

53. Pintor L, Gastó C, Navarro V, Torres X, Fañanas L. Relapse of major depression after complete and partial remission during a 2-year follow-up. Journal of affective disorders 2003;73(3):237-44. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12547292. Accessed March 30, 2015.

54. Ramana R, Paykel ES, Cooper Z, Hayhurst H, Saxty M, Surtees PG. Remission and relapse in major depression: a two-year prospective follow-up study. Psychological medicine 1995;25(6):1161-70. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8637946. Accessed March 27, 2015.

55. McGrath PJ, Stewart JW, Quitkin FM, et al. Predictors of relapse in a prospective study of fluoxetine treatment of major depression. Am J Psychiatry 2006;163 (9):1542-1548.

56. Kessing L V, Hansen MG, Andersen PK, Angst J. The predictive effect of episodes on the risk of recurrence in depressive and bipolar disorders - a life-long perspective. Acta Psychiatrica Scandinavica 2004;109 (0001-690X (Print)):339-344.

57. Rush, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry 2006;163(11):1905-1917.

58. Hamilton M. A rating scale for depression. J Neurol.Neurosurg.Psychiatry 1960;23 (0022-3050 (Print)):56-62.

59. Kuyken W, Hayes R, Barrett B, et al. Effectiveness and cost-effectiveness of mindfulness-based cognitive therapy compared with maintenance antidepressant treatment in the prevention of depressive relapse or recurrence (PREVENT): a randomised controlled trial. The Lancet 2015;386(9988):63-73. doi:10.1016/S0140-6736(14)62222-4.



25

160. Zimmerman M, Posternak MA, Chelminski I. Heterogeneity among depressed outpatients considered to be

in remission. Comprehensive psychiatry 2007;48(2):113-7. doi:10.1016/j.comppsych.2006.10.005.

61. Trivedi MH, Corey-Lisle PK, Guo Z, Lennox RD, Pikalov A, Kim E. Remission, response without remission, and nonresponse in major depressive disorder: impact on functioning. International clinical psychopharmacology 2009;24(3):133-8. doi:10.1097/YIC.0b013e3283277614.

62. Olfson M, Marcus SC. National patterns in antidepressant medication treatment. Archives of general psychiatry 2009;66(8):848-56. doi:10.1001/archgenpsychiatry.2009.81.

63. Glue P, Donovan MR, Kolluri S, Emir B. Meta-analysis of relapse prevention antidepressant trials in depressive disorders. The Australian and New Zealand journal of psychiatry 2010;44(8):697-705. doi:10.3109/00048671003705441.

64. Hansen R, Gaynes B, Thieda P, et al. Meta-analysis of major depressive disorder relapse and recurrence with second-generation antidepressants. Psychiatric services (Washington, D.C.) 2008;59(10):1121-30. doi:10.1176/appi.ps.59.10.1121.

65. NICE. National Institute for Health and Clinical Excellence. In: National Institute for Health and Clinical Excellence. Depression: The Treatment and Management of Depression in Adults (update). Clinical Guidelines: National Institute for Health and Clinical Excellence. Depression: the treatment and management of depression in adults (update).; 2009. Available at: www.nice.org.uk/CG90.

66. Beshai S, Dobson KS, Bockting CL, Quigley L. Relapse and recurrence prevention in depression: current research and future prospects. Clin Psychol.Rev. 2011;31 (8):1349-1360.

67. Cuijpers P, Hollon SD, van Straten A, Bockting C, Berking M, Andersson G. Does cognitive behaviour therapy have an enduring effect that is superior to keeping patients on continuation pharmacotherapy? A meta-analysis. BMJ open 2013;3(4). doi:10.1136/bmjopen-2012-002542.

68. Bockting CLH, Schene AH, Spinhoven P, et al. Preventing relapse/recurrence in recurrent depression with cognitive therapy: a randomized controlled trial. J Consult Clin.Psychol. 2005;73 (4):647-657.

69. Fava GA, Grandi S, Zielezny M, Canestrari R, Morphy MA. Cognitive behavioral treatment of residual symptoms in primary major depressive disorder. American Journal of Psychiatry 1994;151 (9):1295-1299. Available at: http://ajp.psychiatryonline.org/cgi/content/abstract/151/9/1295.

70. Kok G, Burger H, Riper H, et al. The Three-Month Effect of Mobile Internet-Based Cognitive Therapy on the Course of Depressive Symptoms in Remitted Recurrently Depressed Patients: Results of a Randomized Controlled Trial. Psychotherapy and psychosomatics 2015;84(2):90-99. doi:10.1159/000369469.

71. Ma SH, Teasdale JD. Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects. J Consult Clin Psychol. 2004;72 (1):31-40.

72. Segal Z V, Bieling P, Young T, et al. Antidepressant monotherapy vs sequential pharmacotherapy and mindfulness-based cognitive therapy, or placebo, for relapse prophylaxis in recurrent depression. Arch Gen.Psychiatry 2010;67 (12):1256-1264.

73. Teasdale JD, Segal Z V, Williams JM, Ridgeway VA, Soulsby JM, Lau MA. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol. 2000;68 (4):615-623.

74. Berto P, D’Ilario D, Ruffo P, Di VR, Rizzo F. Depression: cost-of-illness studies in the international literature, a review. J Ment.Health Policy Econ. 2000;3 (1091-4358 (Print)):3-10.

75. Greenberg PE, Birnbaum HG. The economic burden of depression in the US: societal and patient perspectives. Expert.Opin.Pharmacother. 2005;6 (1744-7666 (Electronic)):369-376.

76. Smit A, Kluiter H, Conradi HJ, et al. Short-term effects of enhanced treatment for depression in primary care: results from a randomized controlled trial. Psychol.Med 2006;36 (1):15-26.

77. Vasiliadis HM, Dionne PA, Preville M, Gentil L, Berbiche D, Latimer E. The excess healthcare costs associated with depression and anxiety in elderly living in the community. Am J Geriatr.Psychiatry 2013;21 (1545-7214 (Electronic)):536-548.

78. Sobocki P, Jönsson B, Angst J, Rehnberg C. Cost of depression in Europe. The journal of mental health policy and economics 2006;9(2):87-98. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17007486. Accessed March 3, 2015.

79. Kessler RC, Barber C, Birnbaum HG, et al. Depression in the workplace: effects on short-term disability. Health affairs (Project Hope) 18(5):163-71. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10495604. Accessed August 18, 2015.

R1R2R3R4R5R6R7R8R9


Chapter 1

26

80. Stewart WF, Ricci JA, Chee E, Hahn SR, Morganstein D. Cost of lost productive work time among US workers with depression. JAMA 2003;289(23):3135-44. doi:10.1001/jama.289.23.3135.

81. Volksgezondheid NK. Preventie in de Huisartsenpraktijk. 2014. Available at: http://www.nationaalkompas.nl/preventie/in-verschillende-settings/zorg/huisartsenpraktijk/.

82. Starfield B. Primary Care: Concept, Evaluation and Policy. New York: Oxford University Press.

83. Hunkeler EM, Meresman JF, Hargreaves WA, et al. Efficacy of nurse telehealth care and peer support in augmenting treatment of depression in primary care. Archives of Family Medicine 2000;9 (1063-3987 (Print)):700-708.

84. Mynors-Wallis LM, Gath DH, Day A, Baker F. Randomised controlled trial of problem solving treatment, antidepressant medication, and combined treatment for major depression in primary care. BMJ 2000;320 (0959-8138 (Print)):26-30.

85. Drummond M, Sculpher M, Torrance GW, O’Brien B, Stoddart G. Methods for the Economic Evaluation of Health Care Programmes. New York: Oxford University Press Inc.; 2005.

86. Bockting CLH, van Valen E. Ingredients of Mobile Preventive Cognitive Therapy for Recurrent Depression. Groningen: University of Groningen; 2009.

87. Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive Therapy of Depression. New York: Guilford; 1979.

88. De Graaf R, Ten Have M, van Dorsselaer S. The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2): design and methods. International Journal of Methods in Psychiatric Research 2010;19 (1557-0657 (Electronic)):125-141.

2Non-fatal disease burden for subtypes of depressive

disorder:population-based epidemiological study

Karolien E.M. Biesheuvel-LeliefeldGemma D. Kok

Claudi L.H. BocktingRon de Graaf

Margreet ten HaveHenriette E. van der Horst

Anneke van SchaikHarm W.J. van Marwijk

Filip Smit

Accepted for publication in BMC Psychiatry

R1R2R3R4R5R6R7R8R9


Chapter 2

28

ABSTRACT

Background Major depression is the leading cause of non-fatal disease burden. Because major depression is not a homogeneous condition, this study estimated the non-fatal disease burden for mild, moderate and severe depression in both single episode and recurrent depression. All estimates were assessed from an individual and a population perspective and presented as unadjusted, raw estimates and as estimates adjusted for comorbidity.

Methods We used data from the first wave of the second Netherlands-Mental-Health-Survey-and-Incidence-Study (NEMESIS-2, n=6,646; single episode DSM-IV depression, n=115; recurrent depression, n=246). Disease burden from an individual perspective was assessed as ‘disability weight * time spent in depression’ for each person in the dataset. From a population perspective it was assessed as ‘disability weight * time spent in depression *number of people affected’. Statistical adjustments were made for co-morbid mental disorders and physical illnesses. The presence of mental disorders was assessed with the Composite International Diagnostic Interview (CIDI) 3.0.

Results Single depressive episodes emerged as a key driver of disease burden from an individual perspective. From a population perspective, recurrent depressions emerged as a key driver. These findings remained unaltered after adjusting for co-morbidity.

Conclusions The burden of disease differs between subtypes of depression and depends much on the choice of perspective. The distinction between an individual and a population perspective may help to avoid misunderstandings between policy makers and clinicians.


Disease burden for subtypes of depression

29

2

BACKGROUND

Depressive disorders affect 15% of the population on a lifetime basis1 and have a detrimental impact on social, family and professional role functioning2,3,4,5. Depression however, is not a homogeneous condition and its burden of disease might vary across DSM-IV subtypes.

Subtypes of depression can be classified into single episode or recurrent depression and then further graded by severity: mild, moderate or severe6. Burden of disease estimates for subtypes of depression have received remarkably little attention in research. Kruijshaar et al.7 studied the associations of severity and type of depression with functional impairment of the individual in a Dutch general population sample. They concluded that recurrent depression was found not to be associated with more impairment than single episode depression. Higher severity classes however were associated with more impairment. In contrast,8 arrived at the conclusion that recurrent depressions are associated with a greater burden of disease.

It should be noted that burden of disease can be assessed at the individual and at population level. At individual level, clinicians tend to give priority to the disorders that exact the heaviest toll on their patients while from a population perspective the disease burden might be driven by the number of people affected in addition to case severity and disease duration. Indeed, a study by Lokkerbol et al.9 about the non-fatal burden of several mental disorders showed that the rank order of disorders by individual burden is often different from the rank order which is based on the population-level disease burden.

This current study aims to estimate the non-fatal burden of disease for subtypes of depression from both an individual and population perspective. Distinguishing both perspectives may help both clinicians and policy-makers to conduct debates about resource allocation in a clearer way. In addition, we take into account the impact of comorbidity. Estimating the disease burden with and without adjusting for comorbidity addresses two fundamentally different questions. When adjusting for comorbidity, one addresses an (academic) question about a disorder’s unique contribution to the disease burden overall. When also incorporating the additional disability weights of comorbid conditions one addresses a (pragmatic) question how much people suffer from a disorder while taking the realistic perspective that in real life people are not adjusted for comorbidity.

Taking these notions as starting points, we hypothesised that from an individual perspective, single and recurrent depressive episodes exact the same toll on individual patients (H1). However, when assessed from population perspective, we hypothesised that recurrent depression would emerge as health care priority due to the large number of people affected by recurrent depressions (H2). After all, some 45% of the depressed people experience recurrences, usually cumulating to seven or eight depressive episodes over the course of their life7 and spending as much as 21% of their lifetime in a depressed

R1R2R3R4R5R6R7R8R9


Chapter 2

30

condition8. These hypotheses may not be as straightforward as might seem because the duration of depressive episodes might be shorter in recurrent depressions10, which requires an empirical investigation. We also hypothesised that the non-fatal burden of disease across MDD subtypes follows the gradient of symptom severity (H3). Finally, due to comorbid conditions that may lend extra weight to the burden of disease, we hypothesised that after adjusting for comorbidity, the burden of disease for major depression is lower for each subtype of depression (H4).

METHODS

SampleData were collected from the first wave (2007-2009) of the second Netherlands Mental Health Survey and Incidence Study (NEMESIS-2, n=6,646). The methods used have been described elsewhere in detail11,12.

Briefly, a multistage, stratified random sampling procedure was applied. A random sample of 184 of the 443 existing municipalities was drawn. In these municipalities, a random sample of addresses of private households from postal registers was drawn. Based on the most recent birthday at first contact within the household, an individual aged 18–64 with sufficient fluency in the Dutch language was randomly selected for interview. The study was approved by a medical ethics committee and respondents provided written informed consent. Selected households received a letter from the Dutch Minister of Health, Welfare and Sport, in which the study was explained and recommended. Households were contacted by phone or visited in person if no phone number was available, at least ten times during November 2007 to July 2009. The response rate was 65.1%.

The sample was fairly representative of the general Dutch population, but males, younger people (especially in the 18–24 age bracket), people who had attained fewer years of education, and those not in paid employment were somewhat underrepresented. Therefore, post-stratification weights were used in all analyses11,12. After weighting, the sample followed exactly the same multivariate distribution over age, gender, civil status and urbanization as the population according to Statistics Netherlands 2013 (www.statline.cbs.nl). 361 Respondents were diagnosed with MDD in the past 12 months according to the Composite International Diagnostic Interview (CIDI) 3.013.

MeasuresThe presence of mental disorders (major unipolar depression and comorbid mental disorders) was assessed with the Composite International Diagnostic Interview (CIDI) 3.013. The CIDI is developed by the World Health Organization and is a psychiatric interview



31

2

generating 12-month and life-time prevalence rates of the DSM-IV mental disorders. The CIDI 3.0 was first produced in English and underwent a rigorous process of adaptation to obtain a conceptually and cross-culturally comparable Dutch version14,15. Clinical calibration studies in various countries16 found that the CIDI 3.0 assesses anxiety, mood and substance use disorders with generally good validity in comparison to blinded clinical reappraisal interviews. Studies on earlier CIDI versions concluded that the CIDI assesses common mental disorders with generally acceptable reliability and validity17,18. Comorbid mental disorders included any anxiety disorder, any substance use disorder, dysthymia and attention deficit hyperactivity disorder (ADHD). Eating disorders was not diagnosed in NEMESIS-2. A diagnosis of bipolar disorder in the last 12 months by definition excludes the diagnosis of depression in the same period and was therefore not taken into consideration.

The CIDI diagnostic interview was also used to retrospectively classify type of depression (single or recurrent) and assess the duration (in days), spent in depression in the last twelve months. To study co-morbidity and its effects on the burden of disease, we used the CIDI without imposing the rules for the hierarchy among the disorders, meaning that if a person manifests with two disorders, we count this as two distinct disorders and not as a single disorder (e.g. depressive episode as part of bipolar disorder).

Severity of depression was assessed with the widely used and validated Sheehan Disability Scale (SDS)19. The SDS is a self-report measure of condition-specific disability and was incorporated in all diagnostic CIDI sections. It consists of four questions, each asking the respondent to rate, on a scale from 0 to 10, the extent to which a particular disorder ‘interfered with’ activities in one of four role domains (home, work, social, close relationships) during the month in the past year when the disorder was most severe. ‘Severe’ depression cases score in the range of 7-10 in at least two areas of role functioning. ‘Moderate’ depression cases are those who score 4-6 in any domain. Remaining cases are defined ‘mild’20. Depression cases with unspecified severity (missings on all domains of the SDS, n=22) were re-scored, using two questions of the CIDI about possible disruption of work, social contacts or personal relations (question D28) and/or in daily routine (question D28a). ‘Severe’ depression cases score 4 or 5 (D28) and 1 or 2 (D28a). ‘Moderate’ cases score 3 (D28) and 1or 2 (D28a) or 4 (D28) and 3 or 4 (D28a). Remaining cases are defined ‘mild’.

Disability weights (DWs) are weight factors that reflect the severity of the disease on a scale from 0 (perfect health) to 1 (equivalent to death). DWs were obtained from the Medical Outcome Study Short Form 6 Dimensions (SF-6D) using Brazier’s algorithm21, a well-validated algorithm that was applicable to our data. The SF-6D is a much used and well-validated instrument to estimate Health Related Quality of Life valuations (HR QoL) derived from the Medical Outcomes Study22. It is of note that the SF-6D can describe as

R1R2R3R4R5R6R7R8R9


Chapter 2

32

many as 18,000 health states, i.e. all the permutations of the items (1) physical functioning, (2) role limitations, (3) social functioning, (4) pain, (5) mental health, and (6) vitality, each of which has five or six possible answers. Brazier and colleagues used a sub-sample of 249 health states to elicit valuations in a representative sample (N = 836) from the general public in the UK. During a personal interview each respondent was asked to value the selected health states, and valuation was carried out using the standard gamble method, which was originally developed by Von Neumann and Morgenstern23. In standard gamble, individuals are asked to make a hypothetical choice between the certainty of living in that particular health state versus engaging in a treatment entailing a chance of getting well at probability P and dying at probability 1-P. The idea is that people are more willing to accept a risky treatment that involves a higher risk of dying when their HR QoL is poor. Brazier and colleagues used the health state valuations obtained in an econometric model to predict the values of all 18,000 health states that can be described by the SF-6D and to assess DWs.

Covariates included gender, age and level of education. Moreover, we adjusted DWs for the presence of comorbid mental disorders (assessed with the CIDI) and somatic illnesses, which might inflate DWs. These somatic illnesses were based on self-reports and consisted of a list 17 chronic general medical disorders being treated or monitored by a physician in the 12 months preceding baseline, such as asthma, COPD, osteoarthritis, heart disease, peptic ulcer and diabetes. Comparisons between self-reports of chronic physical disorders and medical records show moderate to good concordance24,25,26.

Metrics of non-fatal disease burdenIndividual perspective: Quality adjusted life year (QALY) decrements A common metric to describe an individual’s health-related quality of life is Quality Adjusted Life Year (QALY). A QALY gain is the amount of time (T) spent in a health state, multiplied by a valuation of that health state. This valuation is called ‘utility’ (U), which is anchored between 0 (‘death’) and 1 (perfect health). Utilities can be rescaled such that a higher score signifies poorer health and are then called ‘disability weights’ (DWs). In this burden of disease study we base our calculations on the DWs. As such, the focus of this disease burden paper is at QALY decrements instead of QALY gains. To illustrate, living 5 years with a DW of 0,34 is equivalent to (5*0,34=) 1,7 QALY decrements.

Population perspective: Years Lived with Disability (YLD)At population level we are not looking at one single individual spending time (T) in a certain health state weighted by a disability weight (DW), but at (N) individuals spending variable amounts of time, T, in that health state weighted by DW. This results in Years Lived with Disability (YLD). The amount of time individuals collectively spend in a health condition



33

2

can be described by the number of person-years (PYRS). When we want to describe the disease burden of depressive disorder from a population perspective we compute the number of person years (PYR) and multiply these by the disability weight associated with that health state: YLD = PYR * DW.

Henceforth, we reserve the terms “QALY decrement” for disease burden from individual perspective and “YLD” for disease burden from population perspective.

AnalysisDWs were estimated for the various subtypes of depression (single, recurrent, single mild, single moderate, single severe, recurrent mild, recurrent moderate, recurrent severe) in order to estimate QALY decrements and YLDs. Here, we took two approaches: one without, and another with adjustments for co-morbid mental disorders and somatic illnesses (12-month depression and 12-month comorbid mental disorders).

QALY decrements not adjusted for comorbidity were computed as the average DW of all respondents in a certain subtype of depression multiplied by the average time spent in a depressed health state in the last year. While this approach may portray a realistic picture of the disease burden of a disorder, it can be criticised for overestimating the disease burden attributable to a disorder when there are comorbid conditions that lend extra weight to the DW.

QALY decrements adjusted for comorbidity can be computed in various ways27. In this study we corrected for comorbidity by regressing DWs on all the depression subtypes, other mental disorders and somatic illnesses. The regression coefficients were then interpreted as the DW of one depression subtype adjusted for comorbid mental disorders and somatic illnesses. The intercept (constant) in the regression models could be interpreted as the DW attributable to unobserved factors affecting health-related quality of life such as minor illnesses, accidents and conditions that were not measured. The DW of a disorder is the base-rate DW (intercept, a) plus the adjusted DW corresponding to this disorder (regression coefficient, b), thus: DW=a+b. In this way, the adjusted DWs were computed for all subtypes of depression and multiplied by the average time spent in depressed state to estimate adjusted QALY decrements.

YLDs not adjusted for comorbidity per subtype were computed by multiplying DWs by the number of person-years (PYRs). PYRs were calculated as the number of affected people (1-year prevalence) multiplied by the total time spent in depression in the last year. To facilitate interpretation, results were standardised per one million persons, thus expressed as YLDs/mln.

YLDs adjusted for comorbidity were corrected for comorbidity by regressing DWs on all the subtypes, comorbid mental disorders and somatic illnesses.

R1R2R3R4R5R6R7R8R9


Chapter 2

34

The Brazier algorithm we used was executed in Excel (version 11.0 for Windows, 2003) and can be obtained from John Brazier at Sheffield University. All other analyses (estimation of DW, QALY, YLD, PYRS and standard errors) were conducted in Stata (version 10.1 for Windows). As data were weighted, Stata’s procedure for design-based analysis and robust statistical techniques based on first-order Taylor-series linearization method were used to obtain correct sample errors.

RESULTS

Sample characteristics Demographic characteristics of the total NEMESIS-2 sample and of the 361 individuals with depression are shown in Table 1 (weighted). The depressed population includes more women, is less highly educated than the total population and suffers significantly more from comorbid disorders.

Table 1. Characteristics of the total NEMESIS-2 population (N=6,646) and of respondents diagnosed with depression (last year-prevalence, n=361) with standard errors (SE), weighteda

total population Nemesis-2

depressed populationNemesis-2

SE

n (percentage) 6,646 (100%) 361 (5,2%)mean age (years) 41.6 40.2 1.038females (%) 49.6 60.2 0.035educational level high - cat 4b (%) 28.0 21.45 0.025any 12-month anxiety disorder (%) 10.1 40.0 0.031any 12-month substance use disorder (%) 5.6 10.5 0.019any 12-month somatic illness (%) 31.6 46.9 0.040any 12-month dysthymia (%) 0.89 15.4 0.026any 12-month ADHD (%) 1.2 2.7 0.010mean duration episode last 12 months (days) n/a 120.86 7.15single vs recurrent episodes (%) n/a 32.7 / 67.3 0.038 / 0.040mild vs moderate vs severe (%) n/a 7.6 / 36.1 / 56.3 0.015 / 0.032 / 0.033

Abbreviations: SE=standard error, ADHD= Attention Deficit Hyperactivity Disordera weighting based on city of residence, part of country (north, south, east, west) and on a specific weight-factor to correct for differences in the response rates in several socio-demographic groups and in the probability of selection of respondents within householdsb higher professional or university education



35

2

QALY decrements not adjusted for comorbidity Figure 1 presents the number of affected patients (n), average duration of time spent in depression (T), disability weights (DW) that were not adjusted for comorbidity, individual QALY decrements (QALY) and the number of years lived with disability per one million (YLD) for each of the subtypes in the population in the last year. The whole depression sample (n=361) has, on average, a disability weight of 0.27 (se 0.010), meaning that people who meet the criteria for depression have a health-related quality of life that is 27% lower than the non-depressed population. Single episodes have a longer duration than recurrent depressions (T= 149 and T=107 days respectively). Regarding QALY decrements, a single episode poses a greater burden on individuals than a recurrent depressive episode (QALY=0.111 and 0.078 respectively). Besides, higher levels of symptom severity are associated with higher QALY decrements.

Depressed personsN 361T 121 (7.1)

DW 0.27 (0.010)

QALY 0.089 (0.006)

YLD 4,664 (569)

Single episodeN 115T 149 (13.5)

DW 0.27 (0.014)

QALY 0.111 (0.012)

YLD 1,882 (238)

Recurrent episodesN 246T 107 (8.7)

DW 0.27 (0.013)

QALY 0.078 (0.007)

YLD 2,782 (437)

MildN 12T 121 (35.8)DW 0.23 (0.030)QALY 0.076 (0.025)YLD 94 (27)

ModerateN 39T 143 (21.3)DW 0.27 (0.017)QALY 0.105 (0.017)YLD 540 (93)

MildN 20T 84 (17)DW 0.24 (0.027)QALY 0.055 (0.009)YLD 153 (69)

SevereN 64T 155 (18.0)DW 0.27 (0.020)QALY 0.116 (0.016)YLD 1,232 (197)



Figure 1. Unadjusted characteristics of burden of disease for subtypes of depression* *Number of affected respondents (n), mean time spent in depression in days (T), disability weight (DW), quality of life decrement (QALY), years lived with disability per 1 million population (YLD)

and standard errors (between parentheses) for major depression and subtypes of major depression; unadjusted for comorbid mental disorders and somatic illnesses. Standard errors of YLD were calculated using the standard rules when multiplying two variables, under the assumption that both variables (DW and PYRS/mln) are independent (se YLD/mln= √ ((se DW / DW2)+ (se (PYRS/mln) / (PYRS/mln)2))

R1R2R3R4R5R6R7R8R9


Chapter 2

36

QALY decrements adjusted for comorbidityFigure 2 presents the results when adjustments are made for comorbidity. Both unadjusted and adjusted QALY decrements show more or less the same ranking of results; single depression emerges as the leading causes of non-fatal disease burden (QALY =0.111). It appears that the adjusted QALYs are lower than the unadjusted QALYs by 18% on average. The QALYs for single depression became smaller after adjusting for comorbidity than those for recurrent depression (0.017 versus 0.014 respectively).

Depressed personsN 361T 121 (7.1)

DW 0.22 (0.014)

QALY 0.073 (0.006)

YLD 3,973 (530)

Single episodeN 115T 149 (13.5)

DW 0.23 (0.037)

QALY 0.094 (0.017)

YLD 1,603 (306)

Recurrent episodesN 246T 107 (8.7)

DW 0.22 (0.034)

QALY 0.064 (0.011)

YLD 2,267 (480)

MildN 12T 121 (35.8)DW 0.20 (0.055)QALY 0.066 (0.027)YLD 81 (13)


MildN 20T 84 (17)DW 0.21 (0.039)QALY 0.048 (0.009)YLD 134 (--)




Figure 2. Adjusted characteristics of burden of disease for subtypes of depression**Number of affected respondents (n), mean time spent in depression in days (T), disability weight (DW), quality of life decrement (QALY), years lived with disability per 1 million population (YLD)

and standard errors (between parentheses) for major depression and subtypes of major depression; adjusted for comorbid mental disorders and somatic illnesses. Standard errors of YLD were calculated using the standard rules when multiplying two variables, under the assumption that both variables (DW and PYRS/mln) are independent (se YLD/mln= √ ((se DW / DW2)+ (se (PYRS/mln) / (PYRS/mln)2))

YLDs/mln not adjusted for comorbidity From population perspective, the number of affected people becomes an important driver of disease burden. It appears that from a public health perspective recurrent depression causes a larger disease burden than single episode depression (YLD/mln= 2,782 and 1,882 respectively). Again, higher levels of symptom severity are associated with a greater burden of



37

2

disease. Severe single episode and recurrent depressions emerge as the subtypes associated with the highest levels of YLD disease burden. Both types of depression are associated with a relatively large number of person-years and in addition have a high average DW, making them both leading causes of disease burden as seen from a population perspective.

YLDs/mln adjusted for comorbidityFrom population perspective, YLDs/mln are on average 15% lower after adjustment. Comparable to disease burden at individual level, both the unadjusted and adjusted YLDs show a similar hierarchy: recurrent depression emerges as the leading cause of non-fatal disease burden (YLD/mln=2.267). Single mild depression is the least disabling condition, mainly due to the small number of affected people.

DISCUSSION

Main findingsThis study estimated the non-fatal burden of disease for subtypes of depression such as single episode and recurrent depression, graded by severity (mild, moderate, severe depression). All estimates were assessed from an individual or a population perspective. In addition, the disease burden estimates were adjusted for comorbidity.

We hypothesised that from an individual perspective, the disease burden of a single depressive episode would be equal to the disease burden of a recurrent depression. However, results show that a single episode is associated with greater non-fatal disease burden than a recurrent depression. Apparently, disability weights are equal (both 0.27) but the time spent in a recurrent depression versus a single episode is shorter (107 vs 149 days respectively).

The other three hypotheses were confirmed; from a public health perspective, recurrent episodes are associated with greater disease burden (H2). Even though the mean time spent in a recurrent depression is shorter, the large number of people affected comes into play (n=115 single episode and n=246 recurrent episode). As expected, the burden of disease follows the gradient of symptom severity (H3). Finally, the burden of disease for each subtype of depression is lower after adjusting for comorbidity (H4).

Context and other studiesWe need to place our findings in the wider context of the literature. Depressive disorder has been consistently identified as a leading cause of disability28,29. Currently, depression is the single leading cause of non-fatal YLD disease burden in high-income countries and it is projected to become the second leading cause of DALY disease burden (which also

R1R2R3R4R5R6R7R8R9


Chapter 2

38

accounts for mortality) by 2020, second only to ischemic heart disease30. More recent projections predict that depression might become the single leading cause of DALY disease burden in the high-income countries by the year 203028. The Global Burden of Disease (GBD) studies31 conducted in 1990 and 2000 have quantified non-fatal health outcomes across a range of disorders at the global and regional level. The leading causes of YLDs were much the same in 2010 as they were in 1990, with depressive disorder contributing 8.1% of total YLDs, ranking second after low back pain. Depressive disorder caused 63 million YLDs globally, but this figure was not disaggregated across the various types of depressive disorder. More recently, another Global Burden of Disease publication4 showed that depressive disorders accounted for most YLDs within the group of mental and substance use disorders (42.5%, 95%CI: 33.3–51.7)).

Kruijshaar et al7 studied the associations of severity and type of depression with functional impairment of the individual in a Dutch general population sample. Functional impairment was defined as limitations in physical, psychological and social functioning. Impairment was measured using the Short-Form-36 Health Survey (SF-36)32 and two other indicators of impairment were added to reflect some of the economic consequences of depression as well: the number of days spent in bed due to psychiatric, drug- or alcohol-related problems, and the time missed from work due to these problems. The study did not make a difference between disability at individual or population level. They concluded that higher severity classes were associated with more impairment. However, recurrent depression was found not to be associated with more impairment than single episode depression. In contrast, Vos et al8 arrived at another conclusion. Their study focused on quantifying the burden of disease currently averted in people seeking care for major depression and the amount of disease burden that could be averted in these people under optimal episodic and maintenance treatment strategies. Results suggested that recurrent depression is the key driver of disease burden and that depression should be treated as a chronic episodic disorder in order to reduce this great burden of disease.

A study by Lokkerbol et al9, aimed at estimating the non-fatal burden of disease at both individual and population level due to several mental disorders in a Dutch population sample, estimated an unadjusted DW of 0.35 for depression and an adjusted DW of 0.25. Years lived with depression (YLD) per million were estimated 9,117 and 6,524 respectively. These findings showed that co-morbidity plays an important role in causing disability, analogue to our findings. Contrary to our findings however, Lokkerbol’s adjusted and unadjusted DWs and YLD/mln led to different rank orders. However, Lokkerbol et al. used the first NEMESIS data33 and as a consequence, their analyses were based on DSM-III-R disorders while ours were based on DSM-IV disorders.

The fact that, from an individual perspective, the burden of disease for single depression decreased more than the burden of disease for recurrent depression after adjusting for



39

2

comorbidity (0.017 versus 0.014 respectively) indicates that single depression is more often comorbid with other conditions.

Strengths and limitationsThe strengths of this study include the population-based representative dataset on which the analyses are conducted and using reliable instruments like the CIDI and the SDS.

Another strength is that the disability weights were derived from the general population. The most recent Global Burden of Disease Study34 also accommodated data collection through population-based household surveys rather than from expert panels. This is important, because there are several ways of eliciting these valuations (e.g. from professionals in the medical field), but these are surrounded by controversy, and ultimately we need to understand how people value their own health35.

A third strength is that we could compute both unadjusted and adjusted estimates. Unadjusted QALY decrements and YLDs portray an accurate picture of the burden of disease in subtypes of people that are likely to have comorbid conditions—after all, in real life, we do not encounter people who have been adjusted for comorbidity. Therefore, unadjusted estimates may have value from a public health perspective. However, when the aim is to assess the burden of disease attributable to a specific type of disorder, then adjusted estimates are preferred, because adjusted estimates give information about the level of disability due to a disorder without confounding by co-occurring conditions. In our study, we explored both approaches and were thus able to shed light on the different conclusions that depend on the chosen perspective.

A final strength is the distinction between a clinical perspective on individuals and a public health perspective on populations. This distinction might be important to inform clinicians, decision makers and researchers in the health care sector in a conceptually clear and unambiguous way.

We acknowledge the following limitations of our study. First, we used a one-year timeframe for our analysis. Therefore we might have missed the intermediate dynamics of incidence, disease duration, remission and recurrence within that year. Perhaps more importantly, the longer-term dynamics of the epidemiology of depressive disorder beyond one year were missed in our study. To illustrate, Vos and colleagues8 estimated that in people with a history of depression the risk of a recurrence is 60% within the first year after remission of the index episode, 70% after two years, 80% after 20 years and might be as high as 90% on a lifetime basis. Despite this limitation, our approach is well-accepted and is used in the most recent WHO Global Burden of Disease (GBD) study5, which aims to estimate the burden of disease consistently across diseases, risk factors and regions. To estimate burden of disease the GBD study was based on point (current/past month) or past year prevalence estimates and excluded lifetime estimates as recall bias invalidated them as credible measures of disease burden.

R1R2R3R4R5R6R7R8R9


Chapter 2

40

Second, we focused on the non-fatal disease burden, ignoring excess mortality, but elsewhere we computed that the risk of premature death is a factor 1.65 higher in people with depression as compared to people without depression36. Especially when taking the life-course perspective, mortality may have significant impacts on disease burden, and these impacts were missed due to our focus on ‘instantaneous’ non-fatal disease burden. Third, people with severe conditions may have been unable to participate in this population-based survey, because they were hospitalised, and this is likely to have resulted in an under-estimation of the disease burden in the more severe forms of depression. Fourth, we used the Brazier algorithm to calculate utilities and disability weights, but this algorithm was based on health state valuations in a sample of British people, while our sample was Dutch. This may have distorted our outcomes somewhat, although it is unlikely to change the overall results in a material way, as differences in Western Europe between national value sets, such as the set for the SF-6D, are small37. While we feel that the people should be the ultimate judges of their own health, a panel of lays may be associated with limitations that are worth noting, such as lesser consistency, and the possibility that (healthy) lays have difficulties passing judgments on the severe conditions. This may have caused some under-estimation of the disability weights associated with the more severe disorders. In fact, regarding the more severe conditions, Brazier et al21 pointed out that ‘inconsistent estimates (…) of the value of the poorest health states’ might be seen as a limitation of their method. A fifth limitation is that data on somatic illnesses is based on self-report rather than diagnostic assessment, as well as the fact that we did not adjust for every illness, but only for 17 illnesses. This may have inflated DWs and YLDs estimates for depressive disorder. A sixth limitation is that we did not adjust for severity of comorbidity, which might have provided a more realistic picture. A final point: our analyses were performed on a sample that was diagnosed with depression during the past year, whereas the MOS-SF-36 is designed to assess disability over the period of the past 4 weeks. The depressive episode may have occurred well before the last month as the average duration of an episode is 4–6 months. As a result, past month level of disability may not be truly representative for the disability that was actually experienced during the disorder. The individuals that are no longer suffering from depression may have therefore contributed to an underestimation of DWs, QALY decrements and YLDs. The use of a 12-month timeframe to estimate 1-month disability has the advantage that remitted disorders which continue to have residual adverse effects on disability are included. In sum, we captured the average effect of both acute and remitted episodes in the past 12 months on disability.

In this paragraph we mentioned a number of limitations. If these limitations led to bias overall, this bias will probably have led to an underestimation of the burden of disease but will probably not (or at least to a lesser degree) have impacted the ranking of the subtypes of depressive disorder.



41

2

Implications Major depression is a leading cause of non-fatal disease burden, but it is not a homogeneous condition and assessing the disease burden for depression subtypes may help prioritise treatment allocation.

Overall, we saw that the disease burden differs from one subtype to another and that comorbidity influences the results. In addition, our study showed that the burden of depression poses a substantial challenge both from a clinical perspective (at individual level) as well as seen from a public health perspective (at population level). This justifies the prevailing dichotomy of medicine into clinical medicine (directed at individual patients) and public health (directed at collectives). Health care providers who focus on helping individual patients may conclude from our results that a single episode depression causes a greater burden of disease than a recurrent depression. However, the high prevalence of recurrent depression in the population does raise questions about how to best alleviate the disease burden stemming from recurrent depression when taking the public health perspective. The distinction between a clinical and a public health perspective may cause confusion in debates when not made explicit.

Our message is that individual and population perspectives are neither absolute nor independent concepts. Both perspectives serve different purposes and may require careful alignment when being used jointly. Such an alignment may result in the optimal balance between an individual approach directed, for example, at the episodic treatment of acute single episode depressions, in combination with a public health care approach with an emphasis on the longer-term preventive management of recurrences.

R1R2R3R4R5R6R7R8R9


Chapter 2

42

REFERENCES

1. Bromet, E., Andrade, L. H., Hwang, I., Sampson, N. A. et al. Cross-national epidemiology of DSM-IV major depressive episode BMC Med 2011;9:90-

2. Broadhead, W. E., Blazer, D. G., George, L. K., and Tse, C. K.Depression, disability days, and days lost from work in a prospective epidemiologic survey JAMA 21-11-1990;264(19):2524-2528.

3. Kessler, R. C., Anthony, J. C., Blazer, D. G., Bromet, E. et al. The US National Comorbidity Survey: overview and future directions Epidemiol Psichiatr Soc 1997;6(1):4-16.

4. Whiteford, H. A., Degenhardt, L., Rehm, J., Baxter, A. J. et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010 Lancet 28-8-2013;

5. Ferrari, A. J., Charlson, F. J., Norman, R. E., Patten, S. B. et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010 PLoS Med 2013;10(11):e1001547-

6. World Health OrganizationThe ICD-10 Classification of Mental and Behavioural Disorders: Clinical Description and Diagnostic Guidelines. 1992;

7. Kruijshaar, M. E., Hoeymans, N., Bijl, R. V., Spijker, J. et al. Levels of disability in major depression: findings from the Netherlands Mental Health Survey and Incidence Study (NEMESIS) Journal of Affective Disorders 2003;77(1):53-64.

8. Vos, T., Haby, M. M., Barendregt, J. J., Kruijshaar, M. et al. The burden of major depression avoidable by longer-term treatment strategies Arch Gen Psychiatry 2004;61(11):1097-1103.

9. Lokkerbol, J., Adema, D., de, Graaf R., Ten, Have M. et al. Non-fatal burden of disease due to mental disorders in the Netherlands Soc Psychiatry Psychiatr Epidemiol 10-2-2013;

10. Spijker, J., de, Graaf R., Bijl, R. V., Beekman, A. T. et al. Duration of major depressive episodes in the general population: results from The Netherlands Mental Health Survey and Incidence Study (NEMESIS) Br J Psychiatry 2002;181:208-213.

11. de Graaf, R., Ten Have, M., and van Dorsselaer, S.The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2): design and methods International Journal of Methods in Psychiatric Research 2010;19(3):125-141.

12. de Graaf, R., Ten Have, M., van Gool, C., and van Dorsselaer, S.Prevalence of mental disorders, and trends from 1996 to 2009. Results from NEMESIS-2 Tijdschrift voor Psychiatrie 2012;54(1):27-38.

13. Kessler, R. C. and Ustun, T. B.The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI) International Journal of Methods in Psychiatric Research 2004;13(2):93-121.

14. Alonso J, Angermeyer MC, Bernert S, Bruffaerts R et al. Use of mental health services in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project Acta Psychiatr Scand Suppl 2004;(420):47-54.

15. de Graaf, R., Ormel, J, Ten Have, M., Burger, H. et al. Mental disorders and service use in The Netherlands. Results from the European Study of the Epidemiology of Mental Disorders (ESEMeD). 2008;388-405.

16. Haro, J. M., Arbabzadeh-Bouchez, S., Brugha, T. S., de, Girolamo G. et al. Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys International Journal of Methods in Psychiatric Research 2006;15(4):167-180.

17. Andrews G and Peters LThe psychometric properties of the Composite International Diagnostic Interview Soc Psychiatry Psychiatr Epidemiol 1998;33(2):80-88.

18. Wittchen, H. U.Reliability and validity studies of the WHO--Composite International Diagnostic Interview (CIDI): a critical review J Psychiatr Res 1994;28(1):57-84.

19. Leon, A. C., Olfson, M., Portera, L., Farber, L. et al. Assessing psychiatric impairment in primary care with the Sheehan Disability Scale Int J Psychiatry Med 1997;27(2):93-105.

20. Ten Have, M., Nuyen, J., Beekman, A., and de, Graaf R.Common mental disorder severity and its association with treatment contact and treatment intensity for mental health problems Psychol Med 2013;43(10):2203-2213.



43

2

21. Brazier, J., Roberts, J., and Deverill, M.The estimation of a preference-based measure of health from the SF-36 J Health Econ 2002;21(2):271-292.

22. Ware, J. E., Jr. and Sherbourne, C. D.The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection Med Care 1992;30(6):473-483.

23. Von Neumann, J. and Morgenstern, O.Theory of games and economic behavior 1953;

24. Baker, MM, Stabile, M, and Deri, CWhat do self-reported, objective measures of health measure? 2013;

25. Knight, M., Stewart-Brown, S., and Fletcher, L.Estimating health needs: the impact of a checklist of conditions and quality of life measurement on health information derived from community surveys J Public Health Med 2001;23(3):179-186.

26. National Center for Vital and Health StatisticsEvaluation of national health interview survey diagnostic reporting. Series 2: data evaluation and methods 1994;Hyattsville, Maryland, Department of Health and Human Services(120)

27. Andrews G, Sanderson K, and Beard JBurden of disease. Methods of calculating disability from mental disorder Br J Psychiatry 1998;173:123-131.

28. Mathers, C. D. and Loncar, D.Projections of global mortality and burden of disease from 2002 to 2030 PLoS Med 2006;3(11):e442-

29. Murray, CJL and Lopez, ADThe global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020 1996;(1)

30. Murray, C. J.Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010 Lancet 15-12-2012;380(9859):2197-2223.

31. Vos, T., Flaxman, A. D., Naghavi, M., Lozano, R. et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010 Lancet 15-12-2012;380(9859):2163-2196.

32. Ware, J. E.SF-36 Health Survey, Manual and Interpretation 1993;

33. Bijl, R. V., van, Zessen G., Ravelli, A., de, Rijk C. et al. The Netherlands Mental Health Survey and Incidence Study (NEMESIS): objectives and design Soc Psychiatry Psychiatr Epidemiol 1998;33(12):581-586.

34. Salomon, J. A., Vos, T., Hogan, D. R., Gagnon, M. et al. Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010 Lancet 15-12-2012;380(9859):2129-2143.

35. Saarni, S. I., Suvisaari, J., Sintonen, H., Pirkola, S. et al. Impact of psychiatric disorders on health-related quality of life: general population survey Br J Psychiatry 2007;190:326-332.

36. Cuijpers, P. and Smit, F.Excess mortality in depression: a meta-analysis of community studies J Affect Disord 2002;72(3):227-236.

37. Craig, B. M., Busschbach, J. J., and Salomon, J. A.Keep it simple: ranking health states yields values similar to cardinal measurement approaches J Clin Epidemiol 2009;62(3):296-305.

3Effectiveness of psychological interventions in

preventing recurrence of depressive disorder:

meta-analysis and meta-regression

Karolien E.M. Biesheuvel-Leliefeld Gemma D. Kok

Claudi L.H. BocktingPim Cuijpers

Steven D. HollonHarm W.J. van Marwijk

Filip Smit

Published in: Journal of Affective Disorders; 174 (2015) 400-410

R1R2R3R4R5R6R7R8R9


Chapter 3

46

ABSTRACT

Background Major depression is probably best seen as a chronically recurrent disorder, with patients experiencing another depressive episode after remission. Therefore, attention to reduce the risk of relapse or recurrence after remission is warranted. The aim of this review is to meta-analytically examine the effectiveness of psychological interventions to reduce relapse or recurrence rates of depressive disorder.

Methods We systematically reviewed the pertinent trial literature until May 2014. The random-effects model was used to compute the pooled relative risk of relapse or recurrence (RR). A distinction was made between two comparator conditions: (1) treatment-as-usual and (2) the use of antidepressants. Other sources of heterogeneity in the data were explored using meta-regression.

Results Twenty-five randomised trials met inclusion criteria. Preventive psychological interventions were significantly better than treatment-as-usual in reducing the risk of relapse or recurrence (RR=0.64, 95%CI=0.53-0.76, z= 4.89, p<0.001, NNT=5) and also more successful than antidepressants (RR=0.83, 95%CI=0.70-0.97, z=2.40, p=0.017, NNT=13). Meta-regression showed homogeneity in effect size across a range of study, population and intervention characteristics, but the preventive effect of psychological intervention was usually better when the prevention was preceded by treatment in the acute phase (b=-1.94, SEb=0.68, z=-2.84, p=0.005).

Limitations Differences between the primary studies in methodological design, composition of the patient groups and type of intervention may have caused heterogeneity in the data, but could not be evaluated in a meta-regression owing to poor reporting.

Conclusions We conclude that there is supporting evidence that preventive psychological interventions reduce the risk of relapse or recurrence in major depression.


Psychological interventions in recurrent depression; a meta-analysis

47

3

INTRODUCTION

Major depressive disorder (MDD) affects 16% of the population on a lifetime basis1. Of all people with MDD, at least 45% experience recurrences, typically with seven to eight depressive episodes over the course of their life2 and spending as much as 21% of their lifetime in a depressed condition3. MDD is therefore perhaps best seen as a largely chronically recurrent disorder with much of its disease burden stemming from its recurrent nature4.

Treatments for depression are often delivered in an episodic fashion during the acute disease stage. As a result, the burden of depression is only partly alleviated. Therefore, longer-term strategies to reduce the risk of relapse or recurrence are needed. The National Institute for Health and Clinical Excellence5 recommends to continue antidepressant medication (ADM) in medication-responders for at least 6 months after remission or even to continue for at least 2 years if there is a significant risk of relapse. The strength of these recommendations are under debate as the optimal duration of the continuation- or maintenance phase has not been studied well enough6,7, there is divergent information about discontinuation of continuation- or maintenance ADM and the relation to relapse or recurrence8 and reported levels of non-adherence have been consistently high. Continuation of ADM may therefore not be the most optimal strategy from a clinical and a public health perspective.

Fortunately, increasing attention is being paid to preventive psychological interventions after remission. NICE recommends providing individual cognitive behavioural therapy (CBT) for people who have relapsed despite antidepressants and for people with a significant history of depression and residual symptoms despite treatment. For people who are currently well but have had 3 or more episodes of depression, NICE recommends to provide mindfulness-based cognitive therapy (MCT). These recommendations are based on data from controlled studies on the efficacy of psychological interventions after remission from a depressive episode. In 2007, a meta-analysis on cognitive (behavioural) interventions versus non-active (assessment only) and active controls (ADM and treatment-as-usual (TAU)) in recurrent depression was conducted by Vittengl et al. Their conclusion was that among acute-phase responders, cognitive therapy (CT) reduced relapse–recurrence significantly compared with non-active controls (assessment only) at the end of continuation treatment (21% reduction) and at follow-up (29% reduction). CT also reduced relapse–recurrence compared with active controls at the end of continuation treatment (non-significant reduction, 12%) and at follow-up (significant reduction, 14%). A meta-analysis by Guidi et al9 showed that the sequential administration of psychotherapy (alone or in combination with antidepressant medication) may have a protective effect against relapse or recurrence in MDD versus active control (ADM and TAU). A meta-analysis by Piet et al10 indicates that mindfulness-based cognitive therapy after remission is an effective intervention for

R1R2R3R4R5R6R7R8R9


Chapter 3

48

the prevention of relapse when compared to TAU and placebo. These meta-analyses show that several psychological interventions reduce the risk on relapse or recurrence. However, there are four important drawbacks: 1) Previous meta-analyses do not include all types of psychological interventions. Vittengl11 merely included CBT, Piet10 merely included MCT and Guidi9 included CBT and MCT. In our meta-analysis we expand on these meta-analyses by including more types of psychological interventions (e.g. problem-solving therapy and psychodynamic/ psychoanalytic therapy), and also by including various modes of delivery (e.g. booster sessions and therapy over the Internet). 2) Results of previous meta-analyses are difficult to interpret as their active controls-group are defined rather broadly. Vittengl11 defined active control as ‘ADM or another active therapy (e.g. MCT)’, Guidi9 defined active control as ‘ADM or TAU’ and Piet10 did not include ADM as comparator at all. However, according to NICE guidelines, continuation of ADM is the first step in the treatment of recurrent depression. In order to draw conclusions on the use of ADM versus a preventive psychological intervention in remitted patients, our argument for the present meta-analysis is to more clearly define active control as ‘ADM (possibly plus TAU)’ and non-active controls ‘TAU only’. 3) Two meta-analyses, Vittengl12 and Guidi9, report on the effect of psychological interventions in ‘acute-phase responders’. In our meta-analyses we include patients ‘who are in remission’. These patients may have had a directly preceding acute-phase intervention but this is not necessary. This corresponds best with real life practice where patients recover, stay depression-free for a while - with or without continuation or maintenance treatment- and then relapse or recur. 4) Previous meta-analyses are not up-to-date (2007 and 2011) and need an update including new and current trials in this dynamic field.

In our meta-analysis we summarize the current randomised trial literature of psychological interventions for the prevention of relapse or recurrence following (partial) remission of depressive disorder, and evaluate to what extent these interventions are effective compared to TAU and ADM. Our hypotheses are as follows: (H1) Psychological interventions are superior to TAU; (H2) Psychological interventions are not inferior to ADM and; (H3) The different types of psychological interventions are equally effective. We also assess the effect of several moderators (mean age, number of previous episodes, percentage females, previous acute-phase treatment (if any), duration of preventive intervention, length of follow-up and setting (community, primary care, secondary care)).

This meta-analysis assesses the effects of preventive psychological interventions, ADM and TAU on reducing the risk of relapse of recurrence. Assessing which intervention works best for remitted patients who are at risk for relapse or recurrence may be an important step in improving existing guidelines and ultimately the outcomes of treatment.



49

3

METHODS

Primary studiesWe included studies in the meta-analysis when the following inclusion criteria were met: a) a randomised controlled trial b) examining adult patients in the age bracket of 18-64 year c) with recurrent MDD d) who were in remission (according to their own definition in the individual trial-paper) at randomisation e) receiving a preventive psychological intervention f) with the aim of reducing the risk of relapse or recurrence and g) with a comparison to a control condition. Control conditions could be classed as TAU (routine clinical management, assessments only, no treatment and waiting-list control with unrestricted access to TAU) or ADM. Each study had to report relapse or recurrence rates using established screeners with a pre-defined cut-off point for MDD, such as the Hamilton Rating Scale of Depression (HRSD)13 and Beck Depression Inventory (BDI)14, or a diagnostic interview such as the Structured Clinical Interview for DSM-IV Axis 1 Disorders (SCID-1)15.

Psychological interventions could be classed as ‘cognitive (behavioural) therapy’ (CT), ‘mindfulness-based cognitive therapy’ (MCT), ‘interpersonal therapy’ (IPT), problem-solving therapy (PST) and psychodynamic- (psychoanalytic) therapy (PDT). Besides the more usual mode of delivery of interventions (e.g. group- or face-to-face interventions) we included all modes of delivery like booster-sessions during follow-up and therapy over the Internet. Table 1 summarizes these interventions including a brief overview.

Table 1. Psychological interventions included in the meta-analysis

Name therapy Approach

Cognitive (behavioural) therapy (CT)

Negative automatic thoughts, maladaptive information processing, and avoidance behaviour play a key role in the development and recurrence of depression16

Mindfulness based cognitive therapy (MCT)

Protocol-led, group-based skills training program designed to teach recovered depressed patients how to disengage from automatic, cognitive processing patterns linked to relapse17.

Interpersonal therapy (IPT)

Originates from interpersonal theory by Klerman et al. 18. It links stressful life events and insufficient social support to the development and recurrence of depressive symptoms19.

Problem solving therapy (PST)

Brief treatment focused on strengthening practical problem-solving skills. The goal is to stimulate an active attitude towards everyday problems and, hereby, to achieve a reduction in mental health problems20

Psychodynamic therapy (PDT)

Focuses on the affective, behavioural and cognitive aspects of relationships from a psychodynamic point of view21,22. It comprises intervention methods such as clarification, interpretation and confrontation each addressing intra-psychic conflict and resistance23.

R1R2R3R4R5R6R7R8R9


Chapter 3

50

Central clinical end-termThe central clinical outcome was the relapse or recurrence rate of MDD as defined by study investigators (i.e. crossing the cut-off on a depression rating scale or a change in diagnostic depression status based on clinical assessment). Outcomes were evaluated at the longest available follow-up.

Search methods for identification of studiesA literature search was conducted in May 2014. Free text and MeSH terms were used for searches in Medline, PsycInfo, CINAHL, Embase and the Cochrane database. The studies had to be published in English. Keyword searches were conducted by combining the following main terms: cognitive, cognitive behaviour therapy, mindfulness, mindfulness-based cognitive therapy, interpersonal therapy, problem-solving, problem-solving therapy, psychodynamic, psychodynamic therapy, psychoanalytic, psychoanalytic therapy, continuation, maintenance, relapse, recurrence, prevention, therapy, treatment, recurrent, recurrence, depressive disorder and depression. Additional delimiters were adults and randomised controlled trials (Appendix S1). To supplement the searches of published research, the Internet was also utilised to find additional studies.

Data collection and analysisSelection of trialsStudies were searched, selected and reviewed in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses, PRISMA24 (Appendix S2). The first selection was the responsibility of the first author (KBL) and was made using the title, abstract and keywords whereby the full-text article was retrieved when in doubt. All authors of significant papers in the research field were contacted and asked to complete the list of selected publications. Two independent researchers (KBL and GK) carried out the final selection. Any disagreement was resolved by consensus.

Assessment of risk of bias in included studies Flaws in the design, conduct, analysis, and reporting of randomised trials can cause the effect of an intervention to be underestimated or overestimated. The Cochrane risk-of-bias method25 was applied for assessing risk of bias to make the process clearer and more accurate (Appendix S3). This Cochrane list consists of six items. Two items assess the strength of the randomisation process in preventing selection bias in the assignment of participants to interventions: adequacy of sequence generation and allocation concealment. The third item (masking) assesses the influence of performance bias on the study results. The fourth item assesses the likelihood of incomplete outcome data, which raises the possibility of bias in effect estimates. The fifth item assesses selective reporting, the tendency to preferentially



51

3

report statistically significant outcomes. This item requires a comparison of published data with trial protocols, when such are available. The final item refers to other sources of bias that are relevant in certain circumstances such as sponsorship bias.

Data extractionWe collated an evidence table in which extracted data of each study was recorded. Two reviewers (KBL and GK) extracted the data independently. The initial inter-rater agreement reached in extracting data ranged from 85% to 100%. All disagreements were discussed between the two reviewers and consensus was obtained between these two raters.Extracted data included mean age, number of previous episodes, percentage females, type of previous intervention and comparator (if any), type of current intervention and comparator, type and duration of current intervention, length of follow-up, setting (community, primary care, secondary care), number of patients per study-arm, definition of relapse or recurrence and relapse or recurrence rates.

Data analysisThe primary outcome in this meta-analysis was the reduction in the relapse or recurrence rate in the intervention group as compared with the comparator condition. This gave rise to an effect size called relative risk (RR). A RR below 1 indicates that the intervention is more effective than the comparator condition, because fewer relapses or recurrences occur.

The meta-analysis was based on DerSimonian and Laird’s random-effects model26, because heterogeneity was likely to be substantial in the context of various intervention types and comparator conditions, while follow-up measurements ranged from 17 to 332 weeks. An α-level of 0.05 (2-tailed) was used for hypothesis testing. In addition to the RR, the risk-difference (RD) was calculated and transformed by inversion into the number-needed-to-treat (NNT).

Heterogeneity was evaluated using the I2 statistic27 and can be interpreted as the percentage of between-study variance that cannot be explained by random sample error of the primary studies alone. As a rule, heterogeneity is deemed low, moderate or high when I2 is 25%, 50% or 75%, respectively. The 95% confidence-interval of I2 was estimated using STATA’s downloadable ‘heterogi’-procedure.

The presence of publication bias was evaluated using Duval and Tweedie’s Trim & Fill procedure28. Essentially, this procedure re-estimates the meta-analytically pooled effect size after considering publication bias by imputing missing studies. The bias can then be observed as the difference between the unadjusted pooled effect size and the adjusted one. We also computed the fail-safe N for the pooled RR as another way to gauge the robustness of the pooled RR in the possible presence of publication bias.

R1R2R3R4R5R6R7R8R9


Chapter 3

52

The correlation between the effect size of the interventions and the characteristics of the primary studies was explored using meta-regression. In meta-regression, the effect size (RR) of each the primary studies is regressed on the characteristics of the studies, the study population and the intervention. The meta-analytic regression model contained seven predictor variables: mean age, number of previous episodes, percentage females, type of previous treatment (if any), duration of preventive intervention, length of follow-up and setting (community, primary care, secondary care).

Subgroup analyses were performed to study the results on long effectiveness of psychological interventions (one year or more, two years or more) and on the use of diagnostic interviews at follow up.

The meta-analytic dataset was analysed with help of Comprehensive Meta-Analysis (CMA, Version 2.2.057, 2010) (2012, http://www.meta-analysis.com). Stata (StataCorp, Version 8.2, 2009) was used for carrying out the multivariate meta-regression and calculating the 95% confidence intervals of the I2-statistic. All findings were summarised in a table according to the methodology described by the GRADE working group29 (Appendix S4).

RESULTS

Description of included studiesHaving examined a total of 3,537 abstracts, we retrieved 69 full text papers. Of these, 44 studies were excluded because they did not meet the inclusion criteria (Appendix S5). The remaining 25 studies met all inclusion criteria. Five trials compared 3 conditions, thus testing multiple contrasts. As a result, this meta-analysis was based on 25 studies and 30 contrasts. In case of multiple contrasts, data on variables (sample size of control, mean age, etc.) were adjusted accordingly per arm. Figure 1 depicts the flow chart of the selection process.

No trials evaluating PST or PDT met the inclusion criteria. Sixteen trials (17 contrasts) evaluated preventive CT, 3 trials (6 contrasts) evaluated IPT, and 6 trials (7 contrasts) evaluated MCT. Thirteen contrasts compared a psychological intervention with ADM and 17 contrasts compared a psychological intervention with TAU. Fourteen studies were conducted in Europe, 10 in the United States and 1 in Australia. Duration of follow-up ranged between 17 weeks and 332 weeks (mean follow-up 115 weeks).

CT after remission was delivered through various modes; weekly group sessions, individual sessions, over the Internet and as booster sessions (various number of sessions during various duration of periods with a minimum of 3 sessions). All trials evaluating MCT consisted of two-hour weekly sessions over 8 consecutive weeks, eventually followed by a few booster sessions. IPT was delivered in individual sessions (varying from monthly maintenance sessions over 8 months to weekly maintenance sessions over 4 months).



53

3

Records identified through database searching

Medline (2,384)Psychinfo (855)CINAHL (206)

Embase (1,330)Cochrane (922)

(n = 5,697)

Screen

ing

Includ

edEligibility

Iden

tification

Additional records identified through other sources

(n = 2)

Number of duplicates removed (n = 2,162)

Records excluded(n =3,468)

Full-text articles assessed for eligibility

(n = 69) Full-text articles excluded, with reasons

(n = 44)

No relapse rates, 11Follow up after acute-phase, 8No RCT, 11No psychological treatment, 2No recovered participants, 2Other, 10

Studies included in meta-analysis

(n = 25 )

Records screened by abstract and title(n = 3,537)

Figure 1. PRISMA Flow chart of the literature search

The 25 primary studies encompassed 2,055 patients in total. In 21 contrasts the patients had been recipients of a preceding acute-phase therapy during the same trial, which was either medication or cognitive therapy or a combination of both. At randomisation, all patients were depression free (with or without residual symptoms) and therefore ‘at risk’ of a relapse or recurrence into yet another episode of MDD. A total of 932 patients were randomised to an intervention condition: 529 received preventive CT, 142 IPT and 261 MCT. The remaining 1,123 patients were randomised to comparator conditions: 670 receiving TAU and 453 receiving ADM.

Mean age was 43.3 years (95%CI, 43.11-43.39), 70.4% of the participants was female (95%CI, 69.9-70.91) and the mean number of previous episodes was 3.8 (95%CI, 3.77-3.92). Selected characteristics are presented in Table 2.

R1R2R3R4R5R6R7R8R9


54

Tabl

e 2.

Sel

ecte

d ch

arac

teri

stic

s of

25

incl

uded

stu

dies

a

Aut

hor

Year

Mea

n ag

e %

Fe

mal

e no

prev

ious

ep

isod

es

Prev

ious

in

terv

enti

onb

Prev

ious

co

mpa

rato

rbC

urre

ntin

terv

enti

onC

urre

nt

com

para

tor

Follo

w-u

p (w

ks)

Defi

niti

on

recu

rren

ceSe

ttin

gLe

ngth

of

inte

rven

tion

(w

ks)

Ris

k ra

te

inte

rven

tion

Ris

k ra

te

com

para

tor

Bak

er e

t al 30

1985

4074

n/a

Gro

up C

BT

Gro

up C

BT

CB

T TA

U22

BD

I ≥ 1

7 co

mm

unity

126/

107/

9

Bla

ckbu

rn e

t al

3119

864

842

CT

AD

MC

TA

DM

104

HR

SD ≥

8 a

nd

BD

I ≥ 9

or

retr

eatm

ent

prim

ary

and

seco

ndar

y ca

re26

3/13

7/9

Bla

ckbu

rn e

t al

3219

9740

593

CT

AD

MC

TA

DM

52H

RSD

≥ 1

4se

cond

ary

care

104

4/17

4/13

Boc

ktin

g et

al 33

2009

4573

4n/

an/

aC

T +

TA

UTA

U28

6M

DE

acco

rdin

g to

SC

IDco

mm

unity

, pr

imar

y an

d se

cond

ary

care

869

/88

73/8

4

Bon

dolfi

et a

l 3420

1048

724

n/a

n/a

MC

T +

TA

UTA

U60

MD

E ac

cord

ing

to

SCID

com

mun

ity,

prim

ary

and

seco

ndar

y ca

re

89/

2710

/28

Con

radi

et a

l 3520

0744

61no

t cle

arn/

an/

aC

BT

TAU

156

MD

E (C

IDI)

prim

ary

care

156

21/3

839

/62

Fava

et a

l 3619

9846

68no

t cle

arA

DM

AD

MC

TTA

U33

2M

DE

(RD

C

defin

ed)

seco

ndar

y ca

re20

10/2

015

/20

Fava

et a

l 3720

0244

604

AD

MA

DM

CT

+ A

DM

AD

M

60M

DE

(RD

C

defin

ed)

seco

ndar

y ca

re6

1/4

4/4

Fava

et a

l 3820

0447

604

AD

MA

DM

CT

TAU

33

2M

DE

(RD

C

defin

ed)

seco

ndar

y ca

re20

8/20

18/2

0

Fran

k et

al a

3919

9040

777

AD

M +

IPT

AD

M +

IPT

IPT

TAU

156

MD

E (R

DC

de

fined

) + H

SRD

≥

15 +

Ras

kin

≥7

unkn

own

156

1/26

1/23

Fran

k et

al b

39

1990

4077

7A

DM

+ IP

TA

DM

+ IP

TIP

TA

DM

156

MD

E (R

DC

de

fined

) + H

SRD

≥

15 +

Ras

kin

≥7

unkn

own

156

1/26

0/28

God

frin

et a

l 4020

1046

81no

t cle

arn/

an/

aM

CT

+ T

AU

TAU

56M

DE

acco

rdin

g to

D

SM-I

Vse

cond

ary

care

812

/40

32/4

7

Hol

land

are

et

al 41

2011

4585

6n/

an/

aC

BT

(inte

rnet

)TA

U26

MD

D a

ccor

ding

to

SC

IDco

mm

unity

104/

3814

/37

Hol

lon

et a

l 4220

0540

592

CT

AD

MC

T A

DM

104

MD

E or

HR

SD

≥ 14

, at l

east

2

wee

ks

seco

ndar

y ca

re52

5/20

7/14

Jarr

ett e

t al a

4320

0041

842

CT

TAU

CT

TAU

104

MD

E (R

DC

de

fined

) or

retr

eatm

ent

seco

ndar

y ca

re10

3/7

6/7

Jarr

ett e

t al b

4320

0041

842

CT

AD

MC

TA

DM

104

MD

E (R

DC

de

fined

) or

retr

eatm

ent

seco

ndar

y ca

re10

3/7

4/7


55

3

Aut

hor

Year

Mea

n ag

e %

Fe

mal

e no

prev

ious

ep

isod

es

Prev

ious

in

terv

enti

onb

Prev

ious

co

mpa

rato

rbC

urre

ntin

terv

enti

onC

urre

nt

com

para

tor

Follo

w-u

p (w

ks)

Defi

niti

on

recu

rren

ceSe

ttin

gLe

ngth

of

inte

rven

tion

(w

ks)

Ris

k ra

te

inte

rven

tion

Ris

k ra

te

com

para

tor

Jarr

ett e

t al 44

2001

4373

3C

TC

TC

TTA

U10

4M

DD

(DSM

de

fined

)co

mm

unity

, pr

imar

y an

d se

cond

ary

care

3615

/41

22/4

3

Jarr

ett e

t 4520

1343

674

CT

n/a

CT

AD

M14

0M

DD

(DSM

de

fined

, LIF

E se

cond

ary

care

seco

ndar

y ca

re34

11/2

512

/28

Kle

in e

t al 46

2004

4567

2C

BASP

(CB

T)C

BASP

(CB

T)C

BASP

(CB

T)TA

U52

PSR

=5

or 6

, 2

cons

wee

ks)

MD

D a

nd H

RSD

≥

16 fo

r ≥2

vis

its

seco

ndar

y ca

re52

1/42

8/40

Kle

rman

et

al a

4719

74no

t cl

ear

100

not c

lear

AD

MA

DM

IPT

TAU

35no

t cle

arse

cond

ary

care

364/

259/

25

Kle

rman

et a

l b

4719

74no

t cl

ear

100

not c

lear

AD

MA

DM

IPT

AD

M35

not c

lear

seco

ndar

y ca

re36

4/25

3/25

Kuyk

en e

t al 48

2008

4977

6n/

an/

aM

CT

+ T

AU

AD

M65

MD

E ac

cord

ing

to

SCID

prim

ary

care

5229

/61

37/6

2

Ma

et a

l 4920

0445

763

n/a

n/

aM

CT

+ T

AU

TAU

60M

DE

(DSM

-IV

de

fined

)pr

imar

y ca

re

and

com

mun

ity36

14/3

623

/37

Payk

el e

t al 50

2005

449

2n/

an/

aC

BT

+ A

DM

AD

M27

5M

DD

>4

wee

ks

or H

AM

D ≥

13,

at

leas

t 8 w

eeks

seco

ndar

y ca

re32

48/8

051

/78

Perl

is e

t al 51

2002

4055

5A

DM

AD

MC

T +

AD

MA

DM

28M

DE

at a

ny v

isit,

H

RSD

≥ 1

5 at

two

cons

ecut

ive

visi

ts

seco

ndar

y ca

re26

4/66

5/66

Schu

lber

g et

al

a 52

1996

3883

not c

lear

IPT

TAU

IPT

TAU

17sy

mpt

omat

ic

(HR

SD≥1

3)pr

imar

y ca

re18

17/9

144

/92

Schu

lber

g et

al

b 52

1996

3883

not c

lear

IPT

AD

MIP

TA

DM

17sy

mpt

omat

ic

(HR

SD≥1

3)pr

imar

y ca

re18

17/9

123

/91

Sega

l et a

l a 53

2010

4359

5A

DM

AD

MM

CT

TAU

78H

RSD

≥ 1

6 2

cons

ecut

ive

wee

ks

+ M

DE

on S

CID

com

mun

ity,

prim

ary

and

seco

ndar

y ca

re

810

/26

18/3

0

Sega

l et a

l b 53

2010

4561

5A

DM

AD

MM

CT

AD

M78

HR

SD ≥

16

2 co

nsec

utiv

e w

eeks

+

MD

E on

SC

ID

com

mun

ity,

prim

ary

and

seco

ndar

y ca

re

810

/26

13/2

8

Teas

dale

et a

l 5420

0041

715

n/a

n/a

MC

T +

TA

UTA

U60

reco

very

or

rem

issi

on, H

RSD

-17

<10

com

mun

ity,

prim

ary

and

seco

ndar

y ca

re

2631

/71

38/6

6

a A

bbre

viat

ions

: A

DM

, an

ti-de

pres

sant

med

icat

ion;

BD

I, B

eck

Dep

ress

ion

Inve

ntor

y 14

; C

BASP

; ‘C

ogni

tive

Beh

avio

ural

Ana

lysi

s Sy

stem

of

Psyc

hoth

erap

y’;

C(B

)T,

Cog

nitiv

e (B

ehav

iour

) Th

erap

y; D

SM-I

V, D

iagn

ostic

and

Sta

tistic

al M

anua

l of M

enta

l Dis

orde

rs-I

V; H

RSD

, Ham

ilton

Rat

ing

Scal

e fo

r D

epre

ssio

n13; I

PT, I

nter

pers

onal

The

rapy

; MC

T, M

indf

ulne

ss B

ased

Cog

nitiv

e Th

erap

y; M

DD

, Maj

or D

epre

ssiv

e D

isor

der;

MD

E, M

ajor

Dep

ress

ive

Epis

ode;

Ras

kin,

Ras

kin

Dep

ress

ion

Rat

ing

Scal

e55; R

CT,

rand

omis

ed c

ontr

olle

d tr

ial;

RD

C, R

esea

rch

Dia

gnos

tic C

rite

ria;

SC

ID, S

truc

ture

d C

linic

al In

terv

iew

for

Dep

ress

ion15

; TA

U, t

reat

men

t-as

-usu

al; w

ks, w

eeks

.b A

n in

terv

entio

n in

the

acut

e ph

ase

as p

art o

f thi

s sp

ecifi

c R

CT

R1R2R3R4R5R6R7R8R9


Chapter 3

56

Psychological interventions versus treatment-as-usual Pooled RRWe obtained seventeen randomised trials that compared a psychological intervention with TAU (routine clinical management, assessments only, no treatment, waiting-list control with unrestricted access to TAU). The pooled relative risk was 0.64 (95%CI 0.53-0.76) and was statistically significant (z= 4.89, p<0.001), indicating that psychological interventions were more successful in decreasing the risk of relapse or recurrence than TAU over a mean follow-up time of two-years (115 weeks) (Figure 2).

Figure 2. Forest plot of risk ratios and 95% confidence-intervals for psychological interventions versus treatment-as-usuala

a Abbreviations: CI, confidence interval

Pooled RD and NNTThe risk-difference was 0.19 (95%CI 0.13-0.26), corresponding to a number-needed-to-treat (NNT) of 5.3. In other words, it takes 5 patients to be treated with a psychological intervention (CT, MCT or IPT) rather than TAU to prevent one relapse or recurrence.



57

3

HeterogeneityThe test of heterogeneity indicated that the observed variability in effect sizes across the studies was greater than that expected by chance alone (χ2 =32.41, df=16, p=0.009). The corresponding I2 was 51% (95%CI 14-72%) corresponding with moderate heterogeneity56.

Meta-regression A meta-regression analysis was used to identify sources of heterogeneity in the effects across the studies in terms of sample characteristics (mean age, gender, number of previous depressive episodes) and the study’s methodological characteristics (previous intervention in the acute phase (if any), mean treatment duration, mean follow-up duration and setting). Whether or not there was any intervention in the acute phase (CT, MCT, IPT, ADM or combination) helped partially explain heterogeneity across outcomes: the preventive effect of psychological intervention was usually better when the prevention was preceded by treatment in the acute phase (b=-1.94, SEb=0.68, z=-2.84, p=0.005). The other investigated variables in the meta-regression were not associated with the effect size. Other sources of heterogeneity (e.g. variability in delivery mode of intervention and comorbidity) may also have contributed to the heterogeneity, but were insufficiently reported to be included in the meta-regression.

Publication biasThe Egger test showed that publication bias was likely in this meta-analysis of psychological interventions versus TAU (intercept = -1.65, SE=0.41, p<0.001). This was confirmed by Duvall and Tweedie’s adjusted estimate of RR=0.82 (95%CI 0.68-0.99), based on eight additionally imputed studies. The adjusted estimate differed somewhat from the original (unadjusted) estimate of RR=0.64 (95%CI 0.53-0.76): the 95% confidence intervals have a considerable overlap, but the point estimates of RR fall outside the alternative intervals. That said, the conclusion that psychological interventions are statistically superior to TAU remained unaffected57,58. The robustness of the findings was further supported by a fail-safe N of 197, meaning that 197 undetected studies with no effect (RR=1) need to be included in the meta-analysis before the pooled effect would cease to be statistically significant at p<0.05 (2-tailed).

Psychological interventions versus antidepressant medicationThirteen studies compared psychological interventions with ADM (Figure 3). The pooled effect size was RR=0.83 (95%CI 0.70-0.97), which was statistically significant (z=2.39, P=0.017). No evidence was obtained for heterogeneity (χ2= 9.09, df=12, P=0.695, I2= 0%, 95%CI 0-57), although the wide 95% CI of the I2 leaves room for other interpretations. The mean follow-up period was somewhat less than two years (90 weeks). The risk-difference

R1R2R3R4R5R6R7R8R9


Chapter 3

58

was 0.075 (95%CI 0.001-0.149) and the corresponding NNT was 13.3, demonstrating that 13 patients would need to be treated with a preventive psychological intervention rather than ADM to prevent one relapse or recurrence.

Figure 3. Forest plot of risk ratios and 95% confidence-intervals for psychological interventions versus antidepressant medicationa

a Abbreviations: CI, confidence interval

The Egger test and Duvall and Tweedie’s Trim & Fill did not suggest presence of a significant publication bias (Egger test intercept = -0.47, SE=0.41, P=0.27, Duvall and Tweedie’s adjusted RR=0.86, 95%CI 0.73-1.00, based on two imputed studies).

Three trials comparing preventive psychological interventions with ADM (Paykel50, Perlis51 and Fava37), allowed ADM intake in the intervention group. In a subgroup analysis of preventive psychological interventions versus ADM without these three trials the pooled effect size was RR=0.78 (95%CI=0.625-0.961) and did not change results.

The effectiveness of different types of psychological interventionsSixteen trials (17 contrasts) included CT, 3 trials included IPT (6 contrasts) and 6 trials (7 contrasts) included MCT (Table 3). The effect sizes of the different psychological interventions were roughly similar.



59

3

Table 3. Risk ratios of different types of psychological interventions versus treatment-as-usual (TAU) and antidepressant medication (ADM) according to the random-effects model (DerSimonian and Laird)a

Intervention KRR D-L

(95%CI) Test RD

(95%CI) NNT I2

versus TAU* all 17 0.64(0.53; 0.76)

z=-4.89P=0.000

-0.190(-0.255; -0.125)

5 51%

CT 9 0.68 (0.54; 0.87)

z=-3.12 P=0.002

-0.196(-0.28; -0.11)

5 52%

MCT 5 0.66 (0.53; 0.82)

z=-3.81 P=0.000

-0.205(-0.32; -0.09)

4 0%

IPT 3 0.41(0.27; 0.63)

z=-4.10 P=0.000

-0.160(-0.37; -0.04)

6 0%

versus ADM* all 13 0.83(0.70; 0.97)

z=-2.39P=0.017

-0.075(-0.149; -0.001)

13 0%

CT 8 0.79(0.61; 1.02)

z=-1.80 P=0.072

-0.16 (-0.30; -0.016)

6 9%

MCT 2 0.80(0.60; 1.08)

z=-1.46 P=0.146

-0.11(-0.25; -0.04)

9 0%

IPT 3 0.83(0.50; 1.38)

z=-0.71 P=0.477

-0.002(-0.068; -0.073)

499 0%

a Abbreviations: ADM, anti-depressant medication; CT, Cognitive (Behaviour) Therapy; CI, confidence interval; I2, heterogeneity; IPT, Interpersonal Therapy; K, number of contrasts; MCT, Mindfulness-based Cognitive Therapy; NNT, number-needed-to-treat; RD, risk difference; RR D-L, random-effects according to DerSimonian and Laird; TAU, treatment-as-usual

Subgroup analysesSubgroup analyses of psychological interventions versus TAU and ADM on results on long effectiveness (follow-up more than one year and more than two years) and on the use of diagnostic interviews at follow-up showed no remarkable differences.

Quality of included studies We created GRADE profiles and classified the overall quality of the evidence (high, moderate, low) based on the GRADE system using 6 criteria29; study design (all RCT’s), study limitations, inconsistency, indirectness, imprecision and other bias (e.g. publication bias). Overall, the quality of the studies was low (Appendix S4). We conducted a meta-regression to analyse whether the size of the effects (RR) systematically co-varied with study quality. This was not the case (b= 0.242, SEb=0.945, z=0.26, p=0.798).

R1R2R3R4R5R6R7R8R9


Chapter 3

60

DISCUSSION

Main findingsWe obtained 25 randomised controlled trials with a total of 2,055 patients examining the effect of psychological interventions to prevent the occurrence of yet another depressive episode. The psychological interventions were based on cognitive (behavioural) therapy, mindfulness-based cognitive therapy and interpersonal therapy. We found no prevention trials based on problem-solving therapy or psychodynamic therapy that met inclusion criteria. We hypothesised (H1) that psychological interventions were superior to TAU. Meta-analysis of the available studies indeed demonstrated that psychological interventions were considerably more effective than TAU in preventing relapse or recurrences over two years (RR=0.64, p<0.001; NNT=5). After correcting for publication bias, the adjusted estimate -based on eight additionally imputed studies- appeared less effective (adjusted RR= 0.82, 95%CI 0.68-0.99). The heterogeneity in this meta-analysis was moderate. Second meta-analysis, we hypothesised (H2) that psychological interventions were not inferior to ADM. Indeed, we found that psychological interventions were more effective in reducing the risk of a relapse or recurrence compared to ADM (RR=0.83, p=0.017; NNT=13). After correcting for publication bias, the adjusted estimate -based on two additionally imputed studies- appeared slightly less effective (adjusted RR=0.86, 95%CI 0.73-1.00). In this meta-analysis, we found no evidence of heterogeneity between studies. The effect sizes of the different psychological interventions were roughly similar, as stated in our third hypothesis (H3). An unanticipated finding was that the effectiveness of preventive interventions was enhanced when the patient had received an intervention (psychological intervention, antidepressant medication or both) during the acute phase of the depression in the same trial.

Strengths and limitationsThis is the first systematic review and meta-regression investigating the effect of all types of preventive psychological interventions on relapse or recurrence after remission. Moreover, this meta-analysis includes ADM and TAU as separate control groups which is new and very much awaited in the field, as ADM is a separate guideline-based treatment choice. A large number of studies (25) are included in this review and current state-of-the-art meta analytic techniques are used.

We also recognize a number of limitations. First, there were many differences in methodological design of the included trials such as definition of remission, recovery, relapse and recurrence, type or duration of interventions and whether or not there was a preceding acute intervention in the same trial. For example, treatment-as-usual was often described inadequately and information on the exact method of determining recurrence (interview vs. questionnaire) was sometimes not provided. These differences are likely to



61

3

have caused some heterogeneity in the data, but owing to e.g. poor reporting could not be evaluated in a meta-regression.

A second limitation, which is a limitation of the included studies, is the low overall quality of the studies according to the GRADE evidence profile. This lack of quality limits our confidence in the overall effect size estimates. However, due to the nature of psychological interventions and the difficulty to compare these with ‘placebo’ interventions, highest quality evidence may not be possible in such studies. Factors that lower the quality of evidence, like differences in interventions (e.g. client-therapist relations) and difficulties in blinding of participants, outcome reporters and/or personnel, will probably remain in studies investigating the effect of psychological interventions.

Third, publication bias could not be ruled out in the meta-analysis of psychological interventions versus TAU and it is possible that unpublished trials showed null findings or even adverse outcomes. After correcting for publication bias, the adjusted estimate appeared less effective than the original estimate.

Fourth, it is unlikely that the exact number of previous depressive episodes has been reported reliably in the primary studies. The number of previous episodes is most likely underreported59. The fact that we did not find an effect of number of previous episodes in our meta-analysis might be caused by a downward bias due to under-reporting.

Fifth, we found that the effect sizes of the different psychological interventions were roughly similar. However, this finding is not based on (scarce) head-on comparisons but on the pooled results of studies that compared the preventive psychological interventions with ADM or TAU. Results on head-on comparisons of psychological interventions (e.g. MCT vs IPT) should be analysed in order to draw definite conclusions about the effect sizes of different psychological interventions.

Implications Treatment of acute depression is the core business of mental health care, but this approach is only partially successful in reducing the overall disease burden stemming from depression. As said, depression is characterized by a large number of patients experiencing multiple relapses and recurrences, with patients spending as much as 21% of their lifetime in a depressed condition3. This has important implications for the longer-term management of depression.

The National Institute for Health and Clinical Excellence5 recommends to continue medication in ADM responders for at least 6 months after remission or even to continue for at least 2 years if there is a significant risk of relapse. However, not all remitted patients feel comfortable taking antidepressants for a long period, they may feel unnecessarily dependent on them, and may find it difficult to adhere over extended periods of time60. Thus, an alternative for the longer-term use of ADM is sorely needed.

R1R2R3R4R5R6R7R8R9


Chapter 3

62

Psychological interventions in the acute phase have been studied extensively and more and more attention is being paid to the effects of psychological interventions after remission. Examples are the meta-analyses by Guidi et al9, Vittengl et al11 and Piet et al10, who conclude that (some types of) preventive psychological interventions after remission are effective in reducing relapse or recurrence. Our meta-analysis of 25 studies adds that preventive psychological interventions outperform continuation of ADM, which is now a first step intervention in remitted patients.

In clinical practice however, an important question is: ‘What works for whom?’. Therefore, NICE recommends specific types of psychological interventions (CT/MCT over others), optimal treatment duration (16 to 20 sessions over 3 to 4 months) and format of intervention (individual CBT/group MCT) and takes into account the number of previous episodes. In our meta-regression we found homogeneity in effect size across the different modalities and target populations so we could not make a meaningful distinction between the various psychological interventions, treatment durations, settings nor previous episodes. This seems to suggest that practically no stratification is necessary. The only unanticipated finding in our meta-analysis was that the effectiveness of preventive interventions was enhanced when the patient had received an intervention (psychological intervention, antidepressant medication or both) during the acute phase of the depression in the same trial. In other words, it might be important to look at depression as a series of disease-stages that need to be taken care of both during the acute-phase and also directly after remission in the continuation- or maintenance phase. Treatment after remission should then be embedded in an integrated way, seamlessly following up on the acute-phase treatment.

Finally; against a background of financial constraints in many health care systems, cost of treatment is of great concern. Cost-effective solutions demand an optimal balance between accessible, acceptable, effective and economically affordable treatments for the many patients suffering from recurrent depressions. Possible ways of offering psychological interventions are over the Internet41,61, by a nurse62, by self-help63 or by low intensity psychological interventions64. More trials focussing on the question how to optimise the cost-effectiveness of psychological interventions after remission is worthy of further consideration and should be placed on the research agenda. Also, more attention should be directed at other psychological interventions like problem-solving therapy and psychodynamic therapy as possibly valuable alternatives in the prevention of recurrent depression. A lot of research is needed to shed light on the unresolved issues. Despite that, this meta-analysis has shown that preventive psychological interventions after remission may reduce the highly significant disease burden stemming from recurrent MDD.



63

3

REFERENCES

1. Kessler, R. C., Chiu, W. T., Demler, O., Merikangas, K. R. et al. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication Arch Gen Psychiatry 2005;62(6):617-627.

2. Kruijshaar, M. E., Barendregt, J., Vos, T., de, Graaf R. et al. Lifetime prevalence estimates of major depression: an indirect estimation method and a quantification of recall bias Eur J Epidemiol 2005;20(1):103-111.

3. Vos, T., Haby, M. M., Barendregt, J. J., Kruijshaar, M. et al. The burden of major depression avoidable by longer-term treatment strategies Arch Gen Psychiatry 2004;61(11):1097-1103.

4. Judd, L. L.The clinical course of unipolar major depressive disorders Arch Gen Psychiatry 1997;54(11):989-991.

5. NICE Clinical GuidelinesNational Institute for Health and Clinical Excellence. Depression: the treatment and management of depression in adults (update). www.nice.org.uk/CG90 2009;

6. Kaymaz, N., van, Os J., Loonen, A. J., and Nolen, W. A.Evidence that patients with single versus recurrent depressive episodes are differentially sensitive to treatment discontinuation: a meta-analysis of placebo-controlled randomized trials J Clin Psychiatry 2008;69(9):1423-1436.

7. Geddes, J. R., Carney, S. M., Davies, C., Furukawa, T. A. et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review Lancet 22-2-2003;361(9358):653-661.

8. Bockting, C. L., ten Doesschate, M. C., Spijker, J., Spinhoven, P. et al. Continuation and maintenance use of antidepressants in recurrent depression Psychother Psychosom 2008;77(1):17-26.

9. Guidi, J., Fava, G. A., Fava, M., and Papakostas, G. I.Efficacy of the sequential integration of psychotherapy and pharmacotherapy in major depressive disorder: a preliminary meta-analysis Psychol Med 2011;41(2):321-331.

10. Piet, J. and Hougaard, E.The effect of mindfulness-based cognitive therapy for prevention of relapse in recurrent major depressive disorder: A systematic review and meta-analysis Clin Psychol Rev 2011;31(6):1032-1040.

11. Vittengl, J. R., Clark, L. A., Dunn, T. W., and Jarrett, R. B.Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitive-behavioral therapy’s effects J Consult Clin Psychol 2007;75(3):475-488.

12. Vittengl, Jeffrey R., Clark, Lee Anna, and Jarrett, Robin B.Continuation-phase cognitive therapy’s effects on remission and recovery from depression J Consult Clin Psychol 2009;77(2):367-371.

13. Hamilton, MA rating scale for depression J Neurol Neurosurg Psychiatry 1960;23:56-62.

14. Beck, A. T, Ward, C. H, Mendelson, M, Mock, J et al. An inventory for measuring depression Arch Gen Psychiatry 1961;4:561-571.

15. First, M. B., Gibbon, M., Spitzer, R. L., and Williams, J. B. W.Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV) 1996;

16. Beck, A. T., Rush, A. J., Shaw, B. F., and Emery, G.Cognitive therapy of depression 1979;

17. Segal, Z., Williams, J. M., and Teasdale, J. D.Mindfulness-based cognitive therapy for depression: A new approach to preventing relapse. 2002;

18. Klerman, G. L., Budman, S., Berwick, D., Weissman, M. M. et al. Efficacy of a brief psychosocial intervention for symptoms of stress and distress among patients in primary care Med Care 1987;25(11):1078-1088.

19. Weissman, M., Markowitz, J. C., and Klerman, G. L.Clinician’s Quick Guide to Interpersonal Psychotherapy. New York: Oxford University Press; 2007. 2007;

20. Hawton, K, Salkovskis, P, Kirk, J, and Clark, DProblem-solving; Cognitive behaviour therapy for psychiatric problems 1989;406-426.

21. De Jonghe F., Rijnierse, P., and Janssen, R.Psychoanalytic supportive psychotherapy J Am Psychoanal Assoc 1994;42(2):421-446.

22. De Jonghe F.Kort en Krachtig (Brief and Potent). Short Psychodynamic Supportive Psychotherapy 2013;

23. Watzke, B., Rueddel, H., Koch, U., Rudolph, M. et al. Comparison of therapeutic action, style and content in cognitive-behavioural and psychodynamic group therapy under clinically representative conditions Clin Psychol Psychother 2008;15(6):404-417.

R1R2R3R4R5R6R7R8R9


Chapter 3

64

24. Moher, D., Liberati, A., Tetzlaff, J., and Altman, D. G.Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement Int J Surg 2010;8(5):336-341.

25. Higgins, JPT and Green, SCochrane Handbook for Systematic Reviews of Interventions 2008;4

26. DerSimonian, R and Laird, N. M.Meta-analysis in clinical trials Control Clin Trials 1986;7:177-188.

27. Higgins, J. P. and Thompson, S. G.Quantifying heterogeneity in a meta-analysis Stat Med 15-6-2002;21(11):1539-1558.

28. Duval, S. and Tweedie, R.Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis Biometrics 2000;56(2):455-463.

29. Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R. et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations BMJ 26-4-2008;336(7650):924-926.

30. Baker, A. L. and Wilson, P. H.Cognitive-behavior therapy for depression: The effects of booster sessions on relapse Behav Ther 1985;16(4):335-344.

31. Blackburn, I. M., Eunson, K. M., and Bishop, S.A two-year naturalistic follow-up of depressed patients treated with cognitive therapy, pharmacotherapy and a combination of both J Affect Disord 1986;10(1):67-75.

32. Blackburn, I. M. and Moore, R. G.Controlled acute and follow-up trial of cognitive therapy and pharmacotherapy in out-patients with recurrent depression Br J Psychiatry 1997;171:328-334.

33. Bockting, Claudi L. H., Spinhoven, Philip, Wouters, Luuk F., Koeter, Maarten W. J. et al. Long-term effects of preventive cognitive therapy in recurrent depression: a 5.5-year follow-up study J Clin Psychiatry 2009;70(12):1621-1628.

34. Bondolfi, Guido, Jermann, Francoise, der Linden, Martial Van, Gex-Fabry, Marianne et al. Depression relapse prophylaxis with Mindfulness-Based Cognitive Therapy: replication and extension in the Swiss health care system J Affect Disord 2010;122(3):224-231.

35. Conradi, Henk Jan, de Jonge, Peter, Kluiter, Herman, Smit, Annet et al. Enhanced treatment for depression in primary care: long-term outcomes of a psycho-educational prevention program alone and enriched with psychiatric consultation or cognitive behavioral therapy Psychol Med 2007;37(6):849-862.

36. Fava, G. A., Rafanelli, C., Grandi, S., Canestrari, R. et al. Six-year outcome for cognitive behavioral treatment of residual symptoms in major depression Am J Psychiatry 1998;155(10):1443-1445.

37. Fava, G. A., Ruini, C., Rafanelli, C., and Grandi, S.Cognitive behavior approach to loss of clinical effect during long-term antidepressant treatment: a pilot study Am J Psychiatry 2002;159(12):2094-2095.

38. Fava, Giovanni A., Ruini, Chiara, Rafanelli, Chiara, Finos, Livio et al. Six-year outcome of cognitive behavior therapy for prevention of recurrent depression Am J Psychiatry 2004;161(10):1872-1876.

39. Frank, E., Kupfer, D. J., Perel, J. M., Cornes, C. et al. Three-year outcomes for maintenance therapies in recurrent depression Arch Gen Psychiatry 1990;47(12):1093-1099.

40. Godfrin, K. A. and van, Heeringen C.The effects of mindfulness-based cognitive therapy on recurrence of depressive episodes, mental health and quality of life: A randomized controlled study Behav Res Ther 2010;48(8):738-746.

41. Hollandare, F., Johnsson, S., Randestad, M., Tillfors, M. et al. Randomized trial of Internet-based relapse prevention for partially remitted depression Acta Psychiatr Scand 2011;124(4):285-294.

42. Hollon, Steven D., DeRubeis, Robert J., Shelton, Richard C., Amsterdam, Jay D. et al. Prevention of relapse following cognitive therapy vs medications in moderate to severe depression Arch Gen Psychiatry 2005;62(4):417-422.

43. Jarrett, R. B., Kraft, D., Schaffer, M., Witt-Browder, A. et al. Reducing relapse in depressed outpatients with atypical features: a pilot study Psychother Psychosom 2000;69(5):232-239.

44. Jarrett, R. B., Kraft, D., Doyle, J., Foster, B. M. et al. Preventing recurrent depression using cognitive therapy with and without a continuation phase: a randomized clinical trial Arch Gen Psychiatry 2001;58(4):381-388.

45. Jarrett, R. B., Minhajuddin, A., Gershenfeld, H., Friedman, E. S. et al. Preventing depressive relapse and recurrence in higher-risk cognitive therapy responders: a randomized trial of continuation phase cognitive therapy, fluoxetine, or matched pill placebo JAMA Psychiatry 2013;70(11):1152-1160.



65

3

46. Klein, D. N., Santiago, N. J., Vivian, D., Blalock, J. A. et al. Cognitive-behavioral analysis system of psychotherapy as a maintenance treatment for chronic depression J Consult Clin Psychol 2004;72(4):681-688.

47. Klerman, G. L., Dimascio, A., Weissman, M., Prusoff, B. et al. Treatment of depression by drugs and psychotherapy Am J Psychiatry 1974;131(2):186-191.

48. Kuyken, Willem, Byford, Sarah, Taylor, Rod S., Watkins, Ed et al. Mindfulness-based cognitive therapy to prevent relapse in recurrent depression J Consult Clin Psychol 2008;76(6):966-978.

49. Ma, S Helen and Teasdale, John D.Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects J Consult Clin Psychol 2004;72(1):31-40.

50. Paykel, E. S., Scott, J., Cornwall, P. L., Abbott, R. et al. Duration of relapse prevention after cognitive therapy in residual depression: follow-up of controlled trial Psychol Med 2005;35(1):59-68.

51. Perlis, R. H., Nierenberg, A. A., Alpert, J. E., Pava, J. et al. Effects of adding cognitive therapy to fluoxetine dose increase on risk of relapse and residual depressive symptoms in continuation treatment of major depressive disorder J Clin Psychopharmacol 2002;22(5):474-480.

52. Schulberg, H. C., Block, M. R., Madonia, M. J., Scott, C. P. et al. Treating major depression in primary care practice. Eight-month clinical outcomes Arch Gen Psychiatry 1996;53(10):913-919.

53. Segal, Z. V., Bieling, P., Young, T., MacQueen, G. et al. Antidepressant monotherapy vs sequential pharmacotherapy and mindfulness-based cognitive therapy, or placebo, for relapse prophylaxis in recurrent depression Arch Gen Psychiatry 2010;67(12):1256-1264.

54. Teasdale, J. D., Segal, Z. V., Williams, J. M., Ridgeway, V. A. et al. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy J Consult Clin Psychol 2000;68(4):615-623.

55. Raskin, A., Schulterbrandt, J., Reatig, N., and McKeon, J. J.Replication of factors of psychopathology in interview, ward behavior and self-report ratings of hospitalized depressives J Nerv Ment Dis 1969;148(1):87-98.

56. Higgins, J. P., Thompson, S. G., Deeks, J. J., and Altman, D. G.Measuring inconsistency in meta-analyses BMJ 6-9-2003;327(7414):557-560.

57. Egger, M., Davey, Smith G., Schneider, M., and Minder, C.Bias in meta-analysis detected by a simple, graphical test BMJ 13-9-1997;315(7109):629-634.

58. Rothstein, H. R., Sutton, H. J., and Borenstein, M.Publication Bias in Meta-Analysis 2005;(1):2-7.

59. Andrews, G., Poulton, R., and Skoog, I.Lifetime risk of depression: restricted to a minority or waiting for most? Br J Psychiatry 2005;187:495-496.

60. ten Doesschate, M. C., Bockting, C. L., and Schene, A. H.Adherence to continuation and maintenance antidepressant use in recurrent depression J Affect Disord 2009;115(1-2):167-170.

61. Kelders, S. M., Pots, W. T., Oskam, M. J., Bohlmeijer, E. T. et al. Development of a web-based intervention for the indicated prevention of depression BMC Med Inform Decis Mak 2013;13:26-

62. Bosmans, J. E., Schreuders, B., van Marwijk, H. W., Smit, J. H. et al. Cost-effectiveness of problem-solving treatment in comparison with usual care for primary care patients with mental health problems: a randomized trial BMC Fam Pract 2012;13:98-

63. Biesheuvel-Leliefeld, K. E., Kersten, S. M., van der Horst, H. E., van, Schaik A. et al. Cost-effectiveness of nurse-led self-help for recurrent depression in the primary care setting: design of a pragmatic randomised controlled trial BMC Psychiatry 2012;12:59-

64. Rodgers, M., Asaria, M., Walker, S., McMillan, D. et al. The clinical effectiveness and cost-effectiveness of low-intensity psychological interventions for the secondary prevention of relapse after depression: a systematic review Health Technol Assess 2012;16(28):1-130.

R1R2R3R4R5R6R7R8R9


Chapter 3

66

Appendix S1. Search strategy

The literature search was conducted in May 2014. Searches were performed in the following databases and article indexes (number of atricles retrieved);MEDLINE (2,384)PSYCINFO (855)CINAHL (206)EMABASE (1,330)COCHRANE databases (922)

Controlled vocabulary was utilised where appropriate terms were available, supplemented with keyword searches to ensure accurate and exhaustive results. The studies had to be published in English. Additional delimiters were adults and randomised controlled trials. To illustrate, the following MEDLINE search is indicative of searches performed in other databases.

# 1 depression [MeSH] OR depressive disorder [MeSH]/exp# 2 major depression OR major depressive disorder# 3 prevention [MeSH]/exp# 4 relapse [MeSH] OR recurrence [MeSH] /exp# 5 psychotherapy [MeSH]/exp# 6 mindfullness based therapy [MeSH]/exp# 7 interpersonal therapy [MeSH]/exp# 8 cognitive therapy [MeSH] OR cognitive behavio(u)ral therapy [MeSH]/exp# 9 problem-solving [MeSH] OR problem-solving therapy [MeSH]/exp# 10 psychodynamic therapy [MeSH]/exp# 11 psychoanalytic therapy [MeSH]/exp# 12 #1 OR #2 # 13 #12 AND #3 # 14 #12 AND #4 # 15 #13 OR #14 # 16 #15 AND #5# 17 #15 AND #6# 18 #15 AND #7# 19 #15 AND #8# 20 #15 AND #9# 21 #15 AND #10# 22 #15 AND #11# 23 #16 OR #17 OR #18 OR # 19 OR #20 OR #21 OR #22 AND Limits; Randomised controlled trial, adults



67

3

App

endi

x S2

. PR

ISM

A C

heck

list o

f ite

ms

to in

clud

e w

hen

repo

rtin

g a

syst

emat

ic r

evie

w o

r m

eta-

anal

ysis

Item

Item

no

Con

tent

Page

-nu

mbe

r

Title

1

Iden

tify

the

repo

rt a

s a

syst

emat

ic r

evie

w, m

eta-

anal

ysis

, or

both

.45

Stru

ctur

ed

sum

mar

y 2

Prov

ide

a st

ruct

ured

sum

mar

y in

clud

ing,

as

appl

icab

le: b

ackg

roun

d; o

bjec

tives

; dat

a so

urce

s; s

tudy

elig

ibili

ty c

rite

ria,

pa

rtic

ipan

ts, a

nd in

terv

entio

ns; s

tudy

app

rais

al a

nd s

ynth

esis

met

hods

; res

ults

; lim

itatio

ns; c

oncl

usio

ns a

nd im

plic

atio

ns

of k

ey fi

ndin

gs; s

yste

mat

ic r

evie

w r

egis

trat

ion

num

ber.

46

Rat

iona

le

3D

escr

ibe

the

ratio

nale

for

the

revi

ew in

the

cont

ext o

f wha

t is

alre

ady

know

n.

47,4

8

Obj

ectiv

es

4Pr

ovid

e an

exp

licit

stat

emen

t of q

uest

ions

bei

ng a

ddre

ssed

with

ref

eren

ce to

par

ticip

ants

, int

erve

ntio

ns, c

ompa

riso

ns,

outc

omes

, and

stu

dy d

esig

n (P

ICO

S).

48

Prot

ocol

and

re

gist

ratio

n 5

Indi

cate

if a

rev

iew

pro

toco

l exi

sts,

if a

nd w

here

it c

an b

e ac

cess

ed (e

.g.,

Web

add

ress

), an

d, if

ava

ilabl

e, p

rovi

de

regi

stra

tion

info

rmat

ion

incl

udin

g re

gist

ratio

n nu

mbe

r. n/

a

Elig

ibili

ty c

rite

ria

6Sp

ecify

stu

dy c

hara

cter

istic

s (e

.g.,

PIC

OS,

leng

th o

f fol

low

-up)

and

rep

ort c

hara

cter

istic

s (e

.g.,

year

s co

nsid

ered

, la

ngua

ge, p

ublic

atio

n st

atus

) use

d as

cri

teri

a fo

r el

igib

ility

, giv

ing

ratio

nale

. 49

,50

Info

rmat

ion

sour

ces

7D

escr

ibe

all i

nfor

mat

ion

sour

ces

(e.g

., da

taba

ses

with

dat

es o

f cov

erag

e, c

onta

ct w

ith s

tudy

aut

hors

to id

entif

y ad

ditio

nal

stud

ies)

in th

e se

arch

and

dat

e la

st s

earc

hed.

50

Sear

ch

8Pr

esen

t ful

l ele

ctro

nic

sear

ch s

trat

egy

for

at le

ast o

ne d

atab

ase,

incl

udin

g an

y lim

its u

sed,

suc

h th

at it

cou

ld b

e re

peat

ed.

App

S1;

66

Stud

y se

lect

ion

9St

ate

the

proc

ess

for

sele

ctin

g st

udie

s (i.

e., s

cree

ning

, elig

ibili

ty, i

nclu

ded

in s

yste

mat

ic r

evie

w, a

nd, i

f app

licab

le,

incl

uded

in th

e m

eta-

anal

ysis

). 51

Dat

a co

llect

ion

proc

ess

10D

escr

ibe

met

hod

of d

ata

extr

actio

n fr

om r

epor

ts (e

.g.,

pilo

ted

form

s, in

depe

nden

tly, i

n du

plic

ate)

and

any

pro

cess

es fo

r ob

tain

ing

and

confi

rmin

g da

ta fr

om in

vest

igat

ors.

51

Dat

a ite

ms

11Li

st a

nd d

efine

all

vari

able

s fo

r w

hich

dat

a w

ere

soug

ht (e

.g.,

PIC

OS,

fund

ing

sour

ces)

and

any

ass

umpt

ions

and

si

mpl

ifica

tions

mad

e.

51

Ris

k of

bia

s in

in

divi

dual

stu

dies

12

Des

crib

e m

etho

ds u

sed

for

asse

ssin

g ri

sk o

f bia

s of

indi

vidu

al s

tudi

es (i

nclu

ding

spe

cific

atio

n of

whe

ther

this

was

don

e at

th

e st

udy

or o

utco

me

leve

l), a

nd h

ow in

form

atio

n is

to b

e us

ed in

any

dat

a sy

nthe

sis.

50

,51

Sum

mar

y m

easu

res

13St

ate

the

prin

cipa

l sum

mar

y m

easu

res

(e.g

., ri

sk r

atio

, diff

eren

ce in

mea

ns).

50

Synt

hesi

s of

re

sults

14

Des

crib

e th

e m

etho

ds o

f han

dlin

g da

ta a

nd c

ombi

ning

res

ults

of s

tudi

es, i

f don

e, in

clud

ing

mea

sure

s of

con

sist

ency

(e.g

., I2 )

for

each

met

a-an

alys

is.

51,5

2

Ris

k of

bia

s ac

ross

stu

dies

15

Spec

ify a

ny a

sses

smen

t of r

isk

of b

ias

that

may

affe

ct th

e cu

mul

ativ

e ev

iden

ce (e

.g.,

publ

icat

ion

bias

, sel

ectiv

e re

port

ing

with

in s

tudi

es).

50,5

1

Add

ition

al

anal

yses

16

Des

crib

e m

etho

ds o

f add

ition

al a

naly

ses

(e.g

., se

nsiti

vity

or

subg

roup

ana

lyse

s, m

eta-

regr

essi

on),

if do

ne, i

ndic

atin

g w

hich

wer

e pr

e-sp

ecifi

ed.

52

R1R2R3R4R5R6R7R8R9


Chapter 3

68

Item

Item

no

Con

tent

Page

-nu

mbe

r

Stud

y se

lect

ion

17G

ive

num

bers

of s

tudi

es s

cree

ned,

ass

esse

d fo

r el

igib

ility

, and

incl

uded

in th

e re

view

, with

rea

sons

for

excl

usio

ns a

t eac

h st

age,

idea

lly w

ith a

flow

dia

gram

. Fi

gure

1;

52,5

3

Stud

y ch

arac

teri

stic

s 18

For

each

stu

dy, p

rese

nt c

hara

cter

istic

s fo

r w

hich

dat

a w

ere

extr

acte

d (e

.g.,

stud

y si

ze, P

ICO

S, fo

llow

-up

peri

od) a

nd

prov

ide

the

cita

tions

. Ta

ble

2;

54,5

5

Ris

k of

bia

s w

ithin

stu

dies

19

Pres

ent d

ata

on r

isk

of b

ias

of e

ach

stud

y an

d, if

ava

ilabl

e, a

ny o

utco

me

leve

l ass

essm

ent (

see

item

12)

. 57

Res

ults

of

indi

vidu

al s

tudi

es

20Fo

r al

l out

com

es c

onsi

dere

d (b

enefi

ts o

r ha

rms)

, pre

sent

, for

eac

h st

udy:

(a) s

impl

e su

mm

ary

data

for

each

inte

rven

tion

grou

p (b

) effe

ct e

stim

ates

and

con

fiden

ce in

terv

als,

idea

lly w

ith a

fore

st p

lot.

56-5

8

Synt

hesi

s of

re

sults

21

Pres

ent r

esul

ts o

f eac

h m

eta-

anal

ysis

don

e, in

clud

ing

confi

denc

e in

terv

als

and

mea

sure

s of

con

sist

ency

. 56

-58

Ris

k of

bia

s ac

ross

stu

dies

22

Pres

ent r

esul

ts o

f any

ass

essm

ent o

f ris

k of

bia

s ac

ross

stu

dies

(see

Item

15)

. A

pp S

3; 6

9

Add

ition

al

anal

ysis

23

Giv

e re

sults

of a

dditi

onal

ana

lyse

s, if

don

e (e

.g.,

sens

itivi

ty o

r su

bgro

up a

naly

ses,

met

a-re

gres

sion

[se

e Ite

m 1

6]).

59

Sum

mar

y of

ev

iden

ce

24Su

mm

ariz

e th

e m

ain

findi

ngs

incl

udin

g th

e st

reng

th o

f evi

denc

e fo

r ea

ch m

ain

outc

ome;

con

side

r th

eir

rele

vanc

e to

key

gr

oups

(e.g

., he

alth

care

pro

vide

rs, u

sers

, and

pol

icy

mak

ers)

. 60

Lim

itatio

ns

25D

iscu

ss li

mita

tions

at s

tudy

and

out

com

e le

vel (

e.g.

, ris

k of

bia

s), a

nd a

t rev

iew

-lev

el (e

.g.,

inco

mpl

ete

retr

ieva

l of

iden

tified

res

earc

h, r

epor

ting

bias

). 60

Con

clus

ions

26

Prov

ide

a ge

nera

l int

erpr

etat

ion

of th

e re

sults

in th

e co

ntex

t of o

ther

evi

denc

e, a

nd im

plic

atio

ns fo

r fu

ture

res

earc

h.

61

Fund

ing

27D

escr

ibe

sour

ces

of fu

ndin

g fo

r th

e sy

stem

atic

rev

iew

and

oth

er s

uppo

rt (e

.g.,

supp

ly o

f dat

a); r

ole

of fu

nder

s fo

r th

e sy

stem

atic

rev

iew

.n/

a



69

3

Appendix S3. Risk of bias according to 6 Cochrane criteria

Author Year Adequate sequence generation?

Adequate allocation concealment?

Adequate blinding?a

Incomplete outcomedata addressed?

Free of selective reporting?

Free of other bias?

Baker 1985 Unclear Unclear No No Yes NoBlackburn 1986 Unclear Unclear No No Yes NoBlackburn 1997 Unclear Unclear No Yes Yes YesBockting 2009 Yes Yes Yes Yes Yes YesBondolfi 2010 Yes Yes Yes Yes Yes YesConradi 2007 Yes Yes No Yes Yes YesFava 1998 Unclear Unclear No No Yes UnclearFava 2002 Unclear Unclear No Yes Yes YesFava 2004 Unclear Unclear No Yes Yes YesFrank 1990 Unclear Unclear No No No NoGodfrin 2010 Yes Yes No Yes Yes YesHollon 2005 Unclear Unclear No Yes Yes No Hollandare 2011 Yes Yes No Yes Yes YesJarrett 2000 Unclear Unclear No Yes Yes No Jarrett 2001 Yes Yes No Yes Yes YesJarrettKlein

20132004

YesUnclear

YesUnclear

YesNo

YesYes

YesYes

YesNo

Klerman 1974 Unclear Unclear No No Yes NoKuyken 2008 Yes Yes No Yes Yes YesMa 2004 No Yes No Yes Yes YesPaykel 2005 Yes Yes Yes Yes Yes YesPerlis 2002 Unclear Unclear No Yes Yes No Schulberg 1996 Unclear Unclear No Yes Yes YesSegal 2010 Yes Yes No Yes Yes YesTeasdale 2000 No Yes No Yes Yes Yes

a Even when assessors/interviewers were blind to allocation of the intervention, it is very likely that participants commented spontaneously on their intervention during an interview. In trials including psychological interventions it is therefore very likely that there is a high risk of bias regarding ‘blinding’. When studies include any thing of the following;‘participants were instructed not to reveal this information to the interviewers’, we considered blinding adequate.

R1R2R3R4R5R6R7R8R9


Chapter 3

70

App

endi

x S4

. GR

AD

E ev

iden

ce p

rofil

e; p

sych

olog

ical

inte

rven

tions

ver

sus

com

para

tors

for

patie

nts

with

rec

urre

nt d

epre

ssio

n

Qua

lity

asse

ssm

ent

Sum

mar

y of

find

ings

Num

ber

of p

atie

nts

No

of c

ontr

asts

and

de

sign

St

udy

limita

tions

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nPu

blic

atio

n bi

asIn

terv

entio

nTA

UA

DM

RR

R

DQ

ualit

y

Rel

apse

afte

r FU

a (al

l vs T

AU

)17

RC

T’s

Seri

ous

limita

tions

Seri

ous

inco

nsis

tenc

y b

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

Ver

y se

riou

s lim

itatio

n c1

646

670

n/a

0.64

(0.5

3-0.

76)

-0.1

90(-

0.25

5- -

0.12

5)Lo

w

Rel

apse

afte

r FU

a (al

l vs

AD

M)

13 R

CT’

s Se

riou

slim

itatio

nsN

o se

riou

s in

cons

iste

ncy

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

No

seri

ous

limiti

atio

n c2

46

1n/

a45

30.

83(0

.70-

0.97

)-0

.075

(-0.

149-

-0.0

01)

Low

Rel

apse

afte

r FU

a 9

RC

T’s

(CT

vs T

AU

)Se

riou

slim

itatio

nsSe

riou

s in

cons

iste

ncy

bSe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n 30

432

2n/

a0.

68

(0.5

4-0.

87)

-0.1

96(-

0.28

- -0

.11)

Low

Rel

apse

afte

r FU

a 5

RC

T’s

(MB

T vs

TA

U)

Seri

ous

limita

tion

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n20

020

8n/

a0.

66

(0.5

3-0.

82)

-0.2

05(-

0.32

--0.

09)

Low

Rel

apse

afte

r FU

a 3

RC

T’s

(IPT

vs T

AU

)V

ery

seri

ous

limita

tions

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n14

214

0n/

a0.

41

(0.2

7-0.

63)

-0.1

60(-

0.37

- -0

.04)

Low

Rel

apse

afte

r FU

a 8

RC

T’s

(CT

vs A

DM

)Se

riou

slim

itatio

nsN

o se

riou

s in

cons

iste

ncy

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

Seri

ous

limita

tion

232

n/a

219

0.79

(0.6

1-1.

02)

-0.1

6 (-

0.30

- -

0.01

6)Lo

w

Rel

apse

afte

r FU

a 2

RC

T’s

(MB

T vs

AD

M)

No

limita

tions

No

seri

ous

inco

nsis

tenc

yN

o in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n87

n/a

900.

80(0

.60-

1.08

)-0

.11

(-0.

25-0

.04)

Mod

erat

e

Rel

apse

afte

r FU

a 3

RC

T’s

(IPT

vs A

DM

)V

ery

seri

ous

limita

tions

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n14

2n/

a14

40.

83(0

.50-

1.38

)-0

.002

(-0.

068-

0.07

3)Lo

w

a Fo

llow

-up

b I2

(het

erog

enei

ty) >

50%

c1

Sig

nific

ant E

gger

’s te

st in

terc

ept ;

-1.

65, S

E =

0.4

1, P

< 0

.001

c2 N

on-s

igni

fican

t Egg

er te

st in

terc

ept ;

-0.

47, S

E=0.

41, P

=0.

27



71

3

App

endi

x S4

. GR

AD

E ev

iden

ce p

rofil

e; p

sych

olog

ical

inte

rven

tions

ver

sus

com

para

tors

for

patie

nts

with

rec

urre

nt d

epre

ssio

n

Qua

lity

asse

ssm

ent

Sum

mar

y of

find

ings

Num

ber

of p

atie

nts

No

of c

ontr

asts

and

de

sign

St

udy

limita

tions

Inco

nsis

tenc

yIn

dire

ctne

ssIm

prec

isio

nPu

blic

atio

n bi

asIn

terv

entio

nTA

UA

DM

RR

R

DQ

ualit

y

Rel

apse

afte

r FU

a (al

l vs T

AU

)17

RC

T’s

Seri

ous

limita

tions

Seri

ous

inco

nsis

tenc

y b

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

Ver

y se

riou

s lim

itatio

n c1

646

670

n/a

0.64

(0.5

3-0.

76)

-0.1

90(-

0.25

5- -

0.12

5)Lo

w

Rel

apse

afte

r FU

a (al

l vs

AD

M)

13 R

CT’

s Se

riou

slim

itatio

nsN

o se

riou

s in

cons

iste

ncy

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

No

seri

ous

limiti

atio

n c2

46

1n/

a45

30.

83(0

.70-

0.97

)-0

.075

(-0.

149-

-0.0

01)

Low

Rel

apse

afte

r FU

a 9

RC

T’s

(CT

vs T

AU

)Se

riou

slim

itatio

nsSe

riou

s in

cons

iste

ncy

bSe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n 30

432

2n/

a0.

68

(0.5

4-0.

87)

-0.1

96(-

0.28

- -0

.11)

Low

Rel

apse

afte

r FU

a 5

RC

T’s

(MB

T vs

TA

U)

Seri

ous

limita

tion

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n20

020

8n/

a0.

66

(0.5

3-0.

82)

-0.2

05(-

0.32

--0.

09)

Low

Rel

apse

afte

r FU

a 3

RC

T’s

(IPT

vs T

AU

)V

ery

seri

ous

limita

tions

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n14

214

0n/

a0.

41

(0.2

7-0.

63)

-0.1

60(-

0.37

- -0

.04)

Low

Rel

apse

afte

r FU

a 8

RC

T’s

(CT

vs A

DM

)Se

riou

slim

itatio

nsN

o se

riou

s in

cons

iste

ncy

Seri

ous

indi

rect

ness

Seri

ous

impr

ecis

ion

Seri

ous

limita

tion

232

n/a

219

0.79

(0.6

1-1.

02)

-0.1

6 (-

0.30

- -

0.01

6)Lo

w

Rel

apse

afte

r FU

a 2

RC

T’s

(MB

T vs

AD

M)

No

limita

tions

No

seri

ous

inco

nsis

tenc

yN

o in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n87

n/a

900.

80(0

.60-

1.08

)-0

.11

(-0.

25-0

.04)

Mod

erat

e

Rel

apse

afte

r FU

a 3

RC

T’s

(IPT

vs A

DM

)V

ery

seri

ous

limita

tions

No

seri

ous

inco

nsis

tenc

ySe

riou

s in

dire

ctne

ssSe

riou

s im

prec

isio

nSe

riou

s lim

itatio

n14

2n/

a14

40.

83(0

.50-

1.38

)-0

.002

(-0.

068-

0.07

3)Lo

w

a Fo

llow

-up

b I2

(het

erog

enei

ty) >

50%

c1

Sig

nific

ant E

gger

’s te

st in

terc

ept ;

-1.

65, S

E =

0.4

1, P

< 0

.001

c2 N

on-s

igni

fican

t Egg

er te

st in

terc

ept ;

-0.

47, S

E=0.

41, P

=0.

27

App

endi

x S5

. Exc

lude

d st

udie

s an

d re

ason

of e

xclu

sion

(44

stud

ies)

Prim

ary

auth

orYe

arTi

tle

Rea

son

for

excl

usio

n

1A

alde

ren

van

2011

The

effic

acy

of m

indf

ulne

ss-b

ased

cog

nitiv

e th

erap

y in

rec

urre

nt d

epre

ssed

pat

ient

s w

ith a

nd w

ithou

t a

curr

ent d

epre

ssiv

e ep

isod

e: a

ran

dom

ized

con

trol

led

tria

lno

rel

apse

rat

es

2B

arnh

ofer

2009

Min

dful

ness

-bas

ed c

ogni

tive

ther

apy

as a

trea

tmen

t for

chr

onic

dep

ress

ion:

A p

relim

inar

y st

udy.

no r

elap

se r

ates

3B

ockt

ing

2005

Prev

entin

g re

laps

e/re

curr

ence

in r

ecur

rent

dep

ress

ion

with

cog

nitiv

e th

erap

y: a

ran

dom

ized

con

trol

led

tria

l.ot

her

4B

osm

ans

2012

Cos

t-ef

fect

iven

ess

of p

robl

em-s

olvi

ng tr

eatm

ent i

n co

mpa

riso

n w

ith u

sual

car

e fo

r pr

imar

y ca

re p

atie

nts

with

men

tal h

ealth

pro

blem

s: a

ran

dom

ized

tria

lno

rel

apse

rat

es

5C

arte

r20

11Pa

tient

pre

dict

ors

of r

espo

nse

to c

ogni

tive

beha

viou

r th

erap

y an

d in

terp

erso

nal p

sych

othe

rapy

in a

ra

ndom

ised

clin

ical

tria

l for

dep

ress

ion

othe

r

6C

arva

lho

2010

Effic

acy

of c

ogni

tive-

beha

vior

al th

erap

y fo

r th

e tr

eatm

ent o

f rec

urre

nt d

epre

ssio

n in

adu

ltsfo

llow

-up

afte

r ac

ute

phas

e7

Dek

ker

2013

Wha

t is

the

best

seq

uent

ial t

reat

men

t str

ateg

y in

the

trea

tmen

t of d

epre

ssio

n? A

ddin

g ph

arm

acot

hera

py to

ps

ycho

ther

apy

or v

ice

vers

a?ot

her

8Ev

ans

1992

Diff

eren

tial r

elap

se fo

llow

ing

cogn

itive

ther

apy

and

phar

mac

othe

rapy

for

depr

essi

on.

follo

w-u

p af

ter

acut

e ph

ase

9Fa

va19

94C

ogni

tive

beha

vior

al tr

eatm

ent o

f res

idua

l sym

ptom

s in

pri

mar

y m

ajor

dep

ress

ive

diso

rder

othe

r10

Fava

1998

Prev

entio

n of

rec

urre

nt d

epre

ssio

n w

ith c

ogni

tive

beha

vior

al th

erap

y: p

relim

inar

y fin

ding

sot

her

11Fr

ank

2007

Ran

dom

ized

tria

l of w

eekl

y, tw

ice-

mon

thly

, and

mon

thly

inte

rper

sona

l psy

chot

hera

py a

s m

aint

enan

ce

trea

tmen

t for

wom

en w

ith r

ecur

rent

dep

ress

ion

no R

CT

12G

esch

win

d20

11Ef

ficac

y of

min

dful

ness

-bas

ed c

ogni

tive

ther

apy

in r

elat

ion

to p

rior

his

tory

of d

epre

ssio

n: r

ando

mis

ed

cont

rolle

d tr

ial

no R

CT

13G

olla

n20

06Pr

edic

tors

of d

epre

ssiv

e re

laps

e du

ring

a tw

o ye

ar p

rosp

ectiv

e fo

llow

-up

afte

r co

gniti

ve a

nd b

ehav

iora

l th

erap

ies

follo

w-u

p af

ter

acut

e ph

ase

14G

ortn

er19

98C

ogni

tive-

beha

vior

al tr

eatm

ent f

or d

epre

ssio

n: r

elap

se p

reve

ntio

nfo

llow

-up

afte

r ac

ute

15H

owel

l20

08Pr

even

ting

rela

pse

of d

epre

ssio

n in

pri

mar

y ca

re: a

pilo

t stu

dy o

f the

“Ke

epin

g th

e bl

ues

away

” pr

ogra

m.

othe

r16

Jarr

ett

1998

Is th

ere

a ro

le fo

r co

ntin

uatio

n ph

ase

cogn

itive

ther

apy

for

depr

esse

d ou

tpat

ient

s?

no R

CT

17K

arp

2004

Rel

atio

nshi

p of

var

iabi

lity

in r

esid

ual s

ympt

oms

with

rec

urre

nce

of m

ajor

dep

ress

ive

diso

rder

dur

ing

mai

nten

ance

trea

tmen

t.no

rel

apse

rat

es

18K

aton

2001

A r

ando

miz

ed tr

ial o

f rel

apse

pre

vent

ion

of d

epre

ssio

n in

pri

mar

y ca

reno

psy

chot

hera

py

19K

ings

ton

2007

Min

dful

ness

-bas

ed c

ogni

tive

ther

apy

for

resi

dual

dep

ress

ive

sym

ptom

sno

rel

apse

rat

es20

Kocs

is20

03C

ontin

uatio

n tr

eatm

ent o

f chr

onic

dep

ress

ion:

a c

ompa

riso

n of

nef

azod

one,

cog

nitiv

e be

havi

oral

ana

lysi

s sy

stem

of p

sych

othe

rapy

, and

thei

r co

mbi

natio

nno

RC

T

R1R2R3R4R5R6R7R8R9


Chapter 3

72

Prim

ary

auth

orYe

arTi

tle

Rea

son

for

excl

usio

n

21Ku

hner

1996

Cog

nitiv

e-be

havi

oral

gro

up in

terv

entio

n as

a m

eans

of t

ertia

ry p

reve

ntio

n in

dep

ress

ed p

atie

nts:

A

ccep

tanc

e an

d sh

ort-

term

effi

cacy

no R

CT

22Ku

hner

2005

An

eval

uatio

n of

the

‘Cop

ing

with

Dep

ress

ion

Cou

rse’

for

rela

pse

prev

entio

n w

ith u

nipo

lar

depr

esse

d pa

tient

sno

RC

T

23Ku

pfer

1992

Five

-yea

r ou

tcom

e fo

r m

aint

enan

ce th

erap

ies

in r

ecur

rent

dep

ress

ion.

othe

r24

Lem

men

s20

11Ef

fect

iven

ess,

rel

apse

pre

vent

ion

and

mec

hani

sms

of c

hang

e of

cog

nitiv

e th

erap

y vs

. int

erpe

rson

al th

erap

y fo

r de

pres

sion

: Stu

dy p

roto

col f

or a

ran

dom

ised

con

trol

led

tria

l.no

RC

T

25Lu

dman

2000

The

desi

gn, i

mpl

emen

tatio

n, a

nd a

ccep

tanc

e of

a p

rim

ary

care

-bas

ed in

terv

entio

n to

pre

vent

dep

ress

ion

rela

pse.

othe

r

26M

acca

relli

2002

Mai

nten

ance

inte

rper

sona

l psy

chot

hera

py (I

PT-M

) tre

atm

ent s

peci

ficity

: The

impa

ct o

n le

ngth

of r

emis

sion

in

wom

en w

ith r

ecur

rent

dep

ress

ion

no r

elap

se r

ates

27M

athe

w20

10Th

e lo

ng-t

erm

effe

cts

of m

indf

ulne

ss-b

ased

cog

nitiv

e th

erap

y as

a r

elap

se p

reve

ntio

n tr

eatm

ent f

or m

ajor

de

pres

sive

dis

orde

r.no

RC

T

28M

ynor

s-W

allis

1995

Ran

dom

ised

con

trol

led

tria

l com

pari

ng p

robl

em s

olvi

ng tr

eatm

ent w

ith a

mitr

ipty

line

and

plac

ebo

for

maj

or d

epre

ssio

n in

pri

mar

y ca

reno

rel

apse

rat

es

29M

ynor

s-W

allis

2000

Ran

dom

ised

con

trol

led

tria

l of p

robl

em s

olvi

ng tr

eatm

ent,

antid

epre

ssan

t med

icat

ion,

and

com

bine

d tr

eatm

ent f

or m

ajor

dep

ress

ion

in p

rim

ary

care

no r

elap

se r

ates

30Pa

va19

94In

tegr

atin

g co

gniti

ve th

erap

y an

d ph

arm

acot

hera

py in

the

trea

tmen

t and

pro

phyl

axis

of d

epre

ssio

n: a

no

vel a

ppro

ach

no R

CT

31Pa

ykel

1999

Prev

entio

n of

rel

apse

in r

esid

ual d

epre

ssio

n by

cog

nitiv

e th

erap

y: a

con

trol

led

tria

lot

her

32Q

uiri

ng20

02D

oes

earl

y in

terv

entio

n in

crea

se la

tenc

y to

rel

apse

in m

ajor

dep

ress

ive

diso

rder

? R

e-ev

alua

tion

with

co

gniti

ve b

ehav

ior

ther

apy

follo

w-u

p af

ter

acut

e ph

ase

33Sc

ott

2000

Effe

cts

of c

ogni

tive

ther

apy

on p

sych

olog

ical

sym

ptom

s an

d so

cial

func

tioni

ng in

res

idua

l dep

ress

ion

no r

elap

se r

ates

34Sc

ott

2003

Use

of c

ogni

tive

ther

apy

for

rela

pse

prev

entio

n in

chr

onic

dep

ress

ion.

Cos

t-ef

fect

iven

ess

stud

yno

acu

te p

hase

res

pond

ers

35Sh

ahar

2010

Mec

hani

sms

of c

hang

e in

min

dful

ness

-bas

ed c

ogni

tive

ther

apy

for

depr

essi

on: P

relim

inar

y ev

iden

ce fr

om

a ra

ndom

ized

con

trol

led

tria

l.no

rel

apse

rat

es

36Sh

apir

o19

95Ef

fect

s of

trea

tmen

t dur

atio

n an

d se

veri

ty o

f dep

ress

ion

on th

e m

aint

enan

ce o

f gai

ns a

fter

cogn

itive

-be

havi

oral

and

psy

chod

ynam

ic-i

nter

pers

onal

psy

chot

hera

py.

follo

w-u

p af

ter

acut

e ph

ase

37Sh

awye

r20

12Th

e D

AR

E st

udy

of r

elap

se p

reve

ntio

n in

dep

ress

ion:

des

ign

for

a ph

ase

1/2

tran

slat

iona

l ran

dom

ised

co

ntro

lled

tria

l inv

olvi

ng m

indf

ulne

ss-b

ased

cog

nitiv

e th

erap

y an

d su

ppor

ted

self

mon

itori

ng.

no R

CT

38Si

mon

2002

Cos

t-ef

fect

iven

ess

of a

pro

gram

to p

reve

nt d

epre

ssio

n re

laps

e in

pri

mar

y ca

reno

psy

chot

hera

py



73

3

Prim

ary

auth

orYe

arTi

tle

Rea

son

for

excl

usio

n

39Si

mon

s19

86C

ogni

tive

ther

apy

and

phar

mac

othe

rapy

for

depr

essi

on. S

usta

ined

impr

ovem

ent o

ver

one

year

follo

w-u

p af

ter

acut

e ph

ase

40Sm

it20

06Sh

ort-

term

effe

cts

of e

nhan

ced

trea

tmen

t for

dep

ress

ion

in p

rim

ary

care

: res

ults

from

a r

ando

miz

ed

cont

rolle

d tr

ial

no a

cute

pha

se r

espo

nder

s

41St

ant

2009

Cos

t-ef

fect

iven

ess

of a

psy

choe

duca

tiona

l rel

apse

pre

vent

ion

prog

ram

for

depr

essi

on in

pri

mar

y ca

re.

no r

elap

se r

ates

42 43St

angi

er

Thas

e

2013

1992

Mai

nten

ance

cog

nitiv

e-be

havi

oral

ther

apy

and

man

ualiz

ed p

sych

oedu

catio

n in

the

trea

tmen

t of

recu

rren

t dep

ress

ion:

a m

ultic

ente

r pr

ospe

ctiv

e ra

ndom

ized

con

trol

led

tria

l.

Rel

apse

afte

r co

gniti

ve b

ehav

ior

ther

apy

of d

epre

ssio

n: p

oten

tial i

mpl

icat

ions

for

long

er c

ours

es o

f tr

eatm

ent

othe

rfo

llow

-up

afte

r ac

ute

phas

e

44V

itten

gl20

09C

ontin

uatio

n-ph

ase

cogn

itive

ther

apy’

s ef

fect

s on

rem

issi

on a

nd r

ecov

ery

from

dep

ress

ion

no R

CT

4Cost-effectiveness of preventing depressive recurrences

by psychological interventions;

a population health economic modelling study

Karolien E.M. Biesheuvel-LeliefeldJoran Lokkerbol

Filip Smit

Submitted in: Journal of Psychiatric and Mental Health Nursing

R1R2R3R4R5R6R7R8R9


Chapter 4

76

ABSTRACT

Introduction Largely owing to its recurrent nature, major depressive disorder (MDD) is associated with high economic costs.

Aim Our aim was to investigate whether offering preventive psychotherapies would improve the cost-effectiveness of the Dutch health care system for recurrent MDD.

Method A health economic model was used to assess return on investments (ROI). We compared a base-case scenario (enhanced TAU) with four scenarios in terms of cost-effectiveness: enhanced TAU plus A) cognitive therapy, CT; B) mindfulness-based CT, MCT; C) interpersonal therapy, IPT and D) a hypothetical ‘supported self-help’ based on preventive CT.

Results The ROI of enhanced TAU is €1.30 (£1.10) while the ROI of adding IPT, CT or MCT is €1.31 (£1.10), €1.43 (£1.21) and €1.45 (£1.22) respectively. In order to reach the most competitive ROI (€1.45), the supported self-help needs to reach a relative risk reduction of 0.71.

Discussion Adding CT, MCT or a supported self-help might make the healthcare system for recurrent MDD more cost-effective compared to enhanced TAU only.

Implications for practice Due to an unsustainable expenditure for mental health, governments are forced to allocate resources to interventions that maximise cost-effectiveness. This study shows cost-effective alternatives for the continuation of medication in recurrently depressed patients.


Psychological interventions in recurrent depression; a health economic modelling study

77

4

INTRODUCTION

Major depressive disorder (MDD) affects 16% of the population on a lifetime basis1 and is associated with a high risk of relapse2. Of all people with MDD, at least 80% experience recurrent depressive disorder3 with five4 to nine5 depressive episodes over the course of their life and spend as much as 21% of their lifetime in a depressed condition6. In the specialised mental health care, the percentage of recurrence is even higher, up to 85% over 15 years7. MDD is therefore perhaps best described as a largely chronically recurrent disorder with much of its disease burden stemming from its recurrent nature. Partly due to its recurrent nature, MDD has substantial economic consequences8–11. Reducing the disease burden of depressive disorders by preventing new episodes at affordable costs is therefore of great public health significance. A recent meta-analysis by Biesheuvel et al12 showed that preventive psychological interventions were significantly better in reducing the risk of relapse or recurrence when compared to treatment-as-usual (TAU) (RR=0.64, 95%CI:0.53-0.76, z= 4.89, p<0.001, NNT=5) and even when compared to anti-depressant medication (ADM) (RR=0.83, 95%CI:0.70-0.97, z=2.40, p=0.017, NNT=13).

Though clinical evidence is promising, due to the mounting pressure on healthcare budgets, there is an additional need to also incorporate economic evidence to guide healthcare policies. To be able to assess the cost-effectiveness of (multiple) interventions, Lokkerbol et al 13 developed a health economic (Markov) model that synthesizes clinical, economic and epidemiological evidence. In the model, we used a base-case scenario (enhanced TAU) and compared this with enhanced TAU plus various psychological interventions to prevent recurrence in depression.

In our study we used this Markov model to answer the following research questions: 1. Do preventive psychological interventions for recurrences in depressive disorder

improve the cost-effectiveness over and above enhanced TAU? 2. What preventive psychological intervention makes the largest contribution in improving

the health care system’s cost-effectiveness?

Since the available interventions often rely on intensive use of therapist’s time and are therefore costly, we asked ourselves if there would be room for a (hypothetical) low-cost supported self-help intervention. Therefore, we will also use the Markov model to answer following research question:

3. How effective needs a hypothetical low-cost supported self-help intervention be to become competitive in terms of its cost-effectiveness relative to CT, MCT and IPT?

R1R2R3R4R5R6R7R8R9


Chapter 4

78

METHODS

Cost-effectiveness at health care system levelAn easy indicator to describe the ratio between effectiveness and intervention costs at population level is the return on investment (ROI). The ROI is computed as (DALYs averted * WtP) / (costs enhanced TAU + (per-patient preventive intervention costs * number of patients)). DALY stands for reduction of disability-adjusted life-years, i.e. the reduction of the number of years lost due to disability and due to premature death. In this study we use the lower WtP bound of €20,000 (£16,859) for one DALY averted. Both benefits and costs are expressed in € (£). A higher ROI for an added intervention is more favourable. This indicator helps to make quick and easy comparisons across competing interventions.

Enhanced TAUThe enhanced TAU scenario, which forms the basis for the comparisons, is an evidence-based healthcare system for depressive disorder in full agreement with the Dutch clinical guidelines for the treatment of depression, which is quite similar to the well-accepted NICE guidelines. The enhanced TAU scenario is not only “evidence-based” (guideline congruent), but also “preference-based” in the sense that interventions were endorsed by both patients and health care professionals (see Lokkerbol et al 13 for a full description).

Preventive psychological interventions According to the literature, various interventions in the continuation or maintenance phase are available for the prevention of recurrent depressive disorder (Biesheuvel-Leliefeld et al. 2014). In our study, the reviewed interventions were preventive cognitive therapy (CT), mindfulness-based cognitive therapy (MCT) and interpersonal therapy (IPT). CT is based on Beck’s theory that negative automatic thoughts, maladaptive information processing, and avoidance behaviour play a key role in the development and maintenance of depression 14. MCT is a protocol-led, group-based skills training program designed to teach remitted patients how to disengage from automatic, cognitive processing patterns linked to depressive relapse 15. IPT originates from interpersonal theory 16. It links stressful life events and insufficient social support to the development and maintenance of depressive symptoms 17.Two other known psychological interventions for depression, problem-solving therapy (PST) and psychodynamic therapy (PDT) were not included in the model because no trials could be found that exclusively focus on the prevention of relapse/recurrence in the continuation or maintenance phase.



79

4

A new (hypothetical) supported self-help intervention Since existing therapist-led preventive psychological interventions might add too much pressure on already tight budgets, there is a demand for effective interventions that can be offered at low costs. Therefore we also evaluated the cost-effectiveness of a less costly type of CT, which is a supported self-help intervention with minimal guidance (henceforth: ‘supported self-help’). This supported self-help was recently developed 18, but has not yet been evaluated in a cost-effectiveness trial. Since the effects of this supported self-help are not yet known, we simulated the minimum effectiveness needed for the self-help intervention to be able to successfully compete with the best alternative (i.e. the CT, MCT or ITP intervention).

Clinical evidenceWe used clinical effectiveness data from the review by Biesheuvel-Leliefeld et al 12 (Table 1). The outcome of interest was the relative risk (RR) with lower RR indicating a greater risk reduction for a depressive relapse.

Table 1. Risk ratio and risk difference of psychological interventions versus treatment-as-usual (TAU)a

according to the random-effects model (DerSimonian and Laird)b

Intervention KRR

(95%CI) Test RD

(95%CI) NNT I2

CT 9 0.68 (0.54;0.87)

z=-3.12 P=0.002

-0.196(-0.28; -0.11)

5 52%

MCT 5 0.66 (0.53;0.82)

z=-3.81 P<0.001

-0.205(-0.32; -0.09)

4 0%

IPT 3 0.41(0.27;0.63)

z=-4.10 P<0.001

-0.160(-0.37; -0.04)

6 0%

a TAU in the meta-analysis is defined ‘routine clinical management, assessments only, no treatment and waiting-list control with unrestricted access to TAU.’ b Abbreviations: CT, Cognitive (Behaviour) Therapy; CI, confidence interval; I2, heterogeneity; IPT, Interpersonal Therapy; K, number of studies; MCT, Mindfulness-based Cognitive Therapy; NNT, number-needed-to-treat; RD, risk difference; RR D-L, random-effects according to DerSimonian and Laird; TAU, treatment-as-usual

CostsCosts were estimated by mapping the total time of one intervention (hours) multiplied by the appropriate full economic costs of the healthcare professional according to the Dutch guidelines for health economic evaluations 19 (Table 2). Number of sessions and number of participants and time per session are based on commonly used formats. Costs of referral by a general practitioner were also included. Due to poor reporting, additional costs like costs of material, a possible orientation session or costs due to time spent by the patients at homework or travelling were not taken into account. Costs in euros (€) were converted to pounds sterling (£) using the purchasing power parities reported by the Organisation for

R1R2R3R4R5R6R7R8R9


Chapter 4

80

Economic Cooperation and Development, which convert currencies taking into account the differential buying power across countries. For the reference year 2013, €1 in The Netherlands was equal to a little over £0.84.

Table 2. Per patient costs of the interventionsa,b

format number of sessions

time per session(hours)

costs HCPper hour

(€)

number of participants

referral costs GP (€)

total costs per patient

(€)

CT group 8 2 171 8 28 370MCT group 8 2 171 8 28 370IPT individual 12 1 171 1 28 2.080self-help individual 3 1 50 1 28 178

aAbbreviations: CT, Cognitive (Behaviour) Therapy; HCP, healthcare professional; IPT, Interpersonal Therapy; MCT, Mindfulness-based Cognitive Therapy b Purchasing Power Parities (PPP) over 2013 were used for conversion from euro (€) to UK pound (£), found at the Organisation for Economic Cooperation and Development (OECD). http://stats.oecd.org/Index.aspx?DataSetCode=PPPGDP

Health economic evaluation using DepModDepMod is a health economic (Markov) model. It is developed and described in detail by Lokkerbol et al (2014). DepMod calculates the total healthcare costs and health gains by comparing a base-case scenario (enhanced TAU) with an alternative scenario (TAU plus an intervention). For the current study the alternative scenarios were defined as enhanced TAU plus CT, MCT, IPT or supported self-help.

DepMod assumes a population of 10 million people, aged 18-65 years. This population has an incidence of 158,000 new MDD cases per year with an episode duration of 6 months on average, and a prevalence of 588,600 acute cases annually. In this population, a rather conservative value of 45% of the currently depressed people have a history of previous episodes and are therefore seen as cases of recurrent depression. It should be noted that these epidemiological parameters are obtained from The Netherlands Mental Health Survey and Incidence Study 1 and 2 (NEMESIS 1 and 2), a population-based psychiatric epidemiologic cohort study 21,22. In this study, the time horizon is five years, standardised effect sizes are normally distributed, costs are gamma distributed and include only direct medical costs.

To allow for parameter uncertainty in costs and effects, the model randomly draws a value from the distributions assigned to the parameters and computes the outcome for that configuration of parameter values. This procedure is repeated 1,000 times over all parameters simultaneously. In each run, the outcomes (costs and health gains for each scenarios) are computed and stored in DepMod’s memory. Then, following the methods



81

4

of Briggs et al (2006) all 1,000 simulated outcomes are evaluated simultaneously, thus explicitly accounting for uncertainty in the input parameters. For more details on DepMod’s assumptions and their justifications we refer to Lokkerbol et al (2014). Sensitivity analysisOne of the parameters in our health economic model is the coverage rate (%) of the interventions. In DepMod coverage rates can be used to conduct “what-if” analyses to evaluate the impact of adding or removing an intervention.

Empirical estimates of treatment coverage show much variation due to the availability of mental health services, cultural barriers that limit mental health service use etc. In the ESEMeD study 24, about one in three (36.5%) individuals with a 12-month prevalence of any mood disorder reported using health services. Andrews et al. (2004) found a coverage rate for depression of 31.7%. However, Chisholm et al.(2004) employed a coverage rate of 50% in his modelling study and in the Netherlands 58.5% of the people (aged 18-64 years) diagnosed with MDD, received some kind of treatment in the last year 27.

Given the modest levels of care-seeking and poor recognition rates of symptoms that precursor relapse, coverage rates of interventions aimed at the prevention of relapse or recurrence of depression, are assumed lower than those for acute depression. For example, coverage of e-health interventions for prevention of relapse and recurrence is set at 10% 13. In our study, the basic coverage rate is arbitrarily and conservatively set at 20% of the patients with a history of depression. In order to test the robustness of these results, the analysis will be repeated with a coverage rate of 40%. The results, based on this rather high coverage rate, will help us to better gauge the impacts of the new preventive interventions.

RESULTS

Enhanced TAUWhen valuing each DALY averted at a WtP of €20,000 (£16,859), the healthcare system for the enhanced TAU scenario generates a value of €0.96 billion (£0.81 billion) (±50,000 DALY averted * €20,000) (95%CI: €0.91 – 1.01 billion) at an overall cost of €750 (£632) million. This results in a ROI of €1.30 (95%CI: €1.18-1.44) (£1.10, 95%CI: £0.99-1.21) (Table 3). While a ROI larger than 1 implies that enhanced TAU is cost-effective, it is important to realize that this value is used as a benchmark for the ROI of other scenarios with the new interventions included.

R1R2R3R4R5R6R7R8R9


Chapter 4

82

The following scenarios show the ROI of the added psychological interventions. As said before, in all scenarios the psychological intervention is offered to 20% of the patients with a history of depression and impacts are evaluated over a 5 year period.

Scenario A: Preventive cognitive therapyThe per-patient costs of CT of €370 (£312) do not put a lot of pressure on overall budget. When offering CT health care total expenditure increases with only 3.2%. With an average RR of 0.68 (95%CI:0.54-0.87), CT reduces the disease burden of major depression and leads to an increase in health effects of 13.8%. Since health effects increase more than costs, the ROI for adding CT increases from €1.30 to €1.43 ( 95%CI: €1.30-€1.58) (from £1.10 to £1.21, 95%CI: £1.10-£1.33).

Scenario B: Interpersonal therapyIPT, being a labour-intensive intervention, has a per-patient cost of approximately €2,080 (£1.753) and places substantial pressure on budget. When offering IPT, spending increases with 23.3%. With an average RR of 0.41 (95%CI:0.27-0.63), IPT is effective in increasing health gains with 24.9%. Since costs and health effects increase by roughly the same amount, the ROI for adding IPT is more or less unaltered (€1.31; 95%CI: €1.20-1.45; £1.10, 95%CI: £1.01-1.22).

Scenario C: Mindfulness-based cognitive therapyMCT, always offered in a group format, has a per-patient cost of approximately €370 (£312). When offering MCT health care expenditure increases with only 2.6%. With an average RR of 0.66 (95%CI:0.53-0.82), MCT is effective in reducing the disease burden and thus increases health gains with 14.5%. Since health effects increase more than costs, the ROI for adding MCT increases from €1.30 to €1.45 (95%CI: €1.31-1.60) (from £1.10 to £1.22, 95%CI: £1.10-1.35).

Scenario D: supported self-help PCTSupported self-help, being an intervention that does not rely much on therapist’s time, has a per-patient cost of only €178 (£150). Our simulations indicate that the supported self-help intervention only needs to be moderately effective, with an RR of 0.71, to become as cost-effective as MCT, which showed the best ROI (€1.45; £1.22). This scenarios differs qualitatively from the previous scenarios in the sense that the effectiveness of this intervention is not yet known and we just determined the effectiveness threshold for supported self-help to become the most cost-effective approach.



83

4

Sensitivity analysisIn order to test the robustness of our results the coverage rate of 20% was increased to 40% as the highest attainable coverage. In this scenario, the ROI for IPT stays more or less the same (€1.33; 95%CI: €1.20-1.47; £1.12; 95%CI: £1.01-1.24). However, at a coverage rate of 40% the ROI of CT and MCT increases to €1.55 (95%CI: €1.41-1.71; £1.31; 95%CI: £1.19-1.44) and €1.58 (95%CI: €1.44-1.75; £1.33; 95%CI: £1.21-1.48) respectively. The minimum RR of the supported self-help, needed to compete with the highest ROI is a moderate 0.71.

Table 3. Overview of effects, costs and ROIs per intervention at different coverage rates

Intervention RR95%CI

Costs per patient

ROI (€) ROI(£*)95%CI 95%CI

20% coverage rate

ROI (€) ROI(£*)95%CI 95%CI

40% coverage rate

enhanced TAU n/a n/a 1.301.18 - 1.44

1.100.99-1.21

1.301.18 - 1.44

1.100.99-1.21

CT 0.68 0.54-0.87

370 1.431.30-1.58

1.211.10-1.33

1.551.41-1.71

1.311.19-1.44

IPT 0.410.27-0.63

2.080 1.311.20-1.45

1.101.00-1.22

1.331.20-1.47

1.121.01-1.24

MCT 0.66 0.53-0.82

370 1.451.31-1.60

1.221.10-1.35

1.581.44-1.75

1.331.21-1.48

supported self-help

0.71** 178 1.451.32-1.59

1.221.11-1.34

*Purchasing Power Parities (PPP) over 2013 were used for conversion from euro (€) to UK pound (£), found at the Organisation for Economic Cooperation and Development (OECD). http://stats.oecd.org/Index.aspx?DataSetCode=PPPGDP**Since the effects of this supported self-help are not yet known, we simulated the minimum effectiveness needed for the self-help intervention to be able to dominate the most cost-effective available intervention which is MCT with an ROI of €1.45 (20% coverage rate).

DISCUSSION

Main findingsThis study aims to answer the following research questions:1. Do preventive psychological interventions for recurrences in depressive disorder

improve the cost-effectiveness over and above enhanced TAU? 2. What type of preventive psychological intervention makes the largest contribution in

improving the health care system’s cost-effectiveness? 3. At what effectiveness threshold becomes a hypothetical low-cost supported self-help

intervention the preferred approach?

R1R2R3R4R5R6R7R8R9


Chapter 4

84

The ROI of IPT (€1.31; 95%CI: € 1.20-1.45; £1.10, 95%CI: £1.01 -1.22) more or less equals the ROI of the base-case scenario of ‘enhanced TAU’ (€1.30; 95%CI: €1.18 - 1.44; £1.10, 95%CI: £0.99-1.21). However, CT and MCT are estimated to make the healthcare system more cost-effective with ROIs of €1.43 (95%CI: €1.30-1.58; £1.21, 95%CI: £1.10-1.33) and €1.45 (95%CI: €1.31-€1.60; £1.22, 95%CI: £1.10-1.35), respectively.

To produce a competitive ROI, the effectiveness of a hypothetical low-cost supported self-help intervention only needs to be successful in reducing the risk of a new depressive episode by a moderate RR of 0.71.

Adjusting the rather conservative coverage rate of 20% to a coverage rate of 40% does not impact the ROIs of IPT but makes CT and MCT even more attractive options for health care policy.

Strengths and limitationsIn our study we used a Markov simulation model called DepMod. One of the benefits of a simulation model is that it makes it possible to conduct “if-then” analyses, allowing evaluation even of hypothetical interventions. This is helpful when exploring policy options by healthcare decision makers.That said, our study has a number of limitations that need to be acknowledged.

First, in health economic modelling, much depends on the assumptions made in the model.

Assumptions in DepMod were as conservative as possible such that results are likely to portray not an overly optimistic outcome scenario. It should also be noted that the base-case scenario which forms the basis for the comparisons is an evidence-based healthcare system that is in full agreement with the Dutch clinical guidelines for the treatment of recurrent depression (enhanced TAU. This system is likely to be better than the current Dutch healthcare system. As a consequence, the ROI of the base-case scenario in this study (€1.30; £1.10) is likely to over-estimate the ROI of the present health care system for depressive disorders in the Netherlands.

Another limitation is that the current simulations are based on the Dutch population in the 18–65 year age bracket. Also the (enhanced) care scenario is modelled after the Dutch health care system for depressive disorder. Therefore, some caution should be applied when transferring our results to other countries.

Third, data on effectiveness of psychological interventions are drawn from a meta-analysis by Biesheuvel et al (2014). In this meta-analysis, preventive psychological interventions for recurrent depression are compared to TAU. The definition of TAU in this meta-analysis differs from the definition of ‘enhanced TAU’ in DepMod. The clinical data on effectiveness of the preventive psychological interventions that are used in the model are therefore probably too optimistic.



85

4

Fourth, in our study we did not include the implementation costs of changing from the current health care system to an alternative scenario. Instead, we looked at “steady state” scenarios, after implementation and asked ourselves what scenario would then be most appealing from a cost-effectiveness point of view.

Finally, owing to poor reporting, some aspects of the new interventions like costs of materials (e.g. therapy books) or costs due to time spent by the patients at e.g. homework or travelling, were not taken into account. As a consequence, the ROI of all scenarios is overestimated.

Implications for practiceOnly few studies on cost-effectiveness of psychological interventions for the prevention of depression have been published 28–33. From a health care perspective, all these economic evaluations show that preventive psychological interventions may represent good value for money versus TAU and/or ADM. Our study adds to this by showing that CT and MCT have the potential to improve the cost-effectiveness of the healthcare system. Due to the current economic down-turn, the squeeze on government budgets, and an expenditure for mental health care that is unlikely to be sustainable 34, governments are forced to allocate resources to these interventions that are seen to maximise cost-effectiveness.

Besides cost-effectiveness, patients’ preferences also play an important role in choosing one intervention over another. Currently, maintenance antidepressants are the mainstay approach for the prevention of relapse or recurrence. However, many patients express a preference for psychosocial interventions that provide long-term protection against relapse or recurrence.

Based on both cost-effectiveness and preference, a widespread introduction of MCT and CT may thus be preferred but poses a great challenge; currently, most depression is treated in primary care where there is a lack of cognitive therapists. Therefore, access to these therapies could only improve if primary care health nurses are trained to support the self-help interventions in primary care. Studies already showed that mental health nurses provide high quality psychological interventions in primary care if they are closely supervised by experienced therapists in a collaborative care model 35–37. As a positive side-effect, costs would even decrease and thus improve ROI.

Finally, this study showed that the hypothetical low-cost supported self-help may be a potentially attractive alternative. However, the exact role of the nurse, duration of treatment and follow-up and type of contact (telephone/internet/face-tot-face) all require empirical support. More research should demonstrate the actual value of this intervention.

R1R2R3R4R5R6R7R8R9


Chapter 4

86

REFERENCES

1. Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry 2005;62 (0003-990X (Print)):617-627.


3. Judd LL. The clinical course of unipolar major depressive disorders. Archives of General Psychiatry 1997;54 (0003-990X (Print)):989-991.

4. Kessler RC, Walters EE. Epidemiology of DSM-III-R major depression and minor depression among adolescents and young adults in the National Comorbidity Survey. Depression and anxiety 1998;7(1):3-14. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9592628. Accessed January 13, 2015.

5. Kessler RC, Zhao S, Blazer DG, Swartz M. Prevalence, correlates, and course of minor depression and major depression in the National Comorbidity Survey. Journal of affective disorders 1997;45(1-2):19-30. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9268772. Accessed January 13, 2015.


7. Hardeveld F, Spijker J, De Graaf R, Nolen WA, Beekman ATF. Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatr Scand 2010;122 (3):184-191.





12. Biesheuvel-Leliefeld KEM, Kok GD, Bockting CLH, et al. Effectiveness of psychological interventions in preventing recurrence of depressive disorder: Meta-analysis and meta-regression. Journal of affective disorders 2015;174C:400-410. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25553400. Accessed March 2, 2015.

13. Lokkerbol J, Adema D, Cuijpers P, et al. Improving the cost-effectiveness of a healthcare system for depressive disorders by implementing telemedicine: a health economic modeling study. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry 2014;22(3):253-62. Available at: http://www.sciencedirect.com/science/article/pii/S1064748113000638. Accessed February 3, 2015.


15. Segal Z, Williams JM, Teasdale JD. Mindfulness-Based Cognitive Therapy for Depression: A New Approach to Preventing Relapse. . New York: Guilford; 2002.

16. Klerman GL, Budman S, Berwick D, et al. Efficacy of a brief psychosocial intervention for symptoms of stress and distress among patients in primary care. Med Care 1987;25 (0025-7079 (Print)):1078-1088.

17. Weissman M, Markowitz JC, Klerman GL. Clinician’s Quick Guide to Interpersonal Psychotherapy. New York: Oxford University Press; 2007. . New York: Oxford University Press ; 2007.

18. Biesheuvel-Leliefeld KEM, Kersten SM a, van der Horst HE, et al. Cost-effectiveness of nurse-led self-help for recurrent depression in the primary care setting: design of a pragmatic randomised controlled trial. BMC psychiatry 2012;12(1):59. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3403967&tool=pmcentrez&rendertype=abstract. Accessed February 27, 2013.

19. Oostenbrink J, Bouwmans C, Koopmanschap M, Rutten F. Guideline for Economic Evaluations in Healthcare. Institute for Medical Technology Assessment; 2004.



87

4

20. Lokkerbol J, Adema D, Cuijpers P, et al. Improving the Cost-Effectiveness of a Healthcare System for Depressive Disorders by Implementing Telemedicine: A Health Economic Modeling Study. Am J Geriatr.Psychiatry 2013;(1545-7214 (Electronic)).

21. Bijl R V, de GR, Ravelli A, Smit F, Vollebergh WA. Gender and age-specific first incidence of DSM-III-R psychiatric disorders in the general population. Results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Social Psychiatry and Psychiatric Epidemiology 2002;37 (0933-7954 (Print)):372-379.

22. De Graaf R, Ten Have M, van Gool C, van Dorsselaer S. Prevalence of mental disorders, and trends from 1996 to 2009. Results from NEMESIS-2. Tijdschrift voor Psychiatrie 2012;54 (0303-7339 (Print)):27-38.

23. Briggs AH, Claxton K, Sculpher MJ. Decision Modelling for Health Economic Evaluation. Oxford, England: Oxford University Press; 2006. Available at: http://books.google.com/books?hl=nl&lr=&id=OgJOllOt_dkC&pgis=1. Accessed January 13, 2015.

24. Alonso J, Bernert S, Bruffaerts R, et al. Use of mental health services in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatrica Scandinavica.Supplementum 2004;(0065-1591 (Print)):47-54.

25. Andrews G, Issakidis C, Sanderson K, Corry J, Lapsley H. Utilising survey data to inform public policy: comparison of the cost-effectiveness of treatment of ten mental disorders. British Journal of Psychiatry 2004;184 (0007-1250 (Print)):526-533.

26. Chisholm D, Sanderson K, Ayuso-Mateos JL, Saxena S. Reducing the global burden of depression: population-level analysis of intervention cost-effectiveness in 14 world regions. Br.J Psychiatry 2004;184 (0007-1250 (Print)):393-403.

27. De Graaf R, ten Have M, van Dorsselaer S. De Psychische Gezondheid van de Nederlandse Bevolking. NEMESIS-2: Opzet En Eerste Resultaten. Utrecht: Trimbos Instituut; 2010.

28. Kuyken W, Byford S, Taylor RS, et al. Mindfulness-based cognitive therapy to prevent relapse in recurrent depression. J Consult Clin Psychol. 2008;76 (6):966-978.

29. Lynch FL, Hornbrook M, Clarke GN, et al. Cost-effectiveness of an intervention to prevent depression in at-risk teens. Archives of general psychiatry 2005;62(11):1241-8. Available at: http://archpsyc.jamanetwork.com/article.aspx?articleid=209034. Accessed January 13, 2015.

30. Mihalopoulos C, Vos T, Pirkis J, Smit F, Carter R. Do indicated preventive interventions for depression represent good value for money?. Australian and New Zealand Journal of Psychiatry 2011;45 (1440-1614 (Electronic)):36-44.

31. Smit F, Willemse G, Koopmanschap M, Onrust S, Cuijpers P, Beekman A. Cost-effectiveness of preventing depression in primary care patients: randomised trial. British Journal of Psychiatry 2006;188 (0007-1250 (Print)):330-336.

32. Van den Berg M, Smit F, Vos T, van Baal PHM. Cost-effectiveness of opportunistic screening and minimal contact psychotherapy to prevent depression in primary care patients. Hashimoto K, ed. PloS one 2011;6(8):e22884.

33. Vos T, Corry J, Haby MM, Carter R, Andrews G. Cost-effectiveness of cognitive–behavioural therapy and drug interventions for major depression. Australian and New Zealand Journal of Psychiatry 2005;39(8):683-692. Available at: http://anp.sagepub.com/content/39/8/683. Accessed January 13, 2015.

34. Forti A, Nas C, van Geldrop A, et al. Mental Health Analysis Profiles (MhAPs) Netherlands, Health Working Papers.(2014). Available at: http://www.ggznederland.nl/uploads/assets/OECD 2014 MhAP The Netherlands.pdf.

35. van’t Veer-Tazelaar PJ, van Marwijk HW, van OP, et al. Stepped-care prevention of anxiety and depression in late life: a randomized controlled trial. Archives of General Psychiatry 2009;66 (1538-3636 (Electronic)):297-304.

36. Schreuders B, van MH, Smit J, Rijmen F, Stalman W, van OP. Primary care patients with mental health problems: outcome of a randomised clinical trial. British Journal of General Practice 2007;57 (0960-1643 (Print)):886-891.

37. Van Schaik A, van Marwijk H, Ader H, et al. Interpersonal psychotherapy for elderly patients in primary care. American Journal of Geriatric Psychiatry 2006;14(1064-7481 (Print)):777-786.

5Cost-effectiveness of nurse-led self-help for recurrent

depression in the primary care setting:

design of a pragmatic randomised controlled trial

Karolien EM Biesheuvel-Leliefeld Sandra MA Kersten

Henriette E van der Horst Anneke van Schaik Claudi LH Bockting

Judith E Bosmans Filip Smit

Harm WJ van Marwijk

Published in: BMC Psychiatry, 2012; 12:59

R1R2R3R4R5R6R7R8R9


Chapter 5

90

ABSTRACT

Background Major Depressive Disorder is a leading cause of disability, tends to run a recurrent course and is associated with substantial economic costs due to increased healthcare utilization and productivity losses. Interventions aimed at the prevention of recurrences may reduce patients’ suffering and costs. Besides antidepressants, several psychological treatments such as preventive cognitive therapy (PCT) are effective in the prevention of recurrences of depression. Yet, many patients find long-term use of antidepressants unattractive, do not want to engage in therapy sessions and in the primary care setting psychologists are often not available. Therefore, it is important to study whether PCT can be used in a nurse-led self-help format in primary care. This study sets out to test the hypothesis that usual care plus nurse-led self-help for recurrent depression in primary care is feasible, acceptable and cost-effective compared to usual care only.

Design Patients are randomly assigned to ‘nurse-led self-help treatment plus usual care’ (134 participants) or ‘usual care’ (134 participants). Randomisation is stratified according to the number of previous episodes (2 or 3 previous episodes versus 4 or more). The primary clinical outcome is the cumulative recurrence rate of depression meeting DSM-IV criteria as assessed by the Structured-Clinical-Interview-for-DSM-IV- disorders at one year after completion of the intervention. Secondary clinical outcomes are quality of life, severity of depressive symptoms, co-morbid psychopathology and self-efficacy. As putative effect-moderators, demographic characteristics, number of previous episodes, type of treatment during previous episodes, age of onset, self-efficacy and symptoms of pain and fatigue are assessed. Cumulative recurrence rate ratios are obtained under a Poisson regression model. Number-needed-to-be-treated is calculated as the inverse of the risk-difference. The economic evaluation is conducted from a societal perspective, both as a cost-effectiveness analysis (costs per depression free survival year) and as a cost-utility analysis (costs per quality adjusted life-year).

Discussion The purpose of this paper is to outline the rationale and design of a nurse-led, cognitive therapy based self-help aimed at preventing recurrence of depression in a primary care setting. Only few studies have focused on psychological self-help interventions aimed at the prevention of recurrences in primary care patients.

Trial registration NTR3001 (www.trialregister.nl)


Design of the Parade-study; a randomised trial in primary care

91

5

BACKGROUND

Recurrent major depressive disorderMajor Depressive Disorder (MDD) is a leading cause of disease burden and is associated with significant healthcare costs and costs stemming from productivity losses1,2. MDD’s disease burden stems largely from its recurrent nature3. Each additional depressive episode increases the risk of recurrence by 18 %4. Interventions aimed at the prevention of recurrences in recovered patients may significantly reduce the burden of depression3. Up till now, maintenance treatment has been largely based on antidepressants (AD). However, the evidence-base to support such prolonged treatment is poor5-8 and moreover there is no evidence when to stop AD since most studies restricted their follow up to no longer than 2 years 9. Furthermore, many depressed patients prefer psychological treatments to drugs10 and many are not willing to take AD for prolonged periods of time11,12. Also, adherence in AD users is estimated at only 50 % at best11-13. As such, there is a need for an accessible alternative to maintenance treatment with AD. Psychological interventions might offer an interesting alternative to prevent recurrence of MDD in recovered patients.

TerminologyAt this point it might be well to introduce some terminology. MDD tends to run a relapsing and recurrent course. Both relapse and recurrence refer to the reappearance of a full-blown MDD after a symptom-free period. The essential distinction between both terms is the time at which each event occurs (Figure 1). According to the description by Frank et al.15, relapse is defined as ‘a return of symptoms satisfying full syndrome criteria for an episode that occurs during a period of remission, but before recovery’. As for recurrence, this is defined as ‘the appearance of a new episode of MDD, occurring during recovery’. Conceptually, this represents the beginning of a new episode of an illness. Treatment stages can be defined accordingly: following remission and before recovery ‘continuation treatment’ is offered. Following recovery, treatment enters the maintenance stage. Both treatments have the aim to prevent recurrences.

R1R2R3R4R5R6R7R8R9


Chapter 5

92

Figure 1. Overview of response, remission, recurrence, and relapse in relation to the treatment phase Modified after Tohen et al. (2009)16 © 2009 Blackwell Munksgaard14.

For research purposes a consensus definition recommends that a 6-month threshold is justifiable to distinct between remission and recovery because of the median duration of MDD15. In patients experiencing relapsing/recurring MDD however, this threshold can be debated because of the mostly shorter duration of an episode. In this trial no distinction is made between relapse and recurrence (henceforth called ‘recurrence’) nor between remission and recovery (henceforth ‘recovery’), nor between continuation and maintenance treatment (henceforth ‘maintenance treatment’). Exact definitions and cut-off points of all terms with respect to this trial are handled in detail later (see ‘Eligibility of participants’).

Efficacy of psychological interventionsThe literature shows that psychological interventions may offer a good alternative or a welcome adjunct to AD. Hollon et al.17 concluded in their review that Cognitive Behavioural Therapy (and especially CT) is as efficacious as medications in the treatment of MDD and that CT has an enduring effect that protects against subsequent relapse and possibly recurrence regardless of when it is applied17. Continuation/maintenance CT has been found to reduce risk for relapse/recurrence in MDD in a trial of Jarrett et al.18. A meta-analysis by Vittengl et al.19, including 28 studies and comprising 1,880 adults, demonstrated that among acute-phase treatment responders, continuation of CBT compared to assessment only or clinical management, reduced the number of recurrences substantially from 73 % to 40 %, over a mean of 153 weeks follow-up19. Specific protocols based on CBT have been developed for the prevention of depression, such as Preventive Cognitive Therapy (PCT). For patients with a history of 4 or more depressive episodes, 75 % receiving PCT experienced



93

5

a recurrence over a period of 5,5 years versus 95 % receiving usual care20. PCT targets underlying cognitive vulnerability factors, such as dysfunctional cognitions that are easily reactivated in recovered patients and therefore may cause vulnerability for recurrence. The treatment focuses on identifying and changing these vulnerability factors, while at the same time reinforcing specific memories of positive experience by keeping a diary of positive experiences and formulating specific recurrence prevention strategies21.

The case for guided self-helpIn the Netherlands, evidence-based psychological treatments are less readily accessible in primary care 10 because they require specific expertise, extensive training and draw on scarce resources. Furthermore, reimbursement of psychological treatments in primary care through insurance is becoming limited. Because the vast majority of persons with a high risk of developing a new episode visits - and receives treatment from - their own primary care physician (PCP), this seems to be the most appropriate coordinator of preventive interventions22. These interventions should be cost-effective, readily accessible at the primary care or community level, acceptable for patients and health care providers, and should easily be integrated into current care.

Self-help interventions using self-help books (bibliotherapy) are one of the most accessible forms of psychological interventions for primary care patients. Research indicates that self-help has a moderate to large effect in reducing symptoms of depression and anxiety23-26. PCT can easily be transformed into a self-help intervention because of its structured design. Some form of support however, should be provided to enhance patients’ compliance, which in turn is associated with better treatment response overall26-29. There is growing evidence that mental health nurses or social workers can effectively deliver self-help treatment protocols for depression, particularly in chronic care models30,31. For both economic and pragmatic reasons it is attractive to let nurses play a pivotal and facilitating role in the provision of PCT instead of a psychologist.

Objectives The primary objective of this study is to evaluate whether nurse-led, cognitive treatment based self-help in addition to usual care is cost-effective in preventing recurrences for patients at high risk of recurrent MDD in primary care compared to usual care alone. Furthermore, this study examines whether the addition of nurse-led self-help to usual care for patients with recurrent MDD is effective in improving health related quality of life, in reducing co-morbid distress, anxiety and/or somatisation, in improving self-efficacy and meets with patients’ satisfaction. Finally, we examine which socio-demographic and clinical variables (e.g. pain and fatigue) moderate treatment response.

R1R2R3R4R5R6R7R8R9


Chapter 5

94

DesignThis study is a multi-site, pragmatic randomised controlled trial among primary care patients with recurrent MDD who are currently recovered. Patients are recruited through primary care practices and are randomly assigned to two parallel groups: ‘nurse-led self-help treatment plus usual care’ (134 patients) or ‘usual care alone’ (134 patients). A flowchart is shown in Figure 2. It is not possible to blind neither patients nor healthcare providers to the intervention in this study due to the nature of this self-help intervention.

Search through databases from PC practices, employing a nurse (search ‘ICPC codes for i.e. depression’, ‘antidepressants’ , ‘depression in free field’ , referral)

Eligible patients are sent a short information letter with response form

Interested patients reply and are sent an extensive information letter, response form and informed consent

The SCID-1 is assessed for inclusion / exclusion criteria

Does 1) patient meet full inclusion criteria and 2) sign informed consent?

Baseline measurements and randomisation

yes

Nurse- led self-help + CAU Care as usual (CAU)

Assessments (T = week 1-8 and month 3, 6, 9, 12 and 15 in both arms)

noexclude

Stratification variable;

no of previous episodes

Interested patients reply by response form and signed informed consent

Figure 2. Flowchart

Eligibility of participantsPatients are eligible for participation in the trial when they: 1) are between 18 and 65 years old, 2) have had at least 2 previous depressive episodes 3) are currently recovered, 4) are fluent in reading and speaking Dutch and 5) have access to the internet. Criterion for ‘currently recovered’ includes ‘no diagnose of depression according to the Structured Clinical Interview for DSM-IV (SCID-I)32. The recovered episode should last for longer than 8 weeks and no longer than 2 years.

Participants who have current (hypo) mania or a history of bipolar disease, any current organic brain disorder, psychotic disorder or severe sensory disabilities are excluded. Patients who have drug or alcohol related abuse or dependence as main diagnosis are also excluded. These exclusion criteria are checked in patient files.



95

5

RecruitmentThe PCP screens for eligible patients by searching the database of the primary care practice using the following indicators33: antidepressants prescription, strong free text indication of depression or a history of depression according to the International Classification of Primary Care (ICPC) codes in the PCP’s patient files. Subsequently, the PCP approaches potentially eligible patients with global information about the study, contact information and a global screening form comprising three questions: 1) did you experience 2 or more depressive episodes? 2) are you currently recovered? and 3) did the last episode end longer than 8 weeks- and no longer than 2 years ago?. If a patient is eligible for the trial based on the global screening and if the patient agrees upon receiving more information, he is sent an extensive study information letter and an informed consent form for participation in the trial. Consenting patients are assessed for their eligibility in more detail using the SCID-I. Eligible patients who sign informed consent enter the trial.

The interventionSelf-helpThe intervention consists of nurse-led self-help treatment based on PCT21,34. Patients are offered a self-help book including background literature for further reading and assignments. This self-help book enables patients to follow the treatment in the privacy of their own homes and at a pace that suits them best. The treatment consists of eight weekly modules with a fixed structure and it takes approximately 1.5 hour each week to complete the assignments. Additionally, patients fill out the electronic Q-IDS-SR weekly to monitor the severity of their depressive symptoms.

Nurse-led supportThe nurse lends minimal support to help patients work through the self-help intervention. Prior to the start of the treatment, a face-to-face meeting with the nurse (with a maximum duration of 30 minutes) is planned at the primary care practice. This meeting involves a discussion of current symptoms, motivational interviewing, psycho-education on the course and treatment of recurrent depression and an introduction to the nurse-led self-help treatment on the basis of the self-help book. Afterwards, the self-help treatment starts and eight weekly telephone contacts (at a maximum of 15 minutes) follow, initiated by the nurse. During these telephone contacts the nurse explores how the patient fares with the self-help treatment according to a strict protocol. In the weekly contacts the nurse asks the following questions: 1) did the patient fill out the electronic Q-IDS-SR questionnaire? 2) did the patient read and understand the literature belonging to that week? 3) did the patient complete the accompanying assignments? and 4) what difficulties did the patient experience in his assignments? After answering these questions, patients are shortly introduced to next

R1R2R3R4R5R6R7R8R9


Chapter 5

96

week’s literature and exercises. The contact is supportive, activating and facilitating and the nurse does not engage in a therapeutic relationship with the patient. If a nurse notices the emergence of depressive symptoms during a regular phone-contact or a patient brings up feeling depressed, the nurse emphasizes specific parts of the treatment that may help the patient to better cope with these symptoms in order to prevent recurrence. If a patient expresses suicidal thoughts the PCP is notified immediately. After each contact the nurse summarizes the conversation (including the questions) in an electronic journal using a checklist. This journal is a way to both monitor and promote treatment integrity on the side of the patient (did the patient read and apply the literature) and the nurse (did the nurse go through all the questions).

Training and supervision of nursesThe mental health nurse attends an one-day training, during which attention is paid to the protocol and content of PCT and to guiding a self-help treatment. The nurse is also taught to let recovered patients deal with symptoms. Since participating nurses already have experience with cognitive therapy, an one-day training is sufficient. The training is delivered by two trained psychologist. To detect adherence and/or competence issues, audiotaped telephone contacts with two patients are evaluated during supervision sessions for each nurse, before the actual start of the trial. During the trial, nurses can contact their supervisors at any time for additional questions and feedback.

Usual careIn both treatment conditions, treatment as usual involves usual care (i.e. standard/routine treatment, including no treatment) as typically provided by the PCP according to the Dutch PCP clinical guidelines (NHG-guidelines)35. These guidelines recommend continuation of treatment with AD, preventive psychological treatment or both, depending on the distress, level of dysfunctioning, psychological or physical co-morbidity and preferences of the patient.

In this trial, usual care is not restricted during the period from entry to end of follow-up. By adding no restrictions to usual care, the findings of this study are more generalisable. There usually is some inter-practice variation in treatment despite the clinical guidelines. It is therefore important to obtain a clear understanding of what (additional) treatments are received by patients in both arms of the trial. To that end, data is collected on health care utilization, using the TiC-P36 – the most commonly used health care receipt interview in the Netherlands (see below for more details on the measurements) and the Medication Adherence Questionnaire (MAQ)37. As the use of usual care in both arms in this trial brings the risk of behavioural change by caregivers and patients because of the information that is supplied38, minimum information is supplied to any participating person (see Discussion for further details).



97

5

Sample sizeTo calculate the sample size of the intervention and control group, we combined rudimentary findings from previous randomised controlled trials (RCT’s), which resulted in a mean recurrence rate of 33 % after 2 years of follow-up versus 67 % in the control group (active and non-active control). Based on this we assume a risk reduction of 20 % rate in this study between the two conditions. To detect this 20 % risk reduction in a 2-sided test at alpha = 0.05 and a power of 1-beta = 0.80, 107 patients in each condition are required. Compensating for loss to follow-up of 20 % over the whole 15 months follow-up, requires (107/0.8=) 134 participants at baseline in each trial arm.

RandomisationPatients who are eligible for the trial and who have given their informed consent, are randomised to ‘nurse-led self-help treatment plus usual care’ or ‘usual care alone’. An independent researcher performs randomisation centrally (Random Allocation Software version 1.0.0), using a blocked randomisation scheme with blocks of 2 patients. The researchers receive the participant’s number and automatically random generated condition in the trial by email. Randomisation is pre-stratified for the number of previous depressive episodes (2 or 3 episodes versus 4 or more episodes). The rationale for stratification based on this variable is that several subgroup analyses suggest that PCT is more effective in ‘high risk’ patients, meaning patients with a history of 4 or more episodes on a lifetime basis21,39,40.

Outcome measurementsFor an overview of assessments at baseline, during the intervention and during follow-up, see Table 1. The primary researcher (KB) conducts collection and analysis of the data with help of a research assistant (SK).

Table 1. Overview of assessments

Measure Description T0 w1,8 T1 T2 T3 T4 T5InterviewsSCID-I32 DSM-IV Axis I Disorders + + + + + +Self-report measuresQ-IDS-SR41 Depressive symptoms + + + + + + +EQ-5D42 Quality of life + + + + + +SF-1243 Quality of life + + + + + +TIC-P36 Direct/indirect costs + + + + + +4DSQ44 Comorbid psychopathology + + +General self-efficacy scale45 Self-efficacy + + +FSS46 Severity of fatigue +MPQ-DLV47 Severity/evaluation of pain +MAQ37 Medication Adherence + + + + + +CSQ-848 Satisfaction +

T0 = baseline, w1,8 = week 1–8, T1 = 3 months, T2 = 6 months, T3 =9 months, T4 = 12 months, T5 =15 months

R1R2R3R4R5R6R7R8R9


Chapter 5

98

DemographicsAt baseline the socio-demographic characteristics of participants are collected (age, gender, educational level, marital status, etc.). The number and duration of previous depressive episodes, age of onset of first depressive episode and kind of treatments received are also assessed at baseline.

Primary outcomeThe primary clinical outcome is the cumulative recurrence rate of depression meeting DSM-IV criteria for a major depressive episode49. Recurrence of depression (current or since the last assessment point) is assessed in both arms with the SCID-I 32 at 3, 6, 9, 12 and 15 months follow-up by a trained researcher and research assistant. Interviews are randomly audiotaped and evaluated for integrity reasons. The incidence rate in the intervention group is compared to the incidence rate in the control group and thus expressed as the (cumulative) incidence rate ratio.

Secondary outcomesSecondary clinical outcomes include health related quality of life (measured with both the ‘EuroQol’ (EQ-5D)42 and ‘Short Form-12’ (SF-12)43), severity of depressive symptoms (measured with the Q-IDS-SR 41), co-morbid distress, anxiety and somatisation (measured with the ‘Four Dimension Symptom Questionnaire’, 4-DSQ 44) and self-efficacy (measured with the ‘General Self Efficacy Scale’, GSES 45). Quality adjusted life years (QALY’s) are calculated based on both the EuroQol 42 and SF-12 43, using the Dutch tariff estimated by Lamers et al.50 and using Brazier’s algorithm51, respectively. All secondary clinical outcomes are measured at baseline and at 9 and 15 months follow-up.

Putative effect-moderatorsSeveral risk factors for recurrent depression have been identified52 and may also be relevant for predicting treatment response. These risk factors include non-adherence, demographic factors such as age and gender, high number and longer duration of previous episodes, younger age at the onset of the first depressive episode53, presence of residual symptoms54, low socioeconomic status55, low self-efficacy for managing depression55 and also symptoms of pain56 and fatigue57 appear to be risk indicators for imminent recurrence of depression. The type of treatment received during previous depressive episodes (psychological intervention/AD/no care) may also be relevant for treatment response. Patients who already had a psychological intervention may have the benefit of possible long-term protection for recurrence58. All of these factors are therefore assessed as putative effect-moderators.



99

5

Cost measuresCost-effectiveness is evaluated from a societal perspective meaning that the costs of the intervention, other health care utilization costs, patients’ out-of-pocket costs and costs due to productivity losses are included in the economic evaluation. Health care utilization is measured using the TiC-P36 at baseline and 3, 6, 9, 12 and 15 months follow-up. Medical costs that are assessed include costs related to the intervention, medication use, hospital admissions, and contacts with other healthcare professionals. For the valuation of health care utilization, standard prices published in the Dutch costing guidelines are used59. Medication use is valued using prices of the Royal Dutch Society for Pharmacy, including the costs of prescription by the PCP and the pharmacist’s dispensing costs. A cost price for the nurse-led self-help intervention is calculated using a bottom-up approach and will account for costs for personnel, patient materials and rental of practice spaces. Costs and effects exceeding 12 months follow-up are discounted, in accordance with the Dutch guideline for economic evaluation in health care60. Costs of productivity losses are estimated using the friction cost method61. In a secondary analysis, the human capital method is used to estimate productivity losses.

Analyses of clinical outcomesStandard descriptive methods (e.g. frequencies, percentages and means) are used to summarize the demographic and clinical features of the intervention and control group and to check whether the randomisation has resulted in a well-balanced design.

Analyses are conducted according to the intention-to-treat principle, meaning that all patients who have been randomised are included in the analyses. Missing endpoints are imputed using the Expectation-Maximization (EM) algorithm. To gauge the robustness of the outcomes, this analysis is repeated while using a Multiple-Imputation (MI) approach. All tests are conducted at P <0.05, 2-tailed. Additional completer analyses for all patients that attended at least 80 % of the telephone sessions are performed.

Analysis of primary outcomeCumulative recurrence rate ratios are estimated using a Poisson regression model. Number-needed-to-be-treated (NNT) is calculated as the inverse of the risk difference (RD) which is estimated using a linear probability model. Data-analysis takes into account that data are clustered within primary care practices and patients. The nested data structure entails violation of the usual assumption that data are uncorrelated. Therefore, all analyses are design-based, taking the clustered data structure into account, using Stata’s62 procedures for clustered data63.

R1R2R3R4R5R6R7R8R9


Chapter 5

100

Analysis of secondary outcomesThe effect of the self-help treatment on health related quality of life, symptom severity, co-morbid distress, anxiety and/or somatisation and self-efficacy is analysed by regressing these secondary endpoints with the randomisation condition, while correcting for baseline values.

Analysis of effect-moderationModerator analyses are conducted for socio-demographic and clinical variables. Subgroups that show particularly good response to the intervention are identified by regressing (P < 0.05) depression severity (measured with Q-IDS-SR) on the interaction term of treatment and baseline characteristics of the patients, namely number of previous depressive episodes, previous and current treatment, age of onset of first depression, self-efficacy in managing depression, symptoms of pain and fatigue and socio-demographic characteristics like gender, marital status, age and socio-economic status.

Analysis of economic dataMissing cost- and effect data are imputed using multiple imputation according to the MICE algorithm developed by van Buuren64. Costs typically have a highly skewed distribution. Policy makers want to have information on the difference in mean total costs between the two treatment groups in order to estimate the total health care budget needed for a specific condition65. Therefore, bias-corrected and accelerated bootstrapping with 5000 replications is used to estimate 95 % confidence intervals around the mean difference in total costs between the treatment groups.

Both a cost-effectiveness analysis (CEA, with depression-free person years as the clinical end term) and as a cost-utility analysis (CUA, with incremental costs per quality adjusted life years (QALY) gained as the clinical end-term) are performed. Bootstrapping is used to estimate the uncertainty surrounding the incremental cost-effectiveness ratios (ICERs) which are graphically presented on cost-effectiveness planes. Cost-effectiveness acceptability curves and net monetary benefits are also estimated66.

An incremental net benefit regression (INBR) analysis is conducted to address the research question in which groups the intervention is likely to be particularly cost-effective, analogous to the moderator analyses for clinical endpoints. The same set of variables is used in these INBR analyses. The incremental net benefit is calculated as Eλ – C. The first term is the number of units of effectiveness gained in the intervention group in comparison with the control group multiplied by the amount (λ) society is willing to pay (WTP) for a unit of effect gained. Because λ is unknown, we use a likely WTP-range. The product term is subtracted by the difference in costs between the groups yielding the net benefit expressed in monetary terms. Incremental net-benefits are analysed using a regression analysis approach 67 and helps to identify sub-groups for which the intervention is particularly cost-effective.



101

5

Patients’ satisfactionAt 15 months patients’ satisfaction is assessed using the Client Satisfaction Questionnaire (CSQ-8)48 in both the intervention and control group. Additionally, 20 ‘experimental’ patients who responded best, and 20 ‘experimental’ patients who responded worst in terms of recurrences are approached for an in-depth (qualitative) interview. Comparing responses from both groups may help to increase understanding what aspects of the intervention must change or remain intact.

DISCUSSION

Given its recurrent character, new minimal interventions are needed to prevent new episodes of MDD. As patients with recurrent MDD account for great societal costs, an intervention to prevent recurrence in the maintenance phase will potentially lead to great reduction in health care utilization and costs of absenteeism and presenteeism. This is as far as we are aware of, the first study that examines the cost-effectiveness of a nurse-led, cognitive treatment based self-help for patients with recurrent MDD in primary care. This innovative self-help format ensures that patients can complete the treatment in their own time and at their own homes, which makes it easy accessible. Besides, being led by nurses, the treatment is expected to be economically affordable and sustainable.

The effect of several socio-demographic and clinical variables on treatment response is assessed. This might lead to insights that will lead to the development of more targeted interventions.

The RCT-design of this trial is considered the ultimate test of a medical hypothesis, and is the support of evidence-based medicine. By adding no restrictions to usual care in this RCT, the findings of this study will be more generalisable.

Risks for adverse events in patients in the intervention arm are very low due to the psychological character of the intervention and because there is no restriction to usual care. In the case of a patient expressing suicidal thoughts, the PCP is notified. These procedures are made explicit in the informed consent papers and protocols.

A limitation of this trial is that there might not be a big contrast in primary outcome (cumulative rate of recurrence) between the 2 arms of the trial at the end of follow-up, because both arms include usual care. Another limitation is, as in any trial that involves psychological treatment, that it is not possible, due to the design of the intervention, to blind patients, health care providers and researchers to the patient’s randomised condition. Therefore it is not possible to prevent any behavioural change during the course of the trial. Patients’ behaviour in the control arm might be influenced by reading the study information letter and a PCP’s choice regarding usual care might be influenced by the knowledge of a

R1R2R3R4R5R6R7R8R9


Chapter 5

102

patient starting the intervention. Nurses might be influenced in their support for patients when guiding the self-help treatment because of their knowledge of the patients’ additional care. The information given to PCP’s, nurses and patients in both arms is therefore limited to a minimum to overcome this limitation.



103

5

REFERENCES

1. Murray CJ, Lopez AD: Regional patterns of disability-free life expectancy and disability-adjusted life expectancy: global Burden of Disease Study. Lancet 1997, 349:1347–1352.

2. Ormel J, Petukhova M, Chatterji S, Aguilar-Gaxiola S, Alonso J, Angermeyer MC, Bromet EJ, Burger H, Demyttenaere K: de GG et al.: Disability and treatment of specific mental and physical disorders across the world. Br J Psychiatry 2008, 192:368–375.

3. Vos T, Haby MM, Barendregt JJ, Kruijshaar M, Corry J, Andrews G: The burden of major depression avoidable by longer-term treatment strategies. Arch Gen Psychiatry 2004, 61:1097–1103.

4. Mueller TI, Leon AC, Keller MB, Solomon DA, Endicott J, Coryell W, Warshaw M, Maser JD: Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am J Psychiatry 1999, 156:1000–1006.

5. Paykel ES: Continuation and maintenance therapy in depression. Br Med Bull 2001, 57:145–159.

6. Van Loenen AC: Farmacotherapeutisch kompas, College voor Zorgverzekeringen; 2003.

7. Viguera AC, Baldessarini RJ, Friedberg J: Discontinuing antidepressant treatment in major depression. Harv Rev Psychiatry 1998, 5:293–306.

8. Piek E, van der Meer K, Nolen WA: Guideline recommendations for long-term treatment of depression with antidepressants in primary care–a critical review. Eur J Gen Pract 2010, 16:106–112.

9. Kaymaz N: van OJ, Loonen AJ, Nolen WA: Evidence that patients with single versus recurrent depressive episodes are differentially sensitive to treatment discontinuation: a meta-analysis of placebo-controlled randomized trials. J Clin Psychiatry 2008, 69:1423–1436.

10. van Schaik DJ, Klijn AF, van Hout HP, van Marwijk HW, Beekman AT: de HM, van DR: Patients’ preferences in the treatment of depressive disorder in primary care. Gen Hosp Psychiatry 2004, 26:184–189.

11. Bockting CL, ten Doesschate MC, Spijker J, Spinhoven P, Koeter MW, Schene AH: Continuation and maintenance use of antidepressants in recurrent depression. Psychother Psychosom 2008, 77:17–26.

12. Meijer WE, Heerdink ER, Leufkens HG, Herings RM, Egberts AC, Nolen WA: Incidence and determinants of long-term use of antidepressants. Eur J Clin Pharmacol 2004, 60:57–61.

13. ten Doesschate MC, Bockting CL, Koeter MW, Schene AH: Predictors of nonadherence to continuation and maintenance antidepressant medication in patients with remitted recurrent depression. J Clin Psychiatry 2009, 70:63–69.

14. Möller HJ, Riedel M, Seemüller F: Relapse or recurrence in depression: why has the cutoff been set at 6 months? Medicographia 2011, 33:125–131.

15. Frank E, Prien RF, Jarrett RB, Keller MB, Kupfer DJ, Lavori PW, Rush AJ, Weissman MM: Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Arch Gen Psychiatry 1991, 48:851–855.

16. Tohen M, Frank E, Bowden CL, Colom F, Ghaemi SN, Yatham LN, Malhi GS, Calabrese JR, Nolen WA, Vieta E, et al: The International Society for Bipolar Disorders (ISBD) Task Force report on the nomenclature of course and outcome in bipolar disorders. Bipolar Disord 2009, 11:453–473.

17. Hollon SD, Ponniah K: A review of empirically supported psychological therapies for mood disorders in adults. Depress Anxiety 2010, 27:891–932.

18. Jarrett RB, Kraft D, Doyle J, Foster BM, Eaves GG, Silver PC: Preventing recurrent depression using cognitive therapy with and without a continuation phase: a randomized clinical trial. Arch Gen Psychiatry 2001, 58:381–388.

19. Vittengl JR, Clark LA, Dunn TW, Jarrett RB: Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitive-behavioral therapy’s effects. J Consult Clin Psychol 2007, 75:475–488.

20. Bockting CL, Spinhoven P, Wouters LF, Koeter MW, Schene AH: Long-term effects of preventive cognitive therapy in recurrent depression: a 5.5-year follow-up study. J Clin Psychiatry 2009, 70:1621–1628.

21. Bockting CL, Schene AH, Spinhoven P, Koeter MW, Wouters LF, Huyser J, Kamphuis JH: Preventing relapse/recurrence in recurrent depression with cognitive therapy: a randomized controlled trial. J Consult Clin Psychol 2005, 73:647–657.

R1R2R3R4R5R6R7R8R9


Chapter 5

104

22. Bijl RV, Ravelli A: Psychiatrische morbididteit, zorggebruik en zorgbehoefte. Resultaten van de Netherlands Mental Health Survey and Incidence Study. Tijdschrift voor Gezondheidswetenschappen 1998, 76:446–57.

23. den Boer PC, Wiersma D, Van den Bosch RJ: Why is self-help neglected in the treatment of emotional disorders? A meta-analysis. Psychol Med 2004, 34:959–971.

24. Gregory RJ, Canning SS, Lee TW, Wise JC: Cognitive bibliotherapy for depression: A meta-analysis. Professional Psychology-Research and Practice 2004, 35:275–280.

25. McKendree-Smith NL, Floyd M, Scogin FR: Self-administered treatments for depression: a review. J Clin Psychol 2003, 59:275–288.

26. Van’t Hof E, Cuijpers P, Stein DJ: Self-help and Internet-guided interventions in depression and anxiety disorders: a systematic review of meta-analyses. CNS Spectr 2009, 14:34–40.

27. Marrs RW: A meta-analysis of bibliotherapy studies. Am J Community Psychol 1995, 23:843–870.

28. Newman MG, Erickson T, Przeworski A, Dzus E: Self-help and minimal-contact therapies for anxiety disorders: Is human contact necessary for therapeutic efficacy? J Clin Psychol 2003, 59:251–274.

29. Spek V, Cuijpers P, Nyklicek I, Riper H, Keyzer J, Pop V: Internet-based cognitive behaviour therapy for symptoms of depression and anxiety: a meta-analysis. Psychol Med 2007, 37:319–328.

30. Hunkeler EM, Meresman JF, Hargreaves WA, Fireman B, Berman WH, Kirsch AJ, Groebe J, Hurt SW, Braden P, Getzell M, et al: Efficacy of nurse telehealth care and peer support in augmenting treatment of depression in primary care. Arch Fam Med 2000, 9:700–708.

31. Mynors-Wallis LM, Gath DH, Day A, Baker F: Randomised controlled trial of problem solving treatment, antidepressant medication, and combined treatment for major depression in primary care. BMJ 2000, 320:26–30.

32. First MB, Gibbon M, Spitzer RL, Williams JBW: Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, D.C: American Psychiatric Press, Inc; 1996.

33. Joling KJ, van Marwijk HW, Piek E, van der Horst HE, Penninx BW, Verhaak P, van Hout HP: Do GPs’ medical records demonstrate a good recognition of depression? A new perspective on case extraction. J Affect Disord 2011, 133:522–527.

34. Bockting CL, Spinhoven P, Huibers MJ, Edited by Richards S, Perry M: Relapse prevention in recurrent depression using CBT, In: Relapse prevention for depression. Washington D.C: American Psychological Association; 2010.

35. Van Marwijk HWJ, Grundmeijer HGLM, Bijl D, Van Gelderen MG, De Haan M, Van Weel-Baumgarten EM, Burgers JS, Boukes FS, Romeijnders ACM: NHG-Standaard Depressieve stoornis (depressie) (Eerste herziening). Huisarts & Wetenschap 2003, 46:614–633.

36. Hakkaart-van Roijen L, Van Straten A, Donker M, et al: Manual:Trimbos/iMTA Questionnaire for Costs Associated with Psychiatric Illness (in Dutch). Erasmus University: Rotterdam; 2002.

37. Morisky DE, Green LW, Levine DM: Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care 1986, 24:67–74.

38. Smelt AF, van der Weele GM, Blom JW, Gussekloo J, Assendelft WJ: How usual is usual care in pragmatic intervention studies in primary care? An overview of recent trials. Br J Gen Pract 2010, 60:e305–e318.

39. Teasdale JD, Segal ZV, Williams JM, Ridgeway VA, Soulsby JM, Lau MA: Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol 2000, 68:615–623.

40. Ma SH, Teasdale JD: Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects. J Consult Clin Psychol 2004, 72:31–40.

41. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, Markowitz JC, Ninan PT, Kornstein S, Manber R, et al: The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol Psychiatry 2003, 54:573–583.

42. EuroQol group: EuroQol - a new facility for the measurement of health-related quality of life, Volume 16. Health Policy: The EuroQol Group; 1990:199–208.

43. Ware J Jr: Kosinski M, Keller SD: A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996, 34:220–233.



105

5

44. Terluin B, van Marwijk HW, Ader HJ, de Vet HC, Penninx BW, Hermens ML, van Boeijen CA, van Balkom AJ, van der Klink JJ, Stalman WA: The Four-Dimensional Symptom Questionnaire (4DSQ): a validation study of a multidimensional self-report questionnaire to assess distress, depression, anxiety and somatization. BMC Psychiatry 2006, 6:34.

45. Schwarzer R, Jerusalem M: Generalized Self-Efficacy scale, In Measures in Health Psychology: a user’s portfolio. Windsor, UK: Causal and control beliefs. Edited by Weinman J, Wright S, Johnston M; 1995:35–37.

46. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD: The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989, 46:1121–1123.

47. van der Kloot WA, Oostendorp RA: van der MJ, van den HJ: The Dutch version of the McGill pain questionnaire: a reliable pain questionnaire. Ned Tijdschr Geneeskd 1995, 139:669–673.

48. Larsen DL, Attkisson CC, Hargreaves WA, Nguyen TD: Assessment of client/patient satisfaction: development of a general scale. Eval Program Plann 1979, 2:197–207.

49. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders. Washington DC: American Psychiatric Association; 1994.

50. Lamers LM, Stalmeier PF, McDonnell J, Krabbe PF, van Busschbach JJ: Measuring the quality of life in economic evaluations: the Dutch EQ-5D tariff. Ned Tijdschr Geneeskd 2005, 149:1574–1578.

51. Brazier JE, Roberts J: The estimation of a preference-based measure of health from the SF-12. Med Care 2004, 42:851–859.

52. Le LA, Despiegel N, Francois C, Duru G: Can discrete event simulation be of use in modelling major depression? Cost Eff Resour Alloc 2006, 4:19.

53. Penninx BW, Nolen WA, Lamers F, Zitman FG, Smit JH, Spinhoven P, Cuijpers P, de Jong PJ, van Marwijk HW, der Meer KV, et al: Two-year course of depressive and anxiety disorders: Results from the Netherlands Study of Depression and Anxiety (NESDA). J Affect Disord 2011, 133(1–2):76–85.

54. Paykel ES: Partial remission, residual symptoms, and relapse in depression. Dialogues Clin Neurosci 2008, 10:431–437.

55. Van Voorhees BW, Paunesku D, Kuwabara SA, Basu A, Gollan J, Hankin BL, Melkonian S, Reinecke M: Protective and vulnerability factors predicting new-onset depressive episode in a representative of U.S. adolescents. J Adolesc Health 2008, 42:605–616.

56. Geerlings SW, Twisk JW, Beekman AT, Deeg DJ: van TW: Longitudinal relationship between pain and depression in older adults: sex, age and physical disability. Soc Psychiatry Psychiatr Epidemiol 2002, 37:23–30.

57. Skapinakis P, Lewis G, Mavreas V: Temporal relations between unexplained fatigue and depression: longitudinal data from an international study in primary care. Psychosom Med 2004, 66:330–335.

58. Segal ZV, Kennedy S, Gemar M, Hood K, Pedersen R, Buis T: Cognitive reactivity to sad mood provocation and the prediction of depressive relapse. Arch Gen Psychiatry 2006, 63:749–755.

59. Hakkaart-van Roijen L, Tan SS, Bouwmans CAM: Handleiding voor kostenonderzoek. Rotterdam: Instituut voor Medical Technology Assessment iov CZV; 2010.

60. Oostenbrink JB, Koopmanschap MA, Rutten C: Standardisation of Costs. Adis International: The Dutch Manual for Costing in Economic Evaluations; 2002.

61. Koopmanschap MA, Rutten FF, van Ineveld BM, van Roijen L: The friction cost method for measuring indirect costs of disease. J Health Econ 1995, 14:171–189.

62. Boston RC, Sumner AE: STATA: a statistical analysis system for examining biomedical data. Adv Exp Med Biol 2003, 537:353–369.

63. Cohen SB: An evaluation of alternative PC-based software packages developed for the analysis of complex survey data. The American Statistician 1997, 51:285–292.

64. Van Buuren S, Oudshoorn CGM: Multivariate imputation by chained equations. TNO Preventie en Gezondheid: MICE V1.0 user’s manual. Leiden; 2000.

65. Thompson SG, Barber JA: How should cost data in pragmatic randomised trials be analysed? BMJ 2000, 320:1197–1200.

R1R2R3R4R5R6R7R8R9


Chapter 5

106

66. Fenwick E, O’Brien BJ, Briggs A: Cost-effectiveness acceptability curves - facts, fallacies and frequently asked questions. Health Econ 2004, 13:405–415.

67. Hoch JS, Rockx MA, Krahn AD: Using the net benefit regression framework to construct cost-effectiveness acceptability curves: an example using data from a trial of external loop recorders versus Holter monitoring for ambulatory monitoring of “community acquired” syncope. BMC Health Serv Res 2006, 6:68.

6Effectiveness of supported self-help for recurrent

depression: a randomised controlled trial in primary care

Karolien E.M. Biesheuvel-Leliefeld Sandra Dijkstra-Kersten

Digna J.F. van SchaikHarm W.J. van Marwijk

Filip SmitHenriette E. van de Horst

Claudi L.H. Bockting

Revision under review in: BMC Psychiatry

R1R2R3R4R5R6R7R8R9


Chapter 6

108

ABSTRACT

Purpose: The burden of depression is high, mostly due to its recurrent nature. In this study we evaluated the effectiveness of a supported Self-help Preventive Cognitive Therapy (S-PCT) aimed at the prevention of recurrence, compared to treatment-as-usual (TAU) in primary care.

Methods: We conducted a randomised controlled trial among 248 patients with a history of depression, currently in remission. Participants were randomised to TAU augmented with S-PCT (n=124) or TAU alone (n=124). S-PCT consisted of a self-help intervention, supported by weekly telephone guidance by a counsellor. Primary outcome was incidence of recurrence over the full 12 months follow-up. Secondary outcomes were depressive symptoms, quality of life (EQ-5D and SF-12), co-morbid psychopathology, and self-efficacy.

Results: In the S-PCT group, 44 participants (35.5%) experienced a recurrence, compared to 62 participants (50.0%) in the TAU group (incidence rate ratio=0.71, 95%CI 0.52 to 0.97; risk difference=14, 95%CI 2-24, number needed to treat=8). Compared to the TAU-group, the S-PCT group showed a significant reduction in depressive symptoms over 12 months (mean difference -2.18; 95%CI -3.09 to -1.27) and a significant increase in quality of life (EQ-5D) (mean difference 0.04; 95%CI 0.004 to 0.08). S-PCT had no effect on co-morbid psychopathology, self-efficacy, and quality of life based on the SF-12.

Conclusions: A supported self-help preventive cognitive therapy in primary care proved to be effective in reducing the burden of recurrent depression.


Effectiveness of supported self-help for recurrent depression

109

6

INTRODUCTION

Major depressive disorder (MDD) is a prevalent mental disorder and is associated with a high risk of relapse and recurrence1. MDD is frequently associated with incomplete remission between episodes2–4 and is considered to be among the most disabling illnesses5, negatively affecting many aspects of life6–8. Current guidelines recommend continuation of antidepressant medication (ADM) and/or psychological treatment (e.g. Cognitive (behavioural) Therapy (CT)) to reduce the risk of relapse and recurrence9,10. The most commonly used strategy is continuation of ADM11–14. Yet, the recommendations on ADM are under debate as the optimal duration of the continuation- or maintenance phase has not been studied well enough11,14. Also, there is conflicting evidence about the effect of discontinuation of ADM on relapse or recurrence15 and, furthermore, reported levels of ADM non-adherence have been consistently high16. In conclusion, proactive management based on continuation of ADM alone may not be the most optimal strategy in preventing relapse or recurrence.

Research demonstrates that psychological interventions, specifically aimed at the prevention of relapse and recurrence in patients with a history of depression, offered during the continuation- or maintenance phase, are effective in reducing the risk of relapse and recurrence compared to TAU and/or ADM17–20. These interventions are mostly based on C(B)T21, but add strategies such as modifying dysfunctional meta-cognitions in preventive CT (PCT)22, meditation in Mindfulness Based CT (MBCT)23 and linking stressful life events and insufficient social support to relapse and recurrence in Interpersonal Therapy (IPT)24. The majority of these interventions is offered in secondary care, often relying on intensive use of therapist’s time, and, therefore, are costly. A minimally supported self-help may help to overcome this problem and has already proved as effective as face-to-face treatments in acute depressed patients25. The integration of a supported self-help in primary care, supported by para-professionals25, into current longitudinal primary care systems, would fit best with the recurrent character of depression. In the Netherlands, as in most western countries, primary care professionals have regular contact with the vast majority of the population, learn about the patients’ social situation and provide continuous care26. Besides, the prevalence of patients with MDD or depressive feelings in primary practice is around 21%27. Therefore, in this study, we conducted a randomised controlled trial (RCT) to evaluate the real-life effectiveness of a supported self-help PCT (S-PCT) in primary care in patients with a history of depression, currently in remission.

R1R2R3R4R5R6R7R8R9


Chapter 6

110

METHODS

DesignWe performed a pragmatic randomised controlled trial with two parallel groups of participants comparing TAU augmented with S-PCT, with TAU alone. The design of this study is described in more detail elsewhere28. The study was called the PARADE-study (Prevention of Recurrent DEpression). The study is registered in the Dutch Trial Register, www.trialregister.nl, NTR3001.

EthicsThe Medical Ethics Committee of the VU University Medical Center Amsterdam approved the study protocol and all participants provided written informed consent.

TerminologyTo describe the course of depression, we use the operational criteria of Frank et al 29. According to these criteria, the course of depression is described as a series of disease stages in which a patient can move from a symptom-free stage, to a stage characterized by some symptoms but not meeting the diagnostic criteria, to a stage with the full-blown disorder, after which the patient can go into remission. When a patient stays in remission for a minimum of six months, he or she is considered to be recovered. Subsequently, a relapse is defined as a depressive episode that occurs during remission and before recovery, while a recurrence is defined as a depressive episode that occurs after recovery.

ParticipantsParticipants were recruited through general practices and mental health care services in the Netherlands. To be included in the trial, participants had to a) be 18 years or older, b) be in full or partial remission (meaning the presence of residual symptoms) of recurrent MDD for at least two months, but no longer in recovery than five years according to the Structured Clinical Interview for DSM-IV Axis 1 disorders (SCID-1 3.0) 30 and c) have experienced two or more previous episodes of MDD. The SCID-I interview was conducted over the telephone by trained researchers and psychologists. Exclusion criteria were severe cognitive impairment, current or past mania, hypomania or psychosis, current alcohol or drug abuse, or insufficient mastery of the Dutch language.

CounsellorsTwenty-four counsellors (primary care mental health nurses and psychologists) were trained to guide the intervention. The psychologists were non specialised psychologists (i.e. without postdoctoral training in clinical interventions). All counsellors attended a one-day



111

6

training delivered by experienced clinical psychologists, who developed the intervention and therefore had an intimate knowledge of S-PCT. Before the start of the trial, the trainers evaluated the competence of the counsellors by giving feedback on audiotaped telephone contacts with two pilot-patients for each counsellor during a one-day supervision session. During the trial, counsellors could contact the trainers at any time for additional questions and feedback. To assess adherence, each week, the counsellor completed a checklist with 4 items; (1) the number of that week’s module (1-8), (2) the compliance of the participants in reading the literature of that week (yes/no plus reason), (3) the compliance of the participants in doing the assignments (yes/no plus reason) and (4) the time spent on the call (minutes).

Intervention The intervention is a supported self-help, and is a manualised PCT-based bibliotherapy consisting of a printed self-help book with eight modules and minimal guidance 31. It is based on an effective face-to-face PCT 22,32 and mobile PCT33. PCT is an adapted type of cognitive therapy for acute depression21 and aims to prevent relapse and recurrence in remitted patients with a history of depressive episodes. The intervention prevention program targets underlying cognitive vulnerability factors, such as dysfunctional beliefs. Unlike CT for acutely depressed patients, S-PCT is not primarily directed toward modifying negative thoughts. Instead, it starts with the identification of negative thoughts and dysfunctional attitudes, aided by a self-report questionnaire with examples of attitudes and specific challenging techniques. The focus of the self-help book is then directed on changing these attitudes by using different cognitive techniques such as identification of positive attitudes. Moreover, practice in daily life with alternative attitudes is promoted. Part of the modules is keeping a diary of positive experiences in order to enhance specific memories of positive experiences, instead of retaining overly general memories. Further specific relapse and recurrence prevention strategies are formulated in the last modules of the S-PCT resulting in a personal prevention plan. Like regular CT, PCT follows a fixed structure with agenda setting, review of homework, explanation of the rationale of each session, and the assignment of homework. Participants complete one module per week. Each module includes reading homework plus assignments, to be completed in approximately 60 minutes. In the current project the counsellor explained the rationale of PCT and coming week’s planning in a first contact (by phone or face-to-face), prior to the start of the intervention. Each week, the counsellor contacted the participant by phone to evaluate progress and understanding. This call was strictly protocolled and was designed to last no longer than 15 minutes. The nature of the contact was solely to support the participant and not to actively engage in a therapeutic relationship.

R1R2R3R4R5R6R7R8R9


Chapter 6

112

Treatment-as-usualThere were no restrictions to type of TAU. Care providers were not aware of randomization status.

Current TAU guidelines recommend to encourage a person who has benefited from taking ADM, to continue ADM for at least 6 months after remission of an episode of depression. With respect to psychological interventions, guidelines recommend to offer CBT to persons with a significant history of depression plus residual symptoms, and MBCT to patients with a history of at least three episodes of depression9,34. TAU was recorded using the Trimbos and iMTA self-report questionnaire for costs associated with Psychiatric Illnesses (TiC-P)35.

Treatment allocationAfter inclusion, participants were randomised using computer generated blocks and allocation concealment, stratified by the number of previous depressive episodes (2-3 episodes versus ≥4 episodes) because the number of previous episodes is associated with relapse and recurrence 36.

BlindingInterviewers were blind for randomization status of the participants during all measurements. Due to the nature of the intervention, it was not possible to blind the participants. At the start of each interview, participants were asked not to reveal their allocation status to the interviewers.

Outcome measures Primary outcomePrimary outcome was the incidence rate of relapse or recurrence of depression over the 12 months follow-up period. To reduce recall bias, telephone SCID-interviews were conducted over 6 months, at 6 and 12 months and combined into a single outcome (0=no relapse or recurrence, 1=relapse or recurrence). The incidence rate ratio (IRR) was calculated by comparing the incidence rates of new episodes in both conditions. An IRR < 1 implies a better risk reduction in the intervention group relative to the control group; the intervention is then deemed successful. An IRR = 1 and IRR > 1 imply no effect or an adverse effect, respectively.

Secondary outcomesSecondary outcomes were assessed online at baseline and after 6 and 12 months (depressive symptomatology, health related quality of life,) or after 9 and 12 months (comorbid psychopathology, self-efficacy).



113

6

Depressive symptoms were assessed using the Dutch translation of the Quick Inventory of Depressive Symptomatology Self Report (QIDS-sr) 37. This self-report questionnaire consists of 16 symptom items to be answered on a 4-point Likert-scale. A score of 0-5 is categorised as no depressive symptoms, 6-10 as mild, 11-15 as moderate, 16-20 as severe, and 20-27 as very severe depressive symptoms.

Health related quality of life (HRQoL) was examined using the Dutch translations of the 12-Item Short-Form Health Survey (SF-12) 38 and the European Quality of Life Five Dimensions (three level) health status questionnaire (EQ-5D-3L)39. The SF-12 is a measure of health-related functional status 40 and yields two summary measures of physical and mental health. It is the most commonly used health measure and, therefore, outcomes can be easily compared to other studies using the SF-12. The EQ-5D measures HRQoL on five dimensions (mobility, self-care, usual activities, pain/discomfort and anxiety and depression), combined into one outcome. Each dimension is rated at three levels corresponding to whether a respondent has no problems, moderate or extreme problems. The value of each of the 243 health states is preference weighted using valuations from the Dutch population41. Besides the SF-12, we used the EQ-5D because it is the most commonly used health measure in a cost-effectiveness analysis, which we plan to report.

Comorbid symptoms were measured with the Four Dimensional Symptom Questionnaire (4DSQ)42. The 4DSQ is a self-rating questionnaire that comprises 50 items distributed over four scales (distress, depression, anxiety, and somatisation)

Perceived self-efficacy was assessed with the General Self Efficacy Scale (GSES) 43. The GSES consists of 10 items, scored 1–4. Especially in the case of self-help, self-efficacy might change in the course of the intervention and during follow-up.

Sample sizeWe combined findings from previous research18,44, and assumed a mean relapse or recurrence rate of 40% after 1 year of follow-up versus 60% in the controls. To detect this 20% risk-reduction in a 2-sided test at α= 0.05 and a power of 1-β= 0.80, 107 participants in each condition were required. Compensating for loss to follow-up of 10% over the whole 12 months follow-up, required at least (107/0.90=) 119 participants at baseline in each trial arm. Our own experience with randomization of patients at general practice level45,46 indicates that clustering of patients within practices has no impact on the power of the trial. Therefore, we did not take clustering effects into account.

Statistical analysesWe investigated whether baseline characteristics differed between conditions. In addition, we compared the baseline characteristics of dropouts and those who completed all measurements during 12-month follow-up by performing logistic regression analysis.

R1R2R3R4R5R6R7R8R9


Chapter 6

114

Data were primarily analysed on the basis of the intention-to-treat (ITT) principle. Missing values on outcome measures were imputed using multiple (10-fold) imputation by chained equations (MICE)47. The analyses were performed in each of the 10 datasets, and the results of the analyses were pooled using the Rubin rules48.

To compare risk on relapse or recurrence in both conditions, we performed a Poisson regression analysis of the incidence of relapse or recurrence on the treatment condition. In this manner we obtained an incidence rate ratio (IRR). Because the use of Poisson regression tends to provide conservative results49–51 and overestimates error50, we used the Hubert-White sandwich estimator as implemented in STATA.52 Results were adjusted for baseline (residual) depressive symptoms (QIDS-sr) because these symptoms are well-known to be a risk factor for relapse or recurrence 53–55.

Estimates of the intervention effects on the secondary outcome measures (all continuous) were obtained from linear mixed models (LMM). Randomization status, R, time of measurements, T, and randomization-by-time interaction (RxT) were included as fixed effects in the models. The participants’ identificator, ID, was included as random term, because in the long dataset the same participant could have contributed to the dataset at some or all time points. We assessed the overall effect of the intervention by testing the interaction between randomization and time of measurement that was associated with outcome. Means were adjusted for baseline level of the outcome. In LMM, imputation of missing data is not necessary. The results of the ITT analysis were compared with the results of the per protocol (PP) analysis, including those participants who completed at least 80% of the intervention (5 modules). All analyses were performed with STATA (version 12).

RESULTS

Participants flow and recruitmentDetails of enrolment are shown in Figure 1. Recruitment took place between September 2012 and April 2014. Medical records of 22 family practices and 4 specialised mental health care institutions were screened for eligible patients. This led to the selection of 5.489 patients, who received a short information letter. Finally, 248 patients met all inclusion-criteria and signed informed consent. They were randomly allocated to the S-PCT group (124) or to the TAU group (124). Baseline characteristicsIn Table 1, baseline socio-demographic and clinical characteristics of the ITT group are presented. No relevant baseline imbalances were found. At baseline, all participants were in (partial) remission of recurrent MDD and experienced mild depressive symptoms (mean QIDS-sr=9.2, SD 4.8).



115

6

Analysed intention to treat (n=124) Analysed per protocol (n=101) Analysed complete follow-up (n=95)

♦ Completed PCT + TAU (n=101/117) ♦ Discontinued intervention (n=16) Reasons Too much time (4) Therapy is too emotional (2) Too much therapy already (1) Lack of motivation (2) Too difficult (3) No restrictive rule (2) Other (2) ♦ Completed 12 month FU (n = 95/124) - Completed questionnaires (n = 98/124) - Completed interviews (n = 104/124) ♦ Lost to follow-up (n= 29) Reasons - Unknown (6) - Logistical reason (3) - Lack of motivation (4) - Stopped with intervention also (5) - Too stressful due to personal circumstances (1) - Lack of time/ no priority (2) - Acute depression? (2) - Other (3)

Allocated to PCT+TAU (n=124) ♦ Received PCT +TAU (n=117) ♦ Did not receive PCT+TAU (n=7) Reasons: Too much depressed (2) Takes too much time (1) Therapy is too emotional (2) Therapist advises to cancel (1) Unknown (1)

♦ Completed 12 month FU (n= 93/124) - Completed questionnaires (n = 95/124) - Completed interviews (n = 106/124) ♦ Lost to follow-up (n= 31/124) Reasons - Unknown (8) - Logistical reason (4) - Too stressful to fill in questionnaires (2) - Acute depression? (4) - Too stressful due to personal circumstances (4) - Unable to contact (4) - Lack of motivation (3) - Other (2)

♦ Allocated to TAU (n=124) ♦ Received TAU (n=124) Withdrew in TAU group immediately after start of study (6)

Analysed intention to treat (n=124) Analysed per protocol (n=124 ) Analysed complete follow-up (n=93)

Allocation

Analysis

12 months Follow-Up

Randomized (n=248)

Enrollment

Assessed for eligibility (n=284)

Excluded (n=36) ♦ Declined to participate (n=2) ♦ Not meeting inclusion criteria (n=34) Current/chronic depression (9) Last episode > 5 years (8) No recurrent depression (9) No or one depression (4)

Burn-out only (3) Anxiety only (2)

Current substance abuse (1) Bipolar disorder (4) Psychosis (3)

Figure 1. Participant flow diagram

R1R2R3R4R5R6R7R8R9


Chapter 6

116

Table 1. Baseline demographic and descriptive characteristics of the study population according to randomised group*

Characteristics S-PCT(n=124)

TAU(n=124)

All participants(n=248)

Age, mean (SD) 48.6 (11.9) 48.8 (11.4) 48.7 (11.7)Females, n (%) 89 (71.8%) 84 (67.7%) 173 (69.8%)No previous episodes, %

2 or 3 53.2% 49.9% 51.6%4 or more 46.8% 50.1% 48.4%

Marital status, % Partner 64.9% 64.9% 64.9%Education**, % High education 42.7% 35.5% 39.1%Age of onset, mean (SD) 28.2 (11.4) 27.5 (12.3) 27.8 (11.9)Depressive symptoms (QIDS-sr), mean (SD) 9.6 (4.8) 8.9 (5.0) 9.3 (4.9)Quality of life

Mental health (SF12, mean (SD) 53.6 (12.2) 53.5 (11.6) 53.5 (11.9)Physical health (SF12, mean (SD) 59.4 (11.4) 57.6 (11.7) 58.5 (11.6)

EQ-5D, mean (SD) 0.77 (0.21) 0.78 (0.20) 0.77 (0.2)Comorbid psychopathology (4DSQ)

Anxiety, mean (SD) 3.2 (3.9) 3.2 (4.3) 3.2 (4.1)Distress, mean (SD) 13.0 (7.6) 12.7 (8.0) 12.8 (7.8)

Somatisation, mean (SD) 8.1 (5.5) 8.9 (5.7) 8.5 (5.6)Pain (MPQ), mean (SD) 2.5 (3.6) 3.2 (4.2) 2.8 (3.9)Fatigue (FSS), mean (SD) 3.8 (1.5) 3.9 (1.6) 3.8 (1.6)Self-efficacy (GSES), mean (SD) 28.6 (5.9) 28.3 (6.2) 28.4 (6.0)ADM use past 3 months, % 51.8% 56.7% 54.2%

Abbreviations: ADM, anti-depressant medication; EQ, EuroQol; FSS, Fatigue Severity Scale; GSES, General Self Efficacy Scale; MPQ, MacGill Pain Questionnaire; QIDS-sr, Quick Inventory of Depressive Symptoms self-report; SD, standard deviation; SF, short form 12-Item Short Form Health Survey; S-PCT, supported self-help preventive cognitive therapy; TAU, treatment-as-usual; 4DSQ, Four Dimensional Symptom Questionnaire *Standard deviations for multiply-imputed data were computed from the standard errors:(sd= sqrt( _b[var2] - _b[var]*_b[var])) **Education is defined as bachelor’s or master’s degree

Numbers analysedComplete follow-up data were collected from 95/124 participants (77%) in the intervention group and 93/124 participants (75%) in the control group, which was not statistically different (χ2= 0.088, df=247). Loss to follow up was significantly associated with more fatigue at baseline (difference in mean=0.602, 95%CI 0.115 to 1.089).

Primary outcomeIncidence of relapse or recurrence of depressionTwelve months after randomization, a new relapse or recurrence of depression had occurred in 44 (35.5%) participants in the intervention group and 62 (50.0%) participants in the control group (IRR=0.71, 95%CI 0.52 to 0.97). The risk-difference (RD) between the TAU-group and the S-PCT group was 14% (95%CI 2-24) which corresponds to a number-needed-to treat (NNT) of 8 (Table 2).



117

6

Table 2. Primary outcome and secondary outcomes (ITT analysis)

S-PCT TAU IRR*(95% CI)

RD**

(95% CI)NNT

Primary outcome

Relapse or recurrence

after12 months

44/124 (35,5%)

62/124 (50,0%)

0.71 (0.52;0.97)

14% (2;24)

8

S-PCTmean (SD)

TAUmean (SD)

mean difference

(95% CI) ***

Z-value P

Secondary outcomes

Depressive symptoms(QIDS-sr)

Baseline6 months12 months

9.6 (4.8) 6.3 (4.3)7.2 (4.9)

8.9 (5.0) 8.7 (4.9)7.7 (5.3)

-2.18(-3.09;-1.27)

-4.70 <0.001

Health related quality of life (SF-12, mental)


53.6 (12.2) 53.3 (10.7)53.4 (10.4)

53.5 (11.6) 51.7 (11.8)54.4 (12.2)

0.67(-1.33;2.67)

0.65 0.513

Health related quality of life (SF-12, physical)


59.4 (11.4)58.7 (10.8)60.5 (11.5)

57.6 (11.7) 56.8 (11.5)58.8 (12.6)

1.05(-0.81;2.91)

1.10 0.270

Health related quality of life (EQ-5D)


0.77 (0.21) 0.81 (0.19)0.80 (0.19)

0.78 (0.20) 0.77 (0.20)0.78 (0.24)

0.04(0.004;0.08)

2.18 0.029

Anxiety (4-DSQ)


3.2 (3.9) 3.0 (3.7)2.8 (3.7)

3.2 (4.3) 2.6 (4.1)2.9 (4.2)

-0.05(-0.68;0.59)

-0.14 0.887

Distress (4-DSQ)


13.0 (7.6) 12.3 (8.1)11.9 (8.8)

12.7 (8.0) 11.5 (8.8)11.6 (8.7)

-0.21(-1.81;1.39)

-0.26 0.798

Somatisation (4-DSQ)


8.1 (5.5) 8.2 (5.8)7.6 (5.4)

8.9 (5.7) 8.4 (6.0)7.6 (5.4)

0.38(-0.64;1.39)

0.73 0.464

Self-efficacy (GSES)


28.6 (5.9) 28.8 (6.8)28.6 (7.1)

28.3 (6.2) 29.3 (6.2)28.8 (7.0)

-0.68(-1.91;0.55)

-1.08 0.280

Abbreviations: CI, Confidence Interval; EQ-5D, Five Dimensional EuroQOl; GSES, General Self Efficacy Scale; ITT, Intention-to-treat analysis; IRR, Incidence rate ratio; NNT, number needed to treat; QIDS-sr, Quick Inventory of depressive Symptoms self-report; RD, Risk Difference; S-PCT, supported self-help preventive cognitive therapy; 4-DSQ, Four Dimensional Symptom Questionnaire; SF-12, 12-Item Short Form Health Survey * P=0.032; an IRR<1 means that over 12 months more patients in the TAU group recurred compared to the S-PCT group; scores were adjusted for depressive symptoms at baseline **P=0.025; RD is the percentage risk difference in recurrence rate between S-PCT and TAU over the 12 months follow-up period *** Scores were adjusted for baseline level of the outcome and estimated with linear mixed modeling

R1R2R3R4R5R6R7R8R9


Chapter 6

118

Secondary outcomesDepressive symptom scores in the intervention group decreased significantly compared to TAU over 12 months (-2.18 QIDS-sr points; 95%CI-3.09 to -1.27). Quality of life (EQ-5D) improved significantly (0.04 EQ-5D points; 95%CI 0.004 to 0.08) but not on the SF-12. No significant effects were found on any of the other secondary outcomes (Table 2).

Adherence to the interventionSeven participants did not start the supported self-help. Two participants dropped out after the first contact, and 5 participants dropped out after the first S-PCT meeting. Reasons for drop-out are shown in Figure 1. From the 117 participants who started the intervention, 16 (18.5%) dropped out during the intervention, all before week 6 of the intervention. In total, 101 participants (81%) completed at least 5 modules (80%), and were labelled as “completers”. At baseline, completers experienced more depressive symptoms (mean difference =2.22, 95%CI 0.010 to 4.35), more distress (mean difference =3.46, 95%CI 0.02 to 6.90) and a lower quality of life (EQ5D) (mean difference=-0.11, 95%CI 0.02 to 0.20) than non-completers.

The self-help was led by a primary care mental health nurse (31.5%) or by a non-specialised psychologist (68.5%). The first contact was organised face-to-face (40.3%) or by telephone (59.1%). The mean amount of time spent per phone-call per participant by the counsellor was 13.8 minutes (sd=5.42), totalling a mean of 110.2 minutes of attention per participant per treatment. According to the checklist of the counsellors, in 6% of all contacts the participant had not read the literature belonging to that week’s module. Reasons for not reading literature were (more than one reason per participant was possible): lack of time (19), too difficult (10), practical considerations (7), too depressed (6), did not feel like it (1), other (3). In 11% of all contacts the participants declared they did not complete the assignments for that week’s module. Reasons for not doing assignments were: lack of time (28), too difficult (21), too depressed (11), practical considerations (11), did not feel like (5), physical illness (5), intervention does not meet expectations (2), other (2).

Treatment as usualMedication use (all types of medication including ADM) data for all 12 months were available for 74% of the participants (92/124) in the S-PCT group and 74% of the participants (92/124) in the TAU group. During this period, 62% (57/92) of the participants in the S-PCT group received medication at any 3-month measurement versus 63% (58/92) participants in the TAU group (χ2=0.23,df=1, P=0.8). Data on mental health care use were available for 85% of the participants (105/124) in the S-PCT group and 81% of the participants (100/124) in the TAU group. In the S-PCT group 43% received additional counselling from a psychiatrist/psychologist/psychotherapist versus 40% of the participants in the TAU group (χ2=0.172, df=1, P=0.678).



119

6

Per protocol analysisThe per protocol (PP) analysis included only those participants who completed at least 80% (5 modules) of the intervention (81%; 101/124 participants). The results were roughly similar. The difference in incidence rate of relapse or recurrence between the S-PCT group and the TAU group was more pronounced than in the ITT analysis (incidence rate ratio=0.68, 95%CI 0.50 to 0.93, risk difference=15%, 95%CI 4 to 25, NNT=7) (Table 3). Similar to the ITT analysis, both the depressive symptoms score and quality of life score (EQ5D) changed significantly over 12 months in the intervention group compared to the TAU group (-2.31 QIDS-sr points; 95%CI-3.26 to -1.37 and 0.04; 95%CI 0.003 to 0.81, respectively).

Table 3. Changes in primary and secondary outcome measures (PP-analysis) IRR*

(95% CI)RD**

(95%CI)NNT

Primary outcome

Relapse or recurrence 0.68(0.50-0.93)

15%(4;25)

7

mean difference***(95% CI)

Z-value P

Secondary outcomes

Depressive symptoms(QIDS-sr)

-2.31(-3.26;-1.37)

-4.81 <0.001

Health relatedquality of life (SF-12, mental)

0.44(-1.62;2.50)

0.42 0.675

Health relatedquality of life (SF-12, physical)

0.89(-1.01;2.80)

0.92 0.359

Health related quality of life (EQ-5D)

0.04(0.003;0.81)

2.10 0.036

Anxiety(4-DSQ)

-0.05(-0.71;0.60)

-0.16 0.872

Distress(4DSQ)

-0.25(-1.90;1.41)

-0.29 0.769

Somatisation(4-DSQ)

0.42(-0.63;1.48)

0.79 0.432

Self-efficacy(GSES)

-0.57(-1.81;0.67)

-0.91 0.36

Abbreviations: CI, Confidence Interval; EQ-5D, Five Dimensional EuroQOl; GSES, General Self Efficacy Scale; ITT, Intention-to-treat analysis; IRR, Incidence rate ratio; NNT, number needed to treat; QIDS-sr, Quick Inventory of depressive Symptoms self-report; RD, Risk Difference; S-PCT, supported self-help preventive cognitive therapy; 4-DSQ, Four Dimensional Symptom Questionnaire; SF-12, 12-Item Short Form Health Survey *P=0.017; scores were adjusted for depressive symptoms at baseline **P= 0.011; RD is the percentage risk difference in recurrence rate between S-PCT and TAU over the 12 months follow-up period***Scores were adjusted for baseline level of the outcome and estimated with linear mixed modeling

R1R2R3R4R5R6R7R8R9


Chapter 6

120

DISCUSSION

Main findingsIn this study, we evaluated the real-life effectiveness of a supported self-help preventive Cognitive Therapy (S-PCT) in primary care, in remitted patients with a history of depression. Our analyses showed that S-PCT statistically significantly reduced relapse and recurrence over 12 months with 14% (95%CI 2 to 24) compared to usual care.

Public health significanceMajor depressive disorder (MDD) is a prevalent mental disorder and is associated with a high risk of relapse and recurrence1. MDD is frequently associated with incomplete remission between episodes2–4 and is considered to be among the most disabling illnesses5, negatively affecting many aspects of life6–8. Largely due to this recurrent nature, the economic consequences of MDD are substantial56–59. Due to budget- and time restraints, health care systems have difficulties addressing the demand for care and treatment of people experiencing full-blown mental health problems and can only avert a fraction of the total disease burden that is attributable to mental disorders60,61. Our prevention trial showed that no less than 50% of the participants who received usual care recurred within 12 months. This emphasizes the vulnerability of this patient group and underlines that a preventive, low-cost, accessible minimal intervention, offered in primary care is much needed.

Our randomised controlled trial is one of the first that supplies evidence that such an endeavour might be successful. Our study shows that the risk of relapse or recurrence decreases significantly by 14% (NNT=8) when targeting remitted patients with a history of depression. Yet, the incidence rate ratio of 0.71 that we found, was somewhat higher (i.e. less effective) than the IRR we found in our meta-analysis comparing psychological interventions to usual care (0.64)20. The most likely explanation for this is our short follow-up period of 12 months (52 weeks) in our trial, while the mean follow-up in the meta-analysis was 115 weeks. Time to recurrence might exceed 12 months 62–66, possibly implying that we have missed the interventions’ and TAUs’ impact on later recurrences. Also, on average, participants in our trial experienced a higher level of residual symptoms than participants in the studies that were included in the meta-analysis. Therefore, the a-priori chance of relapse or recurrence might have been higher in our trial. Besides, we chose to offer a bibliotherapeutical self-help intervention with minimal guidance by a counsellor. Though economically attractive, the format of this intervention may have been too light. A possible way to improve outcome is to differentiate between high risk- and ultra-high risk patient-groups 67; in ultra-high risk groups (including patients with multiple previous episodes and higher levels of residual symptoms (as in our sample) a specific psychological intervention might be indicated such as face-to-face PCT and MBCT as several studies indicate 67, while



121

6

for a relatively lower risk group (including patients with a single previous episode and lower levels of residual symptoms) a less specific minimal intervention such as self-help might be optimal. Finally, S-PCT should be offered at the right moment to gain maximum effects. In our trial, the intervention was offered to participants at a random moment during remission or recovery. However, our meta-analysis 20 shows that the preventive effect of a psychological intervention is usually higher when the prevention directly follows the acute phase.

To end with, the statistically significant increase in quality of life in favour of the S-PCT group, found with the EQ-5D, was low (0.004, 95%CI 0.004 to 0.08). Research has shown that the smallest change in utility scores for the EQ-5D that can be regarded as important is 0.07468. Besides, the SF-12 showed no statistically significant change in quality of life. This implies that the improvement of quality of life, measured with the EQ-5D, should be interpreted with caution.

ImplementationThe addition of S-PCT seems a strategy that can be relatively easily implemented into current longitudinal primary care systems. Paraprofessionals in primary care could act as case managers in a model for continued care for recurrent depression. Still, referral to a paraprofessional in primary care requires a careful and timely coordination between health care professionals.

Strengths and limitationsOur operationalization of depression and relapse or recurrence was based on a structured clinical interview (SCID-1). A further strength of this study is that our participants achieved remission on antidepressants, other psychotherapies, psychiatric help, counselling, or no treatment at all, as typically present in clinical practice. Moreover, there were no restrictions in using medication at entry to the study. Therefore, this study was designed to maximise external validity, which suggests good generalizability of the findings.

Our study also has limitations. First, as is common in studies on psychological interventions, it was not possible to blind participants to the condition to which they were assigned. Second, effects were considered over a time-span of one year. A longer follow-up is preferred because it is important to know whether the positive clinical effects will sustain over time. Previous results on longer term effects of PCT are promising; brief CT, started after remission from a depressive episode on diverse types of treatment in patients with multiple prior episodes, has long-term preventive effects for at least 5.5 to 10 years 32,69.

R1R2R3R4R5R6R7R8R9


Chapter 6

122

Conclusions and further researchThis is the first study to evaluate the effects of a supported self-help preventive cognitive therapy (S-PCT) compared to usual care in primary care, over 12 months in remitted patients with a history of depression. We found a significant difference in relapse or recurrence rate between the two groups, in favour of S-PCT. Also, the level of depressive symptoms and quality of life changed significantly in favour of the S-PCT group.

Suggestions for further research include an evaluation of S-PCT with a longer follow-up. Another research questions is how to improve relapse and recurrence prevention strategies by finding out what works for whom and what type of guidance (e.g. frequency, intensity) works best. Also, it would be valuable to make a comparison of S-PCT with mobile, internet based PCT33 and face-to-face PCT69.

AcknowledgementsWe are very grateful to all participants. We would also like to thank all recruitment sites for their efforts: GGZ NHN, GGZ Amstelmere, GGZ Zuiderpoort, GGZ Rivierduinen, de Bosgroep, and the participating general practitioners. We also thank all counsellors for their guidance. Finally, we are grateful to Evelien van Valen for her help.



123

6

REFERENCES


2. Nierenberg AA, Husain MM, Trivedi MH, et al. Residual symptoms after remission of major depressive disorder with citalopram and risk of relapse: a STAR*D report. Psychol.Med 2010;40 (1):41-50.

3. Fava M, Graves LM, Benazzi F, et al. A cross-sectional study of the prevalence of cognitive and physical symptoms during long-term antidepressant treatment. The Journal of clinical psychiatry 2006;67(11):1754-9.

4. Conradi HJ, Ormel J, de Jonge P. Presence of individual (residual) symptoms during depressive episodes and periods of remission: a 3-year prospective study. Psychological medicine 2011;41(6):1165-74.

5. Ustun TB, Ayuso-Mateos JL, Chatterji S, Mathers C, Murray CJ. Global burden of depressive disorders in the year 2000. Br.J Psychiatry 2004;184

6. Ormel J, VonKorff M, Ustun TB, Pini S, Korten A, Oldehinkel T. Common mental disorders and disability across cultures. Results from the WHO Collaborative Study on Psychological Problems in General Health Care. JAMA 1994;272(22):1741-8.

7. Ormel J, Oldehinkel AJ, Nolen WA, Vollebergh W. Psychosocial disability before, during, and after a major depressive episode: a 3-wave population-based study of state, scar, and trait effects. Archives of general psychiatry 2004;61(4):387-92.

8. Wells KB, Stewart A, Hays RD, et al. The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA 1989;262(7):914-9.

9. NICE. National Institute for Health and Clinical Excellence. In: National Institute for Health and Clinical Excellence. Depression: The Treatment and Management of Depression in Adults (update). Clinical Guidelines: National Institute for Health and Clinical Excellence. Depression: the treatment and management of depression in adults (update).; 2009.

10. APA. Practice guideline for the treatment of patients with major depressive disorder (revision). American Journal of Psychiatry 2000;157.

11. Geddes JR, Carney SM, Davies C, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet 2003;361.

12. Glue P, Donovan MR, Kolluri S, Emir B. Meta-analysis of relapse prevention antidepressant trials in depressive disorders. The Australian and New Zealand journal of psychiatry 2010;44(8):697-705.

13. Hansen R, Gaynes B, Thieda P, et al. Meta-analysis of major depressive disorder relapse and recurrence with second-generation antidepressants. Psychiatric services (Washington, D.C.) 2008;59(10):1121-30.

14. Kaymaz N, van OJ, Loonen AJ, Nolen WA. Evidence that patients with single versus recurrent depressive episodes are differentially sensitive to treatment discontinuation: a meta-analysis of placebo-controlled randomized trials. Journal of Clinical Psychiatry 2008;69

15. Bockting CL, ten Doesschate MC, Spijker J, Spinhoven P, Koeter MW, Schene AH. Continuation and maintenance use of antidepressants in recurrent depression. Psychotherapy and Psychosomatics 2008;77.

16. Ten Doesschate MC, Bockting CL, Schene AH. Adherence to continuation and maintenance antidepressant use in recurrent depression. J.Affect.Disord. 2009;115.

17. Guidi J, Fava GA, Fava M, Papakostas GI. Efficacy of the sequential integration of psychotherapy and pharmacotherapy in major depressive disorder: a preliminary meta-analysis. Psychol.Med. 2011;41 (2):321-331.

18. Vittengl JR, Clark LA, Dunn TW, Jarrett RB. Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitive-behavioral therapy’s effects. J.Consult Clin.Psychol. 2007;75.

19. Piet J, Hougaard E. The effect of mindfulness-based cognitive therapy for prevention of relapse in recurrent major depressive disorder: A systematic review and meta-analysis. Clinical Psychology Review 2011;31.

20. Biesheuvel-Leliefeld KEM, Kok GD, Bockting CLH, et al. Effectiveness of psychological interventions in preventing recurrence of depressive disorder: Meta-analysis and meta-regression. Journal of affective disorders 2015;174C:400-410.

R1R2R3R4R5R6R7R8R9


Chapter 6

124



23. Kuyken W, Watkins E, Holden E, et al. How does mindfulness-based cognitive therapy work?. Behav.Res Ther 2010;48 (11):1105-1112.

24. Spanier CA. The prevention of depression: Protective mechanisms of maintenance interpersonal psychotherapy. Dissertation Abstracts International: Section B: The Sciences and Engineering 1998;58 (10-B).

25. Cuijpers P, Donker T, van SA, Li J, Andersson G. Is guided self-help as effective as face-to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychological Medicine 2010;40.



28. Biesheuvel-Leliefeld KE, Kersten SM, van der Horst HE, et al. Cost-effectiveness of nurse-led self-help for recurrent depression in the primary care setting: design of a pragmatic randomised controlled trial. BMC Psychiatry 2012;12.

29. Frank E, Prien RF, Jarrett RB, et al. Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Archives of General Psychiatry 1991;48.

30. First MB, Gibbon M, Spitzer RL, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV) . Washington, D.C: American Psychiatric Press, Inc; 1996.



33. Kok G, Burger H, Riper H, et al. The Three-Month Effect of Mobile Internet-Based Cognitive Therapy on the Course of Depressive Symptoms in Remitted Recurrently Depressed Patients: Results of a Randomized Controlled Trial. Psychotherapy and psychosomatics 2015;84(2):90-99.

34. American Psychiatric Association. Practice Guidelines for the Treatment of Patients with Major Depressive Disorder, Third Edition.; 2010.

35. Hakkaart-van Roijen L, Van Straten A, Donker M. Manual:Trimbos/iMTA Questionnaire for Costs Associated with Psychiatric Illness (in Dutch). Rotterdam: Erasmus University; 2002.


37. Rush AJ, Trivedi MH, Ibrahim HM, et al. The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biological Psychiatry 2003;54.

38. Gandek B, Ware JE, Aaronson NK, et al. Cross-Validation of Item Selection and Scoring for the SF-12 Health Survey in Nine Countries. Journal of Clinical Epidemiology 1998;51(11):1171-1178.

39. Brooks R. EuroQol: the current state of play. Health Policy 1996;37(1):53-72.

40. Ware Jr. J, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Medical Care 1996;34.

41. Lamers LM, Stalmeier PF, McDonnell J, Krabbe PF, van Busschbach JJ. [Measuring the quality of life in economic evaluations: the Dutch EQ-5D tariff]. Ned Tijdschr Geneeskd. 2005;149.

42. Terluin B, van Marwijk HW, Ader HJ, et al. The Four-Dimensional Symptom Questionnaire (4DSQ): a validation study of a multidimensional self-report questionnaire to assess distress, depression, anxiety and somatization. BMC Psychiatry 2006;6.

43. Schwarzer, R., Jerusalem M. Measures in health psychology: A user’s portfolio. Causal and control beliefs. In: J. Weinman, S. Wright, Johnston M, ed. Windsor, England; 1995:35- 37.



125

6

44. Romijn G, Ruiter M, Smit F. Meer Effect Met Depressiepreventie? Strategieën Voor Publieksvoorlichting, Vroegherkenning En Terugvalpreventie. (Reeks: Wetenschappelijke Onderbouwing Depressiepreventie). : Trimbos-Instituut. Utrecht: Trimbos Instituut; 2008.

45. Willemse GR, Smit F, Cuijpers P, Tiemens BG. Minimal-contact psychotherapy for sub-threshold depression in primary care. Randomised trial. British Journal of Psychiatry 2004;185.

46. Bakker IM, Terluin B, van Marwijk HW, et al. A cluster-randomised trial evaluating an intervention for patients with stress-related mental disorders and sick leave in primary care. PLoS.Clin.Trials 2007;2.

47. Van Buuren S, Oudshoorn CGM. Multivariate Imputation by Chained Equations. Leiden; 2000. Available at: http://www.stefvanbuuren.nl/publications/MICE V1.0 Manual TNO00038 2000.pdf.

48. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York, NY US: John Wiley & Sons; 1987.

49. McNutt L-A, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. American journal of epidemiology 2003;157(10):940-3.

50. Zocchetti C, Consonni D, Bertazzi PA. Estimation of prevalence rate ratios from cross-sectional data. International journal of epidemiology 1995;24(5):1064-7.

51. Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occupational and environmental medicine 1998;55(4):272-7.

52. Royall RM. Model robust confidence intervals using maximum likelihood estimators. Int Stat Rev 1986;(54):221-226.

53. Paykel ES, Ramana R, Cooper Z, Hayhurst H, Kerr J, Barocka A. Residual symptoms after partial remission: an important outcome in depression. Psychological medicine 1995;25(6):1171-80.

54. Ormel J, Oldehinkel T, Brilman E, vanden Brink W. Outcome of depression and anxiety in primary care. A three-wave 3 1/2-year study of psychopathology and disability. Archives of general psychiatry 1993;50(10):759-66.

55. Judd LL, Paulus MJ, Schettler PJ, et al. Does incomplete recovery from first lifetime major depressive episode herald a chronic course of illness? The American journal of psychiatry 2000;157(9):1501-4.


57. Greenberg PE, Birnbaum HG. The economic burden of depression in the US: societal and patient perspectives. Expert.Opin.Pharmacother. 2005;6.


59. Vasiliadis HM, Dionne PA, Preville M, Gentil L, Berbiche D, Latimer E. The excess healthcare costs associated with depression and anxiety in elderly living in the community. Am J Geriatr.Psychiatry 2013;21.

60. Chisholm D, Sanderson K, Ayuso-Mateos JL, Saxena S. Reducing the global burden of depression: population-level analysis of intervention cost-effectiveness in 14 world regions. Br.J Psychiatry 2004;184.

61. Andrews G, Issakidis C, Sanderson K, Corry J, Lapsley H. Utilising survey data to inform public policy: comparison of the cost-effectiveness of treatment of ten mental disorders. British Journal of Psychiatry 2004;184.

62. Kanai T, Takeuchi H, Furukawa TA, et al. Time to recurrence after recovery from major depressive episodes and its predictors. Psychol.Med 2003;33 (5):839-845.

63. Eaton WW, Shao H, Nestadt G, et al. Population-based study of first onset and chronicity in major depressive disorder. Archives of general psychiatry 2008;65(5):513-20. doi:10.1001/archpsyc.65.5.513.

64. Van Londen L, Molenaar RP, Goekoop JG, Zwinderman AH, Rooijmans HG. Three- to 5-year prospective follow-up of outcome in major depression. Psychological medicine 1998;28(3):731-5.

65. Jarrett RB, Kraft D, Doyle J, Foster BM, Eaves GG, Silver PC. Preventing recurrent depression using cognitive therapy with and without a continuation phase: a randomized clinical trial. Arch Gen.Psychiatry 2001;58 (4):381-388.

66. Thase ME, Simons AD, McGeary J, et al. Relapse after cognitive behavior therapy of depression: potential implications for longer courses of treatment. Am.J.Psychiatry 1992;149.

R1R2R3R4R5R6R7R8R9


Chapter 6

126

67. Bockting C, Hollon SD, Jarrett RB, Kuyken W, Dobson K. A lifetime approach to major depressive disorder: The contributions of psychological interventions in preventing relapse and recurrence. Clinical psychology review 2015.

68. Walters SJ, Brazier JE. Comparison of the minimally important difference for two health state utility measures: EQ-5D and SF-6D. Quality of Life Research 2005;14.

69. Bockting CLH, Smid NH, Koeter MWJ, Spinhoven P, Beck AT, Schene AH. Enduring effects of Preventive Cognitive Therapy in adults remitted from recurrent depression: A 10 year follow-up of a randomized controlled trial. Journal of affective disorders 2015;185:188-194.

7Cost-effectiveness of supported self-help for

recurrent depression in primary care

Karolien E.M. Biesheuvel-Leliefeld Judith E. Bosmans

Filip SmitClaudi L.H. BocktingDigna J.F. van Schaik

Harm W.J. van MarwijkHenriette E. van de Horst

Submitted in: Plos ONE

R1R2R3R4R5R6R7R8R9


Chapter 7

128

ABSTRACT

Background Major depression is a prevalent mental disorder with a high risk of relapse and recurrence. It is associated with a considerable burden for patients and substantial costs for society. Only few studies have focused on the cost-effectiveness of interventions aimed at the prevention of relapse and recurrence of depression in primary care.

Aim The aim of this study was to evaluate the cost-effectiveness of a supported Self-help Preventive Cognitive Therapy (S-PCT) added to treatment-as-usual (TAU) compared with TAU alone for patients with a history of depression, currently in remission.

Methods An economic evaluation alongside a multi-center randomised controlled trial was performed (n = 248) over a 12-month follow-up. Outcomes included relapse and recurrence of depression and quality-adjusted-life-years (QALYs) based on the EuroQol-5D. Analyses were performed from both a societal and healthcare perspective. Missing data were imputed using multiple imputations. Uncertainty was estimated using bootstrapping and presented using the cost-effectiveness plane and the Cost-Effectiveness Acceptability Curve (CEAC). Cost estimates were adjusted for baseline costs.

Results S-PCT statistically significantly decreased relapse and recurrence with 15% (95%CI 3;28) compared to TAU. Mean total societal costs were €2,114 higher (95%CI -112;4261). From a societal perspective, the ICER for recurrence of depression was 13,515. At a Willingness To Pay (WTP) of 22,000 €/recurrence prevented, the probability that S-PCT is cost-effective, in comparison with TAU, is 80%. From a healthcare perspective, the WTP at a probability of 80% should be 11,500 €/recurrence prevented. The ICER for QALYs was 63051. The CEA curve indicated that at a WTP of 30,000 €/QALY gained, the probability that S-PCT is cost-effective compared to TAU is 21%. From a healthcare perspective, at a WTP of 30,000 €/QALY gained, the probability that S-PCT is cost-effective compared to TAU is 46%.

Conclusions Though ultimately depending on the WTP of decision makers, we expect that for both relapse and recurrence and QALYs, S-PCT cannot be considered cost-effective compared to TAU.


Cost-effectiveness of supported self-help for recurrent depression

129

7

INTRODUCTION

Major depression is a prevalent mental disorder that often runs an intermittent lifelong course1, is associated with a high risk of relapse and recurrence2 and with frequently incomplete remission between episodes3–5. It is considered to be among the most disabling illnesses6, and negatively affects many aspects of life7–9.

Largely due to its recurrent nature, the economic consequences of MDD are substantial 10–13. It has been estimated that 1-2% of national healthcare expenses in Western countries is spent on the treatment of depressive disorders14,15. Important factors contributing to the considerable healthcare costs associated with depression are the high prevalence, early age of onset and large risk of relapse and recurrence. However, the majority of costs associated with MDD are due to loss of productivity 14,16–18. Therefore, effective preventive interventions may be beneficial from the viewpoint of patients and society 19,20.

A randomized clinical trial by Biesheuvel et al. (submitted 21) showed superior effect of a supported Self-help Preventive Cognitive Therapy (S-PCT) plus treatment-as-usual (TAU) over TAU alone in preventing relapse and recurrence in patients with a history of depressive episodes. Due to the current economic down-turn and the scarcity of resources available for healthcare, information on the cost-effectiveness of interventions is highly relevant for decision-makers. Alongside this trial, we now investigate the cost-effectiveness of S-PCT compared with TAU from both a societal and a healthcare perspective, in patients with a history of depression.

METHODS

DesignAn economic evaluation was performed alongside a pragmatic randomised controlled trial with two parallel groups of participants. The design of this study is described in more detail elsewhere22. The study is registered in the Dutch Trial Register, www.trialregister.nl, NTR3001.

EthicsThe Medical Ethics Committee of the Vrije Universiteit medical center (VUmc) approved the study protocol and all participants provided written informed consent.

R1R2R3R4R5R6R7R8R9


Chapter 7

130

TerminologyTo describe the course of depression, we use the operational criteria of Frank et al.23. According to these criteria, the course of depression is described as a series of disease stages in which a patient can move from a symptom-free stage, to a stage characterized by some symptoms but not meeting the diagnostic criteria, to a stage with the full-blown disorder, after which the patient can go into remission. When a patient stays in remission for a minimum of six months, he or she is considered to be recovered. A relapse is defined as a depressive episode that occurs during remission and before recovery, while a recurrence is defined as a depressive episode that occurs after recovery.

Treatment allocationAfter inclusion, participants were randomised using computer generated blocks and allocation concealment, stratified by the number of previous depressive episodes (2-3 episodes versus ≥4 episodes) because the number of previous episodes is associated with the risk of relapse and recurrence 24.

BlindingInterviewers were blind for randomization status of the participants during all measurements. Due to the nature of the intervention, it was not possible to blind the participants. At the start of each interview, participants were asked not to reveal their allocation status to the interviewers.

ParticipantsParticipants were recruited through general practices and mental health care services in the Netherlands. To be included in the trial, participants had to a) be 18 years or older, b) be in remission of recurrent MDD for at least two months, but no longer recovered than five years according to the Structured Clinical Interview for DSM-IV Axis 1 disorders (SCID-1 3.0)25, and c) have experienced two or more previous episodes of MDD. The SCID-I interview was conducted by telephone by trained researchers and psychologists. Exclusion criteria were severe cognitive impairments, current or past mania, hypomania or psychosis, current alcohol or drug abuse, or insufficient mastery of the Dutch language.

CounsellorsTwenty-four counsellors (primary care mental health nurses and psychologists) were trained to support the intervention. The psychologists were non-specialised psychologists (no postdoctoral training in clinical interventions). All counsellors attended a one-day training delivered by experienced clinical psychologists, who developed the intervention and, therefore, had an intimate knowledge of PCT. To detect competence issues, audiotaped



131

7

telephone contacts with two participants of each counsellor were evaluated during a one-day supervision with the trainer(s) before the actual start of the trial. During the trial, counsellors could contact the trainers at any time for additional questions and feedback.

Intervention The self-help preventive cognitive therapy is a manualised PCT-based bibliotherapy consisting of a printed self-help book with eight modules and minimal guidance 26. It is based on an effective face-to-face PCT 27,28 and mobile PCT29. PCT for the prevention of depression is an adapted type of Cognitive Therapy (CT) for acute depression30 and aims to prevent relapse and recurrence in remitted patients with a history of depressive episodes. Like regular CT, PCT follows a fixed structure, with agenda setting, review of homework, explanation of the rationale of each session, and the assignment of homework. Participants complete one module per week. Each module includes both reading plus assignments to be completed in approximately 60 minutes. During the first meeting (by phone or face-to-face), the counsellor explained the rationale of S-PCT and the planning for the coming week. Each week the counsellor contacted the participant by phone to evaluate progress and understanding. This call was strictly protocolled and was designed to last no longer than 15 minutes. The nature of the contact was solely to support the participant, and not to actively engage in a therapeutic relationship. Adherence to the intervention protocol was assessed using a checklist. Each week, the counsellor completed this checklist with 4 items; (1) the number of that week’s module (1-8), (2) did the participants read the literature of that week (yes/no plus reason), (3) did the participant do the assignments (yes/no plus reason) and (4) time spent on the call (minutes).

Treatment-as-usualThere were no restrictions to type of TAU. Care providers were not aware of randomization status unless participants informed them. Current TAU guidelines suggest continuation of ADM for at least 6 months after remission31. People at significant risk of relapse and recurrence, should be offered individual Cognitive (Behavourial)Therapy (C(B)T) or Mindfulness-Based Cognitive Therapy (MBCT). TAU (e.g. ADM or psychiatric counselling) was recorded using the Trimbos and iMTA questionnaire for costs associated with Psychiatric Illnesses (TiC-P)32.

Clinical outcome measuresRelapse or recurrence Primary outcome was the incidence of relapse or recurrence of depression over the 12 months follow-up period. To reduce recall bias, telephone SCID-interviews were conducted at both 6 and 12 months and combined into a single outcome (0=no relapse or recurrence, 1=relapse or recurrence).

R1R2R3R4R5R6R7R8R9


Chapter 7

132

Quality of lifeHealth related quality of life (HRQoL) was evaluated using the Dutch translations of the EuroQol-5D (three levels) questionnaire (EQ-5D-3L)33. The EQ-5D measures Health Related Quality of Life (HRQoL) on five dimensions (mobility, self-care, usual activities, pain/discomfort and anxiety and depression), combined into one outcome. Each dimension is rated at three levels corresponding to whether a respondent has no problems, moderate or extreme problems. Utility scores for the EQ-5D health states were estimated using preference weights obtained from the Dutch population 34. Quality-adjusted life-years (QALYs) were calculated by multiplying the utility scores belonging to a health state by the amount of time spent in this health state using linear interpolation between time points.

CostsCosts were assessed every 3 months during the 12-month follow-up period and categorised into 8 groups: primary care costs, secondary care costs, mental health care costs (both in primary and secondary care), home care costs, medication costs, costs due to lost productivity, informal care costs and intervention costs.

Health care utilization costsThe Trimbos/iMTA questionnaire for costs associated with Psychiatric Illnesses (TiC-P) 32 was used to assess the utilization of formal health care services (primary care, secondary medical care, day care, hospital admissions) and informal care during the last 3 months. Costs were computed by multiplying the units of health care use (visits, consultations, sessions, hospital days) by the standard cost prices of these services as reported in the Dutch costing guideline35. Costs of medication were valued using daily defined doses (DDD) and the cost prices of the Royal Dutch Society for Pharmacy 36 to which the pharmacist’s dispensing costs were added. All costs were indexed for the year 2013.

Lost productivity costsThe TiC-p was used to measure productivity losses due to both absenteeism (absent from work) and presenteeism (less productive at work). Costs of absenteeism from paid work were calculated according to both the human capital-, and the friction cost approach, using the mean age- and sex specific income of the Dutch population37. In the human capital method, any hour not worked counts as an hour lost. By contrast, according to the friction cost approach, a sick employee is replaced after a certain amount of time (the friction period; currently 160 days) after which there are no longer productivity losses, because the absent employee will be replaced by someone else. Costs of presenteeism were calculated by asking participants how many working hours should have been replaced due to less productivity at work. Lost productivity due to presenteeism was valued using the mean



133

7

age- and sex specific income of the Dutch population37. Costs of productivity losses due to absenteeism from unpaid work and informal care were calculated using the standard wage of a professional housekeeper38. Discounting was unnecessary, because neither costs nor benefits were recorded beyond 12 months.

Cost price supported self-help PCTA cost price for S-PCT was calculated using a bottom-up approach. Cost price for S-PCT totaled €388.15 per recipient. The average costs of screening the participants was estimated to be €46.12, consisting of 45 minutes of time by the counsellor (€35.92) and the time spent by the participant to complete the interview (€10.20). Costs that were included in the cost price calculation of S-PCT were the printing costs of the bibliotherapeutic booklet including sending it by mail (€2.13), the costs of support and administration time invested by the counsellor (€ 143.67), telephone costs by the counsellor (€5.40) the time costs by participants (on the phone €40.80 + assignments €54.40) and the per-participants costs associated with training (€63.75) and supervising (€31.88) the counsellor. We deliberately included time costs by participants, because in a self-help intervention, the idea is that much of the therapeutic work is done by the participants themselves.

Missing dataMissing data on costs and outcomes were imputed using multiple imputations with chained equations (MICE) using predictive mean matching and fully conditional specification in STATA 12 39,40. An imputation model was created that contained variables related to missing data and variables that significantly differed at baseline between the groups. Ten imputed datasets were created, resulting in loss of efficiency of less than 2.5 % for all outcomes39. The 10 imputed datasets were analyzed separately and the results of the analyses were pooled using Rubin’s rules41.

Statistical analysisThe statistical analyses were conducted according to the intention-to-treat (ITT) principle. In the cost-effectiveness analysis (CEA), we estimated Incremental Cost-Effectiveness Ratios (ICERs) as the ratio of the difference in mean costs divided by the difference in mean effects on depressive relapse or recurrence between S-PCT and TAU. In the cost-utility analysis (CUA), the ICER was defined by the ratio of the difference in costs between the S-PCT and the TAU group by the difference in QALYs. Clinical effects were adjusted for depressive symptoms at baseline, which is a main predictor of relapse and recurrence. Cost differences were adjusted for baseline costs.

Seemingly unrelated regression was used to estimate differences in costs and effects while accounting for potential correlation between costs and effects 42. Costs generally

R1R2R3R4R5R6R7R8R9


Chapter 7

134

have a highly skewed distribution. Therefore, non-parametric bootstrapping with 5000 replications was used to estimate bias-corrected and accelerated confidence intervals around cost differences43 and to estimate uncertainty surrounding the ICERs. The bootstrapped cost-effect pairs were plotted on a cost-effectiveness plane 44 and used to estimate cost-effectiveness acceptability curves (CEACs). CEACs show the probability that the intervention is cost-effective in comparison with the control treatment for a range of ceiling ratios. The ceiling ratio is defined as the amount of money society is willing to pay (WTP, λ) to gain one unit of effect 45.

In the CEA, WTP-thresholds that were associated with probabilities of 80% and 95%, that the intervention was cost-effective compared to TAU, were assessed. In the CUA, the probabilities that the intervention was cost-effective in comparison with TAU at the commonly used WTP thresholds of €20.000 and €30.000 were assessed46.

All analyses were performed with STATA (version 12) and SPSS (IBM SPSS Statistics 20).

Sensitivity analysesTwo sensitivity analyses were performed. The first sensitivity analysis was performed from a health care perspective (HC). From a healthcare perspective, which is used in countries like the United Kingdom by the National Institute for Health and Clinical Excellence31 (NICE), only direct healthcare costs are included, thus excluding costs related to informal care and costs related to lost productivity. In the per-protocol analysis (PP), statistical analysis was restricted to patients who completed at least 80% if the intervention (n=101 out of 124 intervention participants).

RESULTS

Participants flow and recruitmentDetails of enrolment are shown in Figure 1. Recruitment took place between September 2012 and April 2014. Twenty-two databases of primary care practices and 4 databases of specialised mental health care practices were screened for eligible patients. This led to the selection of 5,489 potential participants who received a short information letter. Finally, 248 patients met all inclusion-criteria and signed informed consent. They were randomly allocated to the S-PCT group (124) or to the TAU group (124). Complete effect data over 12 months follow-up were available for 81.5% (101/124 participants). Complete cost data over 12 months follow-up were available for 79.8% (99/124) of the S-PCT patients and 75.8% (94/124) of the TAU patients. Participants without complete cost data experienced significantly more fatigue (0.60, 95%CI: 0.09;1.10) at baseline.



135

7

Baseline characteristicsIn Table 1, baseline socio-demographic, clinical characteristics and costs of the ITT group are presented. Costs at baseline were assessed over the last three months.

Analysed intention to treat (n=124) Analysed per protocol (n=101) Analysed complete follow-up (n=95)

♦ Completed PCT + TAU (n=101/117) ♦ Discontinued intervention (n=16) Reasons Too much time (4) Therapy is too emotional (2) Too much therapy already (1) Lack of motivation (2) Too difficult (3) No restrictive rule (2) Other (2) ♦ Completed 12 month FU (n = 95/124) - Completed questionnaires (n = 98/124) - Completed interviews (n = 104/124) ♦ Lost to follow-up (n= 29) Reasons - Unknown (6) - Logistical reason (3) - Lack of motivation (4) - Stopped with intervention also (5) - Too stressful due to personal circumstances (1) - Lack of time/ no priority (2) - Acute depression? (2) - Other (3)

Allocated to PCT+TAU (n=124) ♦ Received PCT +TAU (n=117) ♦ Did not receive PCT+TAU (n=7) Reasons: Too much depressed (2) Takes too much time (1) Therapy is too emotional (2) Therapist advises to cancel (1) Unknown (1)

♦ Completed 12 month FU (n= 93/124) - Completed questionnaires (n = 95/124) - Completed interviews (n = 106/124) ♦ Lost to follow-up (n= 31/124) Reasons - Unknown (8) - Logistical reason (4) - Too stressful to fill in questionnaires (2) - Acute depression? (4) - Too stressful due to personal circumstances (4) - Unable to contact (4) - Lack of motivation (3) - Other (2)

♦ Allocated to TAU (n=124) ♦ Received TAU (n=124) Withdrew in TAU group immediately after start of study (6)

Analysed intention to treat (n=124) Analysed per protocol (n=124 ) Analysed complete follow-up (n=93)

Allocation

Analysis

12 months Follow-Up

Randomized (n=248)

Enrollment

Assessed for eligibility (n=284)

Excluded (n=36) ♦ Declined to participate (n=2) ♦ Not meeting inclusion criteria (n=34) Current/chronic depression (9) Last episode > 5 years (8) No recurrent depression (9) No or one depression (4)

Burn-out only (3) Anxiety only (2)

Current substance abuse (1) Bipolar disorder (4) Psychosis (3)

Figure 1. Participant flow diagram

R1R2R3R4R5R6R7R8R9


Chapter 7

136

Table 1. Baseline characteristics and societal costs by allocation group1

Characteristics S-PCT(n=124)

TAU(n=124)

All (n=248)

Age, mean (SD)2 48.6 (11.9) 48.8 (11.4) 48.7 (11.7)Females, n (%) 89 (71.8%) 84 (67.7%) 173 (69.8%)No previous episodes, %- 2 or 3 53.2% 49.9% 51.6%- 4 or more 46.8% 50.1% 48.4%Marital status, % (with partner) 64.9% 64.9% 64.9%Education2, % (high education) 42.7% 35.5% 39.1%Age of onset, mean (SD) 28.2 (11.4) 27.5 (12.3) 27.8 (11.9)Depressive symptoms (QIDS-sr), mean (SD) 9.6 (4.8) 8.9 (5.0) 9.3 (4.9)Quality of life- Mental health (SF12, mean (SD) 53.6 (12.2) 53.5 (11.6) 53.5 (11.9)- Physical health (SF12, mean (SD) 59.4 (11.4) 57.6 (11.7) 58.5 (11.6)- EQ-5D (0-1), mean (SD) 0.77 (0.21) 0.78 (0.20) 0.77 (0.2)Comorbid psychopathology (4DSQ)- Anxiety (0-24), mean (SD) 3.2 (3.9) 3.2 (4.3) 3.2 (4.1)- Distress (0-32), mean (SD) 13.0 (7.6) 12.7 (8.0) 12.8 (7.8)- Somatisation (0-32), mean (SD) 8.1 (5.5) 8.9 (5.7) 8.5 (5.6)Pain (MPQ), mean (SD) 2.5 (3.6) 3.2 (4.2) 2.8 (3.9)Fatigue (FSS; 1-7), mean (SD) 3.8 (1.5) 3.9 (1.6) 3.8 (1.6)Self-efficacy (GSES; 10-40), mean (SD) 28.6 (5.9) 28.3 (6.2) 28.4 (6.0)ADM use past 3 months, % 51.8% 56.7% 54.2%Societal costs3 (SD) 1,620 (3,370) 1,185 (2,405) 1,406 (2,944)

Abbreviations: ADM, anti-depressant medication; EQ, EuroQol; FSS, Fatigue Severity Scale; GSES, General Self Efficacy Scale; MPQ, MacGill Pain Questionnaire; QIDS-sr, Quick Inventory of Depressive Symptoms self-report; SD, standard deviation; SF, short form 12-Item Short Form Health Survey; S-PCT, supported self-help preventive cognitive therapy; TAU, treatment-as-usual; 4DSQ, Four Dimensional Symptom Questionnaire 1 Standard deviations were computed from the standard errors: (sd= sqrt( _b[var2] -_b[var]*_b[var])) 2 Education is defined as bachelor’s or master’s degree 3 Mean societal costs per person over three months prior to baseline, euros (€)

Clinical OutcomesClinical effects are presented in Table 2. In the S-PCT group, 44 participants (35%) experienced a relapse or recurrence into a depression compared to 62 participants (50%) in the TAU group over 12 months. This difference was statistically significant (risk difference 0.15, 95%CI 0.03;0.28). The mean pooled difference in QALYs in the S-PCT group compared to the TAU group after 12 months was 0.03 (95%CI 0.0006;0.07).



137

7

Table 2. Multiply imputed pooled clinical outcomes and costs over 12-month follow-upa

S-PCT(n=124)

TAU(n=124)

Difference 95%CIb

Clinical outcomesRelapse or recurrence 0.35 0.50 0.15c 0.03;0.28QALYc 0.80 0.78 0.03c 0.0006;0.07Annual costs (2013, euros)d

- Primary care 648 (89) 517 (72) 131 -107;378- Secondary care 1,680 (462) 871 (250) 810 -40;2246- Mental health care 581 (85) 626 (110) -44 -351;179- Home care 127 (52) 117 (34) 10 -90;148- Medication 291 (34) 215 (26) 77 2.8;172- Lost productivity 3,919 (775) 2,538 (636) 1,381 -544;3148

absenteeism 2,648 (521) 1,411 (468) 1,236 -119;2376presenteeism 1,271 (293) 1,127 (326) 144 -772;1086

Informal care 260 (36) 181 (25) 80 9;169Intervention costs 388 (39) 0 (0) 388 n/aTotal costs, unadjusted 7,897 (1.015) 5,065 (960) 2,832 479;5497

Abbreviations: S-PCT; supported self-help preventive cognitive therapy, QALY; Quality-Adjusted Life-Years, TAU; treatment-as-usual. a Presented are means and mean differencesb 95% confidence intervals obtained by bias corrected and accelerated bootstrappingc Adjusted for depressive symptomsd Annual costs per person, measured at the end of each three-month period during 12 months follow-up

CostsTable 2 presents the multiply imputed and pooled cumulative annual costs in the S-PCT and TAU group over 12 months. After adjustment for baseline costs, the difference in total costs was €2,114 (95%CI -112;4261). For all cost-categories, mean costs for in the S-PCT group were higher than in the TAU group, except for costs of mental health care. There were no patients in the S-PCT group, nor in the TAU group who were absent for more than 160 subsequent days during the 12-month follow-up. Therefore, results of the human capital approach equaled the friction cost approach and are not reported separately.

Cost-effectivenessThe results of the CEA are presented in Table 3, Figure 2a and Figure 3a. The ICER for depressive relapse or recurrence was 13,515, indicating that €13,515 should be invested to prevent 1 relapse or recurrence in the S-PCT group in comparison with the TAU group. The CE-plane shows that 96% of the bootstrapped cost-effect pairs were located in the NE quadrant (more effective and more expensive). The accompanying CEAC showed that the probability that the intervention was considered cost-effective was 4% if WTP is 0 €/ relapse or recurrence prevented, 80% if WTP is 22,000 €/ relapse or recurrence prevented, and 95% if WTP is 39,500 €/ relapse or recurrence prevented.

R1R2R3R4R5R6R7R8R9


Chapter 7

138

Tabl

e 3.

Diff

eren

ces

in r

elap

se o

r re

curr

ence

and

cos

ts o

ver

12 m

onth

s be

twee

n S-

PCT

and

TAU

; IC

ER, C

E-pl

anes

qua

dran

ts, a

nd a

ccep

tabi

lity

Ana

lysi

sO

utco

me;

rel

apse

or

recu

rren

ce

Cos

ts Δ

1 (9

5% C

I)Ef

fect

Δ2

(95%

CI)

ICER

Cos

t-ef

fect

iven

ess

plan

ePr

obab

ility

tha

t S-

PCT

is c

ost-

effe

ctiv

e co

mpa

red

to T

AU

at

WTP

= €

0

Min

imum

will

ingn

ess

to

pay

for

the

inte

rven

tion

to

be

cost

-eff

ectiv

e co

mpa

red

to T

AU

at

give

n pr

obab

iliti

es

NE

SESW

NW

0.80

(80%

)0.

95(9

5%)

ITT

anal

ysis

(so

ciet

al

pers

pect

ive)

2,11

4 (-

112;

4261

)0.

15 (0

.03;

0.28

) 13

,515

96%

3%0

%1%

4%€2

2,00

0€4

0,00

0

Per-

prot

ocol

ana

lysi

s (s

ocie

tal p

ersp

ectiv

e)1,

808

(-49

5;40

25)

0.17

(0.0

3;0.

30)

10,6

0294

%5%

0%1%

6%€1

7,50

0€3

0,00

0

ITT

anal

ysis

(he

alth

care

sy

stem

per

spec

tive)

1,10

7 (7

5;23

22)

0.15

(0.0

3;0.

28)

7,07

997

%3%

0%1%

4%€1

1,50

0€2

1,50

0

Abb

revi

atio

ns:

CE,

cos

t-ef

fect

iven

ess;

CI,

confi

denc

e in

terv

al; C

UA

, cos

t-ut

ility

ana

lysi

s; IC

ER, i

ncre

men

tal c

ost e

ffect

iven

ess

ratio

; ITT

, int

entio

n-to

-tre

at; N

E,

nort

h ea

st; N

W, n

orth

wes

t; SE

, sou

th e

ast;

S-PC

T, s

uppo

rted

sel

f-he

lp p

reve

ntiv

e co

gniti

ve th

erap

y; S

W, s

outh

wes

t; TA

U, t

reat

men

t-as

-usu

al; W

TP, w

illin

gnes

s-to

-pay

1 Adj

uste

d fo

r ba

selin

e co

sts

2 Adj

uste

d fo

r de

pres

sive

sym

ptom

s



139

7

Tabl

e 4.

Diff

eren

ces

in Q

ALY

s an

d co

sts

over

12

mon

ths

betw

een

S-PC

T an

d TA

U; I

CER

, CE-

plan

es q

uadr

ants

, acc

epta

bilit

y an

d W

TP

Ana

lysi

sO

utco

me;

QA

LYC

osts

Δ1

(95%

CI)

Effe

ct Δ

2 (9

5% C

I)IC

ERC

ost-

effe

ctiv

enes

s pl

ane

Prob

abili

ty t

hat

S-PC

T is

cos

t-ef

fect

ive

com

pare

d to

TA

U a

t gi

ven

WTP

NE

SESW

NW

WTP

€0W

TP€2

0.00

0W

TP€3

0.00

0

ITT

anal

ysis

(s

ocie

tal p

ersp

ectiv

e)2,

114

(-11

2;42

61)

0.03

(0.0

006;

0.07

)63

,051

95%

3%0%

2%4%

13%

21%

Per-

prot

ocol

ana

lysi

s (s

ocie

tal p

ersp

ectiv

e)1,

808

(-49

5;40

25)

0.04

(0.0

08:0

.08)

41

,952

94%

5%0%

1%6%

23%

36%

ITT

anal

ysis

(he

alth

care

sy

stem

per

spec

tive)

1,10

7 (7

5;23

22)

0.03

(0.0

006;

0.07

)33

,025

95%

3%0%

2%3%

28%

46%

Abb

revi

atio

ns:

CE,

cos

t-ef

fect

iven

ess;

CI,

confi

denc

e in

terv

al; C

UA

, cos

t-ut

ility

ana

lysi

s; IC

ER, i

ncre

men

tal c

ost e

ffect

iven

ess

ratio

; ITT

, int

entio

n-to

-tre

at; N

E,

nort

h ea

st;

NW

, no

rth

wes

t; Q

ALY

, qu

ality

-adj

uste

d-lif

e-ye

ar;

SE,

sout

h ea

st;

S-PC

T, s

uppo

rted

sel

f-he

lp p

reve

ntiv

e co

gniti

ve t

hera

py;

SW,

sout

h w

est;

TAU

, tr

eatm

ent-

as-u

sual

; WTP

, will

ingn

ess-

to-p

ay1 A

djus

ted

for

base

line

cost

s2 A

djus

ted

for

depr

essi

ve s

ympt

oms

R1R2R3R4R5R6R7R8R9


Chapter 7

140

Cost-utilityThe results of the CUA are presented in Table 4, Figure 2b and Figure 3b. The ICER for QALYs was 63,051 indicating that €63,051 should be invested to gain 1 QALY in the S-PCT group as compared to the TAU group. The CE-plane showed that 96% of QALY cost-effect pairs were located in the NE quadrant (more effective and more expensive) (Fig 2). The CEAC showed that the probability that the intervention was considered cost-effective was 13% if WTP is 20.000 €/QALY gained and that this slowly increased to 21% if WTP is 30,000 €/QALY gained.

Fig 2a. Cost-effectiveness plane (CE-plane) Fig 2b. Cost-effectiveness plane (CE-plane)

CE plane; NE; more expensive, more effective, SE; less expensive, more effective, SW; less expensive, less effective, NW; more expensive, less effective

Fig 3a. Cost-effectiveness acceptability curves Fig 3b. Cost-effectiveness acceptability curve(CEAC) (CEAC)



141

7

Sensitivity analysesThe results of the sensitivity-analyses are presented in Table 3 (relapse or recurrence), and Table 4 (QALY). The analysis from healthcare perspective showed that adjusted costs in the S-PCT group were statistically significantly lower than in the TAU group (mean difference €1,107, 95%CI 75;2322). The effect differences are the same as in the ITT analysis from the societal perspective. The ICER for depressive relapse or recurrence was 7,079 indicating that €7,079 should be invested to prevent 1 relapse or recurrence in the S-PCT group in comparison with the TAU group. The majority of the cost-effect pairs were again located in the NE quadrant (97%). The accompanying CEAC showed that the probability that the intervention was cost-effective was 4% if WTP is 0 €/ relapse or recurrence prevented, 80% if WTP is 11,500 €/ relapse or recurrence prevented, and 95% if WTP is 21,500 €/ relapse or recurrence prevented. From a healthcare perspective, the ICER for QALYs was 33,025, indicating that €33,025 should be invested to gain 1 QALY in the S-PCT group as compared to the TAU group. The probability that S-PCT was considered cost-effective in comparison with TAU for QALYs at WTP thresholds of €20.000 and €30.000 were 28% and 46%, respectively.

A per protocol (PP) analysis with only those participants who completed at least 80% (5 modules) of the intervention (81%; 101 out of 124 participants), showed a mean difference in costs between the S-PCT group and the TAU group of €1,808 (95%CI €-495;4025). S-PCT decreased relapse or recurrence by 17% (95%CI 3;30) and gained 0.04 QALY (95%CI 0.008;0.08) compared to the TAU group. The ICER for depressive relapse or recurrence was 10,602. Again, the cost-effect pairs were mostly located in the NE quadrant of the CE plane (94%). The accompanying CEAC showed that the probability that the intervention was cost-effective was 6% if WTP is 0 €/relapse or recurrence prevented, 80% if WTP is 17,500 €/relapse or recurrence prevented, and 95% if WTP is 30,500 €/relapse or recurrence prevented. In the PP analysis, the ICER for QALYs was 41,952. The probability that S-PCT was considered cost-effective in comparison with TAU for QALYs at WTP thresholds of €20,000 and €30,000 were 23% and 36%, respectively.

DISCUSSION

Main findingsWe evaluated whether treatment-as-usual (TAU) augmented with a supported self-help Preventive Cognitive Therapy (S-PCT) in primary care was cost-effective in comparison with TAU alone for patients with recurrent depression. In the S-PCT group, statistically significantly fewer patients experienced a relapse or recurrence of depression than in the TAU group. For QALYs, a statistically significant difference between the two conditions

R1R2R3R4R5R6R7R8R9


Chapter 7

142

was also observed, in favor of the S-PCT group. However, mean societal costs adjusted for baseline costs were higher in the S-PCT group than in the TAU group over 12 months, though not statistically significantly. Willingness-to-pay values should be quite high to reach an acceptable probability that S-PCT is considered cost-effective in comparison with TAU (80% at a WTP of 22,000 €/relapse or recurrence prevented). For the commonly accepted WTP value of 20,000 €/QALY gained, the probability that S-PCT was cost-effective in comparison with TAU was 13%. Sensitivity analyses showed similar results.

Interpretation and explanation of main findingsThe main contributors to the difference in mean societal costs were the costs due to secondary medical care (mental health care not included) and due to lost productivity. It is unclear how these cost differences can be explained as there were no relevant differences in baseline characteristics, there were no clear outliers for costs and there was no association between costs and relapse or recurrence in the S-PCT group.

Previous research shows that interventions (usually CBT) aimed at the prevention of the onset of MDD in patients with sub-threshold depression (depressive symptoms but insufficient to warrant MDD), generally show good value for money 47–51. However, the few economic evaluations of preventive interventions to reduce the risk of relapse or recurrence of MDD, show mixed results. Stant et al. 52 found that TAU enriched with CBT might be cost-effective in preventing relapses in primary care patients with depression compared to TAU alone. Psychodynamic

counselling for relapse prevention proved to be cost-effective in comparison with usual care 53. The study by Kuyken et al. 54 is the most recently published cost-effectiveness analysis, indicating that MBCT plus the tapering of ADM is not cost-effective in comparison with maintenance ADM in terms of either relapse or recurrence or QALYs in patients with a history of three or more depressive episodes and on a therapeutic dose of maintenance antidepressant drugs medication at baseline. A closer look at their costs reveals that the mean societal costs over 24 months per participant in the intervention group and control group in the study by Kuyken et al. (£3,204/€4,415 and £2,754/€3,795 respectively) were much lower than the mean societal costs over 12 months per participant in our study (€7,897 and €5,065, respectively). Kuyken et al. used other ways of measuring productivity losses and healthcare costs, which might explain some of this difference. However, for example, their mean number of fulltime days off work was half the number of days in our study. An explanation for this difference in days off work is lacking.



143

7

Strengths and limitationsThis study has several strengths. To minimise recall bias, costs over the 12-month follow-up were assessed with questionnaires at 3,6,9 and 12 months. Furthermore, our measurement of depression was based on a well-validated and reliable structured clinical interview (Structured Clinical Interview for DSM-IV Disorders-1)25. A further strength of this study is that the trial participants achieved remission or recovery on antidepressants, other psychotherapies, psychiatric help, counselling, or no treatment at all, as is typically the case in clinical practice. Moreover, there were no restrictions in the use of medication at entry to the study. This was done to maximise the external validity of the study, resulting in high generalizability of the findings.

A limitation of our study is, as is common in studies on psychological interventions, that it was not possible to blind participants to the condition to which they were assigned. Also, the follow-period of 1 year might have been too short to capture all the changes in effects and costs due to the intervention. Time to recurrence might exceed 12 months, possibly implying that we have missed the interventions’ and TAUs’ impact on later recurrences and associated health-economic consequences (such as productivity and health care use).

Conclusions and future researchOur study shows that the larger effects of S-PCT are associated with higher costs as compared to TAU. A WTP of 22,000 €/relapse or recurrence prevented to reach a 80% probability that S-PCT is cost-effective in comparison with TAU is expected to be too high for decision makers. We recommend an extended follow-up in future studies and a more in-depth evaluation of costs. Furthermore, we recommend offering S-PCT directly after treatment in the acute phase55. In addition, it is crucial to study what works for whom.

Acknowledgements We are very grateful to all participants for their participation in the study. Moreover, we would like to thank all recruitment sites for their efforts: GGZ NHN, GGZ Amstelmere, GGZ Zuiderpoort, GGZ Rivierduinen, de Bosgroep, and participating general practitioners. We also thank all counsellors for their support. Finally, we are grateful to Evelien van Valen –who trained our counsellors- for her help.

R1R2R3R4R5R6R7R8R9


Chapter 7

144

REFERENCES

1. Vos T, Haby MM, Barendregt JJ, Kruijshaar M, Corry J, Andrews G. The burden of major depression avoidable by longer-term treatment strategies. Archives of General Psychiatry. 2004;61 (0003-990X (Print)):1097-1103.

2. Burcusa SL, Iacono WG. Risk for recurrence in depression. Clin PsycholRev. 2007;27 (8):959-985.

3. Nierenberg AA, Husain MM, Trivedi MH, et al. Residual symptoms after remission of major depressive disorder with citalopram and risk of relapse: a STAR*D report. PsycholMed. 2010;40 (1):41-50.

4. Fava M, Graves LM, Benazzi F, et al. A cross-sectional study of the prevalence of cognitive and physical symptoms during long-term antidepressant treatment. The Journal of clinical psychiatry. 2006;67(11):1754-1759.

5. Conradi HJ, Ormel J, de Jonge P. Presence of individual (residual) symptoms during depressive episodes and periods of remission: a 3-year prospective study. Psychological medicine. 2011;41(6):1165-1174.

6. Ustun TB, Ayuso-Mateos JL, Chatterji S, Mathers C, Murray CJ. Global burden of depressive disorders in the year 2000. BrJ Psychiatry. 2004;184 (0007-1250 (Print)):386-392.

7. Ormel J, VonKorff M, Ustun TB, Pini S, Korten A, Oldehinkel T. Common mental disorders and disability across cultures. Results from the WHO Collaborative Study on Psychological Problems in General Health Care. JAMA. 1994;272(22):1741-1748.

8. Ormel J, Oldehinkel AJ, Nolen WA, Vollebergh W. Psychosocial disability before, during, and after a major depressive episode: a 3-wave population-based study of state, scar, and trait effects. Archives of general psychiatry. 2004;61(4):387-392.

9. Wells KB, Stewart A, Hays RD, et al. The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA. 1989;262(7):914-919.

10. Berto P, D’Ilario D, Ruffo P, Di VR, Rizzo F. Depression: cost-of-illness studies in the international literature, a review. J MentHealth Policy Econ. 2000;3 (1091-4358 (Print)):3-10.

11. Greenberg PE, Birnbaum HG. The economic burden of depression in the US: societal and patient perspectives. ExpertOpinPharmacother. 2005;6 (1744-7666 (Electronic)):369-376.

12. Smit A, Kluiter H, Conradi HJ, et al. Short-term effects of enhanced treatment for depression in primary care: results from a randomized controlled trial. PsycholMed. 2006;36 (1):15-26.

13. Vasiliadis HM, Dionne PA, Preville M, Gentil L, Berbiche D, Latimer E. The excess healthcare costs associated with depression and anxiety in elderly living in the community. Am J GeriatrPsychiatry. 2013;21 (1545-7214 (Electronic)):536-548.

14. Greenberg PE, Kessler RC, Birnbaum HG, et al. The economic burden of depression in the United States: how did it change between 1990 and 2000? The Journal of clinical psychiatry. 2003;64(12):1465-1475.

15. Polder J, Takkern J, Meerding W, Kommer G, Stokx L. Cost of illness in the Netherlands. In: Vol Bilthoven: RIVM; 2002.

16. Egede LE. Major depression in individuals with chronic medical disorders: prevalence, correlates and association with health resource utilization, lost productivity and functional disability. General hospital psychiatry. 2007;29(5):409-416.

17. Broadhead WE, Blazer DG, George LK, Tse CK. Depression, disability days, and days lost from work in a prospective epidemiologic survey. JAMA. 1990;264 (0098-7484 (Print)):2524-2528.

18. Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380 (1474-547X (Electronic)):2163-2196.

19. Smit F, Ederveen A, Cuijpers P, Deeg D, Beekman A. Opportunities for cost-effective prevention of late-life depression: an epidemiological approach. Archives of general psychiatry. 2006;63(3):290-296.

20. Smit F, Beekman A, Cuijpers P, de Graaf R, Vollebergh W. Selecting key variables for depression prevention: results from a population-based prospective epidemiological study. Journal of affective disorders. 2004;81(3):241-249.



145

7

21. Biesheuvel-Leliefeld K, Dijkstra-Kersten S, van Schaik D, et al. Effectiveness of a supported self-help in recurrent depression: a randomised controlled trial in primary care.

22. Biesheuvel-Leliefeld KE, Kersten SM, van der Horst HE, et al. Cost-effectiveness of nurse-led self-help for recurrent depression in the primary care setting: design of a pragmatic randomised controlled trial. BMC Psychiatry. 2012;12 (1471-244X (Electronic)):59.

23. Frank E, Prien RF, Jarrett RB, et al. Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Archives of General Psychiatry. 1991;48 (0003-990X (Print)):851-855.

24. Beshai S, Dobson KS, Bockting CL, Quigley L. Relapse and recurrence prevention in depression: current research and future prospects. Clin PsycholRev. 2011;31 (8):1349-1360.

25. First MB, Gibbon M, Spitzer RL, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV) . Washington, D.C: American Psychiatric Press, Inc; 1996.


27. Bockting CLH, Spinhoven P, Wouters LF, Koeter MWJ, Schene AH. Long-term effects of preventive cognitive therapy in recurrent depression: a 5.5-year follow-up study. J ClinPsychiatry. 2009;70 (12):1621-1628.

28. Bockting CLH, Schene AH, Spinhoven P, et al. Preventing relapse/recurrence in recurrent depression with cognitive therapy: a randomized controlled trial. J Consult ClinPsychol. 2005;73 (4):647-657.

29. Kok G, Burger H, Riper H, et al. The Three-Month Effect of Mobile Internet-Based Cognitive Therapy on the Course of Depressive Symptoms in Remitted Recurrently Depressed Patients: Results of a Randomized Controlled Trial. Psychotherapy and psychosomatics. 2015;84(2):90-99.


31. NICE. National Institute for Health and Clinical Excellence. In: National Institute for Health and Clinical Excellence. Depression: The Treatment and Management of Depression in Adults (update). Vol Clinical Guidelines: National Institute for Health and Clinical Excellence. Depression: the treatment and management of depression in adults (update).; 2009.

32. Hakkaart-van Roijen L, Van Straten A, Donker M. Manual:Trimbos/iMTA Questionnaire for Costs Associated with Psychiatric Illness (in Dutch). Rotterdam: Erasmus University; 2002.

33. Brooks R. EuroQol: the current state of play. Health Policy. 1996;37(1):53-72.

34. Lamers LM, Stalmeier PF, McDonnell J, Krabbe PF, van Busschbach JJ. [Measuring the quality of life in economic evaluations: the Dutch EQ-5D tariff]. Ned Tijdschr Geneeskd. 2005;149 (0028-2162 (Print)):1574-1578.

35. Oostenbrink JB, Bouwmans C, Koopmanschap MA, C., Rutten. Standardisation of Costs: The Dutch Manual for Costing in Economic Evaluations. Vol 7. Adis International; 2002.

36. Z-index. G-Standaard. 2012.


38. CAK. Centraal Administratie Kantoor, Jaarverslag 2008. Den Haag; 2009.

39. Sterne JAC, White IR, Carlin JB, et al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ (Clinical research ed). 2009;338:b2393.

40. White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Statistics in medicine. 2011;30(4):377-399.

41. Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York, NY US: John Wiley & Sons; 1987.

42. Willan AR, Briggs AH, Hoch JS. Regression methods for covariate adjustment and subgroup analysis for non-censored cost-effectiveness data. Health economics. 2004;13(5):461-475.

43. Efron B. Missing data, imputation and the bootstrap. J Am Stat Assoc. 89(426):463-475.

44. Black WC. The CE plane: a graphic representation of cost-effectiveness. Medical decision making : an international journal of the Society for Medical Decision Making. 10(3):212-214.

45. Fenwick E, O’Brien BJ, Briggs A. Cost-effectiveness acceptability curves--facts, fallacies and frequently asked questions. Health Economics. 2004;13 (1057-9230 (Print)):405-415.

R1R2R3R4R5R6R7R8R9


Chapter 7

146

46. National Institute for Health and Care Excellence (NICE). Guide to the Methods of Technology Appraisal. London: National Institute for Health and Care Excellence; 2013.

47. Smit F, Willemse G, Koopmanschap M, Onrust S, Cuijpers P, Beekman A. Cost-effectiveness of preventing depression in primary care patients: randomised trial. British Journal of Psychiatry. 2006;188 (0007-1250 (Print)):330-336.

48. Lynch FL, Hornbrook M, Clarke GN, et al. Cost-effectiveness of an intervention to prevent depression in at-risk teens. Archives of general psychiatry. 2005;62(11):1241-1248.

49. Mihalopoulos C, Vos T, Pirkis J, Smit F, Carter R. Do indicated preventive interventions for depression represent good value for money?. Australian and New Zealand Journal of Psychiatry. 2011;45 (1440-1614 (Electronic)):36-44.

50. Spek V, Cuijpers P, Nyklícek I, et al. One-year follow-up results of a randomized controlled clinical trial on internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years. Psychological medicine. 2008;38(5):635-639.

51. Van den Berg M, Smit F, Vos T, van Baal PHM. Cost-effectiveness of opportunistic screening and minimal contact psychotherapy to prevent depression in primary care patients. Hashimoto K, ed. PloS one. 2011;6(8):e22884.

52. Stant AD, TenVergert EM, Kluiter H, Conradi HJ, Smit A, Ormel J. Cost-effectiveness of a psychoeducational relapse prevention program for depression in primary care (Provisional abstract). Journal of Mental Health Policy and Economics. 2009;12 (4):195-204.

53. Simon GE, Von KM, Ludman EJ, et al. Cost-effectiveness of a program to prevent depression relapse in primary care. Medical Care. 2002;40 (0025-7079 (Print)):941-950.

54. Kuyken W, Hayes R, Barrett B, et al. Effectiveness and cost-effectiveness of mindfulness-based cognitive therapy compared with maintenance antidepressant treatment in the prevention of depressive relapse or recurrence (PREVENT): a randomised controlled trial. The Lancet. 2015;386(9988):63-73.

55. Biesheuvel-Leliefeld KEM, Kok GD, Bockting CLH, et al. Effectiveness of psychological interventions in preventing recurrence of depressive disorder: Meta-analysis and meta-regression. Journal of affective disorders. 2015;174C:400-410.

8General discussion

R1R2R3R4R5R6R7R8R9


Chapter 8

148


General discussion

149

8

Background and research questionsMajor Depressive Disorder (MDD) is prevalent1, has a high risk of relapse and recurrence2 and is therefore potentially, a chronic, lifelong illness for many people3. In order to stop the rhythm of MDD, prevention of relapse and recurrence is of importance. Though continuation of antidepressant medication (ADM) is recommended for the prevention of relapse and recurrence4, and therefore is the mostly used strategy, it may not be the most optimal preventive strategy5–10. Preventive psychological interventions4 might be a better option.

There were several knowledge gaps regarding recurrent MDD and the prevention of relapse and recurrence by psychological interventions in people with a history of depression. To add to the knowledge about prevention of relapse and recurrence in recurrent MDD we tried to answer the following research questions:

1) What is the burden of disease of recurrent depression compared to single episode depression?2) What is the effectiveness of existing psychological interventions compared both to usual care and the continuation of ADM, to prevent relapse and recurrence in recurrent depression?3) What is the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care?4) What is the (cost-)effectiveness of supported self-help in primary care, for the prevention of relapse and recurrence in recurrent depression?

Outline of discussionIn the present chapter, we summarise the main findings, thereby answering the research questions. Further, the main findings are described in light of previous research and we comment on some methodological considerations, associated with the studies in this thesis. Finally, we reflect on the implications for clinical practice and conclude with recommendations for further research.

R1R2R3R4R5R6R7R8R9


Chapter 8

150

MAIN FINDINGS

Chapter 2, research question 1Single depressive episodes emerge as a key driver of disease burden from an individual perspective. From a population perspective, recurrent depressions emerge as a key driver.For this study, we used data from the first wave of the second Netherlands-Mental-Health-Survey-and-Incidence-Study11 (NEMESIS-2, n=6,646; single episode DSM-IV depression, n=115; recurrent depression, n=246). Disease burden from an individual perspective was assessed as ‘disability weight * time spent in depression’ for each person in the dataset, resulting in Quality Adjusted Life Year (QALY) decrements. We found that a single episode poses a greater burden on individuals than a recurrent depressive episode (QALY decrement=0.111 and 0.078 respectively). From a population perspective, disease burden was assessed as ‘disability weight * time spent in depression *number of people affected’ resulting in Years Lived with Disease (YLD). From this perspective, recurrent depression causes a larger disease burden than single episode depression (YLD/mln= 2,782 and 1,882 respectively. This implies that the burden of disease differs between subtypes of depression and depends much on the choice of perspective. Both perspectives serve different purposes and should be made explicit to avoid misunderstandings between policy makers and clinicians.

Chapter 3, research question 2Preventive psychological interventions are significantly better than treatment-as-usual (TAU) in reducing the risk of relapse or recurrence and also more successful than the continuation of ADM.For this meta-analysis and meta-regression, we systematically reviewed the pertinent trial literature until May 2014 and found 25 studies that met inclusion criteria; a) a randomised controlled trial b) examining adult patients in the age bracket of 18-64 year c) with recurrent MDD d) who were in remission (according to their own definition in the individual trial-paper) at randomisation e) receiving a preventive psychological intervention f) with the aim of reducing the risk of relapse or recurrence and g) with a comparison to a control condition. Control conditions could be classed as TAU (routine clinical management, assessments only, no treatment and waiting-list control with unrestricted access to TAU) or anti-depressant medication (ADM). The psychological interventions were diverse and consisted of Internet-based treatments, booster sessions, group and individual treatment. We included trials on Cognitive (Behavioural) Therapy, Interpersonal Therapy (IPT) and Mindfulness Based CT (MBCT). We found no randomised controlled trials on other psychotherapies such as problem-solving therapy and psychodynamic therapy.


General discussion

151

8

Preventive psychological interventions were significantly better than TAU in reducing the risk of relapse or recurrence (RR=0.64, 95%CI=0.53-0.76, z= 4.89, p<0.001, NNT=5) and also more successful than ADM (RR=0.83, 95%CI=0.70-0.97, z=2.40, p=0.017, NNT=13). Meta-regression showed that the preventive effect of psychological interventions was usually better, when the prevention was preceded by treatment in the acute phase (b=-1.94, SEb=0.68, z=-2.84, p=0.005).

Chapter 4, research question 3Adding preventive Cognitive Therapy (CT) and Mindfulness Based Cognitive Therapy (MBCT) to enhanced TAU, might make the healthcare system for recurrent depression more cost-effective compared to enhanced TAU alone.A health economic (Markov) model that synthesizes clinical, economic and epidemiological evidence was used to assess return on investments (ROI)12. The model calculated the total healthcare costs and health gains by comparing a base-case scenario (enhanced TAU) with four scenarios in terms of cost-effectiveness: enhanced TAU plus A) CT; B) MBCT and C)IPT. Enhanced TAU is a hypothetical, evidence-based healthcare system for depressive disorder in full agreement with the Dutch clinical guidelines for the treatment of depression. We used clinical effectiveness data from our review on the effectiveness of psychological interventions on relapse and recurrence13. Costs were estimated by mapping the total time of one intervention (hours) multiplied by the appropriate full economic costs of the healthcare professional according to the Dutch guidelines for health economic evaluations14. The ROI of enhanced TAU is €1.30 while the ROI of adding CT, MBCT or IPT is €1.43, €1.45 and €1.31 respectively.

Chapter 6, research question 4Adding a supported self-help preventive cognitive therapy (S-PCT) to TAU for people with a history of depressions is more effective than TAU alone in the prevention of relapse and recurrence over 12 months. We conducted a randomised controlled trial, evaluating the (cost-)effectiveness of TAU augmented with S-PCT, compared with TAU alone for people with a history of depressions, currently in remission (n=248). Details of this study are described in Chapter 5. We found that in the S-PCT group, 44/124 participants (35.5%) relapsed compared to 62/124 participants (50.0%) in the TAU group. Therefore, our analyses showed a significant risk difference in relapse rate of 14% (95%CI 2-24, number needed to treat=8) in favour of the S-PCT group after 12 months (incidence rate ratio=0.71, 95%CI 0.52 to 0.97).

R1R2R3R4R5R6R7R8R9


Chapter 8

152

The cost-effectiveness of S-PCT was assessed in two ways, which resulted in two (different) findings.A) Chapter 4, research question 4In order to reach the most competitive ROI (€1.45), a hypothetical S-PCT needs to reach a relative risk reduction of 0.71. This relative risk reduction is feasible (Chapter 6). A hypothetical S-PCT was added to the health economic Markov model. We found that in order to reach the most competitive ROI (MBCT, €1.45), the hypothetical S-PCT needs to reach a relative risk reduction of 0.71. Chapter 6, assessing the real-life effectiveness of S-PCT in a pragmatic randomised controlled trial (RCT), shows that a relative risk reduction of 0.71 is feasible. B) Chapter 7, research question 4 Adding S-PCT to TAU for people with a history of depressions is not cost-effective compared to TAU alone, in the prevention of relapse and recurrence over 12 months.An economic evaluation alongside a RCT was performed over a 12-month follow-up. Mean total societal costs were €2,114 higher (95%CI -112;4261). From a societal perspective, the ICER for recurrence of depression was 13,515. At a Willingness To Pay (WTP) of 22,000 €/relapse or recurrence prevented, the probability that S-PCT is cost-effective, in comparison with TAU, is 80%. From a healthcare perspective, the WTP at a probability of 80% should be 11,500 €/relapse or recurrence prevented. The ICER for QALYs was 63,051. The cost-effectiveness acceptability curve indicated that at a WTP of 30,000 €/QALY gained, the probability that S-PCT is cost-effective in comparison with TAU is 21%. From a healthcare perspective, at a WTP of 30,000 €/QALY gained, the probability that S-PCT is cost-effective in comparison with TAU is 46%.


General discussion

153

8

INTERPRETATION AND EXPLANATION OF MAIN FINDINGS

The burden of disease of recurrent depression compared to single episode depressionMDD was the second leading cause of total burden of disease in 201015, accounting for 8.2% of the global burden of disease, behind low back pain. Within the group of mental and substance use disorders, 41% of the global burden of disease was caused by depressive disorders16. However, depressive disorder is not a homogenous condition and burden of disease estimates might vary across subtypes (e.g. single and recurrent episodes). Kruijshaar et al.17 studied the associations of type of depression with functional impairment of the individual in a Dutch general population sample. They concluded that recurrent depression was not associated with more impairment than single episode depression. In contrast, Vos et al.18 found that recurrent depression is the key driver of disease burden and that depression should be treated as a chronic episodic disorder in order to reduce this great burden of disease. These studies took either the individual- or the population perspective when assessing burden of disease. From an individual perspective, clinicians tend to give priority to the disorders that exact the heaviest toll on their patients, while from a population perspective the disease burden might be driven by the number of people affected, in addition to case severity and disease duration. Both perspectives may lead to difference results in terms of burden of disease. Indeed, a study by Lokkerbol et al. 19 into the non-fatal burden of several mental disorders showed that the rank order of disease burden of several disorders at an individual level is often different from the rank order at the population level. Using data of the second Netherlands Mental Health Survey and Incidence Study (NEMESIS-2, n=6,646)11, we showed that single depressive episodes emerge as a key driver of disease burden from an individual perspective and that recurrent depressions emerge as a key driver from a population perspective (Chapter 2). Both perspectives serve different purposes and may require careful alignment when being used jointly. Such an alignment may result in the optimal balance between an individual approach directed, for example, at the episodic treatment of acute single episode depressions, in combination with a public health care approach with an emphasis on the longer-term preventive management of recurrences.

In the study in Chapter 2, we assessed the burden of disease due to depressive relapse or recurrence. However, we know that approximately one third of the recurrently depressed people experience residual symptoms during remission or recovery20–22. Residual symptoms of depression probably reflect persistence of the original disorder in a milder form and cause significant functional impairment23,24. Unfortunately, we had no data on burden of disease due to residual symptoms of depression during remission or recovery. In future studies, when assessing the burden of disease of recurrent depression it would be interesting to take the burden of disease due to all stages (i.e. acute phase, remission, recovery) into account.

R1R2R3R4R5R6R7R8R9


Chapter 8

154

Effectiveness of preventive psychological interventionsPreventive psychological interventions were effective in reducing relapse and recurrence over a mean follow up of two years versus TAU and ADM treatment with a relative risk of 0.64 and 0.83, respectively (Chapter 3). These results are an extension to previous research which demonstrated that C(B)T, including MBCT, after remission might be equally effective in reducing the risk of depressive relapse and recurrence as ADM and more effective than TAU25–30. Further, we found that the effectiveness of the psychological interventions increased when the interventions directly followed acute phase treatment.

Previous research demonstrated that the effectiveness of MBCT and preventive C(B)T was limited to patients with a higher number of previous episodes27,28,30–35. An explanation for this might be that preventive treatments function to disrupt the internal depressive associations which the sufferer tends to make during the course of his condition36. Our findings, both in the meta-analysis (Chapter 3) and in the RCT (Chapter 6), did not demonstrate this; preventive psychological interventions were effective, irrespective of the number of previous episodes. Current national and international clinical practice guidelines state that prevention of relapse with preventive C(B)T of MBCT is especially effective in patients with three or more episodes and that the number of previous depressive episodes should be taken into account when deciding on relapse prevention25,27,28,37–39. Our results suggest that prevention of relapse and recurrence can be advised to all patients with recurrent MDD, irrespective of their depression history. Cost-effectiveness of preventive psychological interventionsThe return on investments (ROI) of enhanced TAU was €1.30 while the ROI of adding CT, MCT or IPT was €1.43, €1.45 and €1.31 respectively. In other words, we found that adding CT or MCT might make the healthcare system for recurrent MDD more cost-effective than enhanced TAU. It should be noted that assumptions in the model are conservative and that the base-case scenario is likely to be more effective than the current Dutch healthcare system. As a consequence, the ROI of the base-case scenario (€1.30) is likely to over-estimate the ROI of the present health care system for depressive disorders in the Netherlands. This implies that results might be more optimistic.

As far as we know, this is the first modelling article on the cost-effectiveness of preventive psychological interventions. With regard to real-life effectiveness, only few RCTs on the cost-effectiveness of preventive psychological interventions versus TAU and ADM have been assessed and results of these studies are mixed29,40–43. As the substantial economic consequences of MDD are largely due to its recurrent character44–47, cost-effective interventions for the prevention of relapse and recurrence are sorely needed. Therefore, we strongly recommend to conduct real-life studies on the cost-effectiveness of preventive interventions for relapse and recurrence.


General discussion

155

8

Supported self-help for recurrent depression Effectiveness of S-PCTMany randomised trials and meta-analyses have shown that (supported) self-help is effective in the treatment of acute depression and other disorders48–56. Our study is the first showing that a self-help for remitted patients in primary care is effective in preventing relapse and recurrence compared to TAU. The effect was even more pronounced in participants that completed a higher number of modules (≥5) (risk difference=15%, 95%CI 4-24, NNT=7). The latter finding fits previous research on acute treatment of depression demonstrating that the effectiveness of minimally supported interventions might depend on treatment adherence57.

The self-help in this study was supported by primary care mental health nurses and clinical psychologists. Studies already showed that mental health nurses are capable of providing high quality psychological interventions in primary care58–60. We think the type of counsellor might not have impacted findings on a large scale for three reasons. First, in a meta-analysis on moderators of self-help interventions, Gellatly et al.51 found that there were no significant relationships between effect size and personnel type, content of the guidance and mode of the guidance. Second, the backgrounds and educational level of nurses is diverse as both literature61–63 and our study showed. Finally, the support given by the counsellor was primarily of supportive and facilitative nature and meant to support the participant in working through the manual. Also, the support of the intervention was strictly protocolled leaving no room for personal contributions.

Cost-effectiveness of S-PCTIn this thesis, we assessed the cost-effectiveness of S-PCT in two ways. The positive results of our modelling work did not correspond to the negative results of the cost-effectiveness analysis (CEA) alongside the RCT. We found two possible explanations for these different results. Pragmatic RCTs are designed to evaluate the effectiveness of interventions in real-life routine practice conditions, whereas modelling studies aim to test whether an intervention works under certain conditions. Assumptions in our model were as conservative as possible. Still, these assumptions may have portrayed an overly optimistic outcome scenario and might not have resembled clinical practice in a credible way. Another explanation may be the different cost-price of S-PCT between the studies which is €178 per participant in the modelling study and €388.15 per participant in the CEA. This difference is caused by a different approach in calculating these costs. However, an additional analysis showed that replacing the higher costs of €388.15 by €178 euro in the CEA, did not impact much on the results. To end with, in the CEA costs of the S-PCT group were substantially higher. An obvious explanation for these higher costs is lacking. As a consequence, an explanation for the different results between the modelling study and the CEA is also lacking.

R1R2R3R4R5R6R7R8R9


Chapter 8

156

EnrolmentOnly 4,5% (248 out of 5.489) of the persons that were invited, were enrolled. Both methodological (selecting the ‘right’ patient) and motivational (motivating a patient to participate) problems may have played a role here. The methodological problems should be placed in the context of conducting a trial while motivational problems might play a role in the context of conducting a trial but also in the context of offering a preventive intervention for recurrent depression in clinical practice.

Medical records of primary care practices and mental health care practices were screened for eligible persons. Obviously, selecting persons who are in remission of recovery from a database is quite difficult and therefore, many non-eligible persons (e.g. chronically depressed patients) were invited. Direct referral of primary are patients into the trial by the general practitioner after an acute phase treatment would have been preferred, because relapse prevention shows better results when offered directly after acute phase treatment13 (Chapter 3) and because there are some indications that patients referred to self-help by their general practitioner benefit most from these treatments, compared with patients who are referred by mental health professionals and self-referrals from the community64. However, due to time restraints, we chose to screen medical records of primary care practices and mental health care practices.

Poor recruitment to RCTs is a widespread problem65. The decision to enter a trial or accept a treatment involves a judgement between risk and reward by the patient66. A combination of depressive feelings with positive expectations about the outcomes of the trial, will probably positively affect recruitment67. However, participants considered for a prevention trial or treatment are not currently diagnosed as having an illness. Typically, they are at risk of a disorder and may not be aware of this. This may influence their attitudes towards a trial or treatment and hence their behaviour68. A self-help intervention might facilitate the judgement between risk and reward by lowering the threshold of requesting for help for those who are not aware of their risk of a disorder or unwilling to participate in formal methods of treatment. Finally, due to the very same reasons that constitute their vulnerability for depression69, such as lack of concentration and confidence and low motivation, it might be difficult for remitted patients to enrol in a trial or to accept a treatment.

Adherence and acceptability In pragmatic trials, an important outcome is the level to which the intervention was implemented according to protocol70. We found that adherence to the intervention protocol by the counsellors was high. The mean duration of weekly calls of 13.8 minutes seems to imply that the counsellors kept to the protocol and that the support was only motivational without an in-depth and therapeutic conversation. However, some counsellors needed


General discussion

157

8

extra support to keep the ‘supporting trail’ and not to go into detail with a participant (based on feedback during supervision.

Of all 124 participants in the S-PCT group, 101 participants completed at least 5 modules and no participant dropped-out of the intervention after module 5. Therefore, adherence, defined as the proportion of participants that started the first module of and completed the final module71, was 81.5% (101 out of 124 participants). This adherence was much higher the adherence rate of self-help psychological interventions over the Internet, reporting adherence rates of ‘only’ 65,2%-72%72. The relatively high adherence rate of S-PCT might be explained by the total counsellor support time per participant which was moderately to high in our study (110.2 minutes), compared to other studies in depressed patients and patients in partial remission of depression (21-150 minutes)73–76. It might also be explained by the intervention being developed in such a way that each module could be completed in someone’s own time and place within approximately 60 min and without the need for a computer with access to the Internet.

According to the checklist of the counsellors, in 6% of all contacts, the participant had not read the literature belonging to that week’s module. In 11% of all contacts, the participants declared they did not complete the assignments for that week’s module. These numbers suggest that the acceptability of the therapy was high.

R1R2R3R4R5R6R7R8R9


Chapter 8

158

METHODOLOGICAL CONSIDERATIONS

Deviations from the study protocolBetween April 2012 and April 2014, we sent 5.489 letters and enrolled 248 patients from 22 primary care practices and 4 specialized mental health care practices. One major deviation from the protocol (Chapter 5) occurred in the first part of the study. The inclusion of participants in primary care, as intended, proved difficult. Explanations for this can be found in the previous paragraph (Interpretation and explanation of main findings; enrolment). The number of patients per practice that was both eligible and willing to participate was very low (1-5 per practice). In order to reach the numbers needed to be able to study the effect of the intervention, we decided to recruit patients from secondary care as well. A meaningful distinction between characteristics of participants from primary care or secondary care could not be made; participants that were selected in a primary care database might have been under treatment in secondary care and vice-versa. Also, selected participants might not have been under treatment at all.

Another deviation concerns the support of the self-help intervention. It proved difficult to involve primary care mental health nurses that were willing to support the self-help in this trial. In the Netherlands, a primary care mental health nurse is operative to a maximum of 9 hours per week per average general practice of 2.350 patients77. Because of this restraint, nurses, and the general practitioners they work with, need to set priorities in the allocation of their time spent on mentally ill patients. As a direct consequence, we experienced problems involving mental health nurses for our trial on relapse prevention. Approaching clinical psychologists to support the intervention, proved a successful strategy. Patient selectionWith regard to the generalizability of the findings, the participants recruited within the context of our RCT were willing to accept help for the prevention of relapse or recurrence. Therefore, this sample might not reflect the whole population with recurrent depression. We think our inclusion criteria, (i.e. at least 2 previous episodes, the last one ending no longer than 5 years ago, age 18+, fluent in Dutch) maximised generalizability of the results, because there were no restrictions with respect to the level of depressive symptoms, contrary to other prevention studies28,31,78. Finally, we should keep in mind that most participants that dropped out of follow-up are the ones with poorer outcomes79. Leaving these patients out of the analyses may cause bias and, because not all data are used, reduced statistical power. Fortunately, because ‘missingness’ of data, i.e. due to drop out, could be predicted from available data, both multiple imputation and linear mixed models including all patients, yielded unbiased results.


General discussion

159

8

Time horizonFor trials on (cost-)effectiveness it is recommended to choose a follow-up period that is long enough to capture all (costs and) effects80. In our RCT, a follow-up of 12 months was chosen. Information on time to relapse or recurrence is divergent. Besides, the level of residual symptoms proved to influence the time to relapse or recurrence. A 12-year study by Judd and colleagues81, showed that patients with residual sub threshold symptoms, when compared with asymptomatic patients, had a relapse or recurrence of the next major depressive episode more than 3 times faster. A study by van Londen et al. found that patients with residual symptoms particularly relapse in the first 4 months after remission, while patients without residual symptoms mainly experience a recurrence after 12 months of remission82. Two studies showed that about 90% of the patients who remitted rapidly and fully with CT remained well for at least 1 year after acute phase therapy83,84. These findings possibly imply that we have missed the interventions’ and TAUs’ impact on later recurrences. Therefore, it would be valuable to find out if the positive clinical effects of the intervention change or sustain over a longer time. Sustained positive effects in favour of PCT over 5,5 and 10 years were found in a previous study on the long term effects of PCT in group format30,32. Another preliminary study suggested long-term effects of cognitive therapy over 6 years as applied in remitted recurrently depressed patients (n=40)85. Also, it would be valuable to assess the health-economic consequences (such as productivity and health care use) that are associated with these longer term effects. Pragmatic trialThe pragmatic study that is evaluated in this thesis resembled clinical practice to a high degree. This increases the generalizability of the results to broader populations. Because usual care is offered in both arms, the contrast between the intervention group and the control group may be small. Even in these circumstances, S-PCT significantly reduced relapse and recurrence compared to usual care. However, the results from the cost-effectiveness analysis were less convincing and a clear explanation for the results was lacking. One option is to change the design of the study to find out if a clearer contrast between the intervention group (e.g. S-PCT only) and the control group (MBCT or ADM only) would result in other effect and costs and thus, economic outcomes. However, the growing number of economic evaluations is not only caused by a widespread interest in both economic and clinical information on medical interventions, but also by reimbursement requirements in many countries. To make sure that policy decisions are made that apply to clinical practice, policy makers need both economic and clinical information on the outcomes of a new medical intervention in the ‘real world’. Therefore it is recommended to employ a pragmatic study design for economic evaluation86, like we did.

R1R2R3R4R5R6R7R8R9


Chapter 8

160

Supported self-helpWe do not know whether it was the intervention part of S-PCT, the counsellors’ support or a combination that caused the positive effect. The examination of the effectiveness of the separate parts may help deciding on which treatments strategy to apply. An additional complicating factor for usability is that patients could drop-out of the intervention group, from the counsellors’ support, and the assessments. It would be important to differentiate between these different types of drop-outs to examine which factors influence treatment outcome. Further, while we estimated modules to be completed in 60 minutes per week, we have no information available on the exact amount of time participants spent per module. Neither do we know what proportion of the completers finished the 8 modules within 8 weeks or e.g. 3 months. Gathering information on the time needed to finish modules is relevant because in face-to-face psychotherapy, there are indications that a higher intensity of treatment is associated with a higher effect size87.

AssessmentsThe number of previous episodes of depression was retrospectively assessed with the SCID-1 and recall could have been affected by memory bias. Depressive symptoms were assessed at fixed time points, and therefore, variations in-between could have been missed. Finally, many variables, others than the once studied, could be of importance. For example, information on genes, family history or social support could help throwing light on the pathways to specific outcomes in recurrent depression.

There are several ways to collect cost data in cost-effectiveness studies alongside clinical trials, such as insurance data, medical records, interviews with patients, questionnaires and cost diaries. Because insurance data or medical records do not cover all medical resource utilisation of study patients, information on resource utilisation should preferably be obtained through self-reporting. Since questionnaires and interviews are by nature retrospective, they are subject to recall bias. Cost diaries are completed prospectively and, therefore, less subject to recall bias88 and the most reliable estimate of resource use. However, it is questionable whether depressed patients are up to completing the diaries prospectively, since depression is associated with loss of energy and a diminished ability to think and concentrate89. Since residual symptoms of depression during remission or recovery are prevalent20–22, and probably reflect persistence of the original disorder in a milder form23,24, this probably also counts for the participants in our trial.

In our RCT we made use of self-report questionnaires (Tic-P) to measure resource use. These questionnaires were assessed every three months. Recall periods of 1 and 2 months are mentioned as the optimal recall period in the literature90,91. Therefore, the three months-interval between interviews might have caused some recall-bias.


General discussion

161

8

CLINICAL IMPLICATIONS

Current guidelines on the prevention of relapse and recurrence recommend to encourage a person who has benefited from taking ADM, to continue ADM for at least 6 months after remission of an episode of depression. With respect to psychological interventions, guidelines recommend to offer CBT to persons with a significant history of depression plus residual symptoms, and MBCT to patients with a history of at least three episodes of depression4,37. Our studies13 confirm that C(B)T and MBCT should be offered to all remitted persons, however, irrespective of the type of previous acute-phase treatment, the previous numbers of depressive episodes (at least 2), and the level of residual symptoms. In addition, IPT can be advised as well (Chapter 3). Also, it is recommended to offer C(BT), MBCT and IPT, directly after the acute-phase treatment to increase effectiveness. Further, a self-help intervention like S-PCT can be offered in primary care and might be an effective way for the prevention of relapse and recurrence in persons with at least 2 depressive episodes, with or without residual symptoms during remission or recovery (Chapter 6). With respect to cost-effectiveness, CT and MBCT could be recommended when added to usual care (Chapter 4).

What works for whomThe prevention of relapse and recurrence of depression is highly important for persons who are in remission of recovery. However, the treatment needs of a person may depend on his risk factors profile for relapse (e.g. number of previous episodes and level of residual symptoms) and may vary on the stage of the depressive disorder (i.e. acute phase, remission, recovery). On the basis of the information gathered by profiling and staging, the best treatment can be matched to patients in order to improve outcomes and reduce the burden of depression.

Profiling Preventive interventions are most likely to be effective when targeted at those with a high a priori risk profile of developing the disorder92,93. For example, research demonstrated that in patients with less than 5 episodes, psycho-education led to the same reduction in risk or relapse or recurrence as preventive CBT34. However, in patients with 5 or more previous episodes, preventive CBT resulted in lower relapse and recurrence rates than psycho-education. While replication is needed, this might indicate that more specific treatment is required when patients are at higher risk for relapse or recurrence2,21,31,94–97. Applying personalised prevention strategies, with the choice of treatment depending on the presence of risk factors to depressive relapse or recurrence, might increase the effectiveness of treatment. Current specific preventive interventions like PCT and MBCT might be implemented to prevent depressive relapse and recurrence when an ultra-high risk for relapse or recurrence is present, whereas a less specific intervention, such as psycho-education or perhaps even only monitoring of mood, might be enough to prevent relapse when the risk of relapse or recurrence is relatively less high98.

R1R2R3R4R5R6R7R8R9


Chapter 8

162

Staging Besides by profiling, we also suggest that the type of treatment should match the current stage of the disorder. For example, during remission there is more room for training the recall of positive experiences, and the need for support is suggested to be lower during remission than during acute-phase of depression. In this case, a minimally supported intervention like S-PCT, may benefit persons while in the acute phase, face-to-face MBCT might be the best approach.

Finally, it is questionable whether the response to treatment will remain stable across different episodes of depression99. For example, response to a specific ADM during the first episode is no guarantee for response to the same ADM during the fourth episode. Everything we experience changes us as a person100 and risk factors for relapse or recurrence might change with each consecutive episode. This fits a dynamic treatment approach, where the choice of treatment is not predetermined but depends on the needs of individuals, the stage of the disorder25 and also previous experiences.

In our RCT, we did not have information on the exact number of previous episodes which is a main risk factor for relapse and recurrence. Besides, it was unclear if participants were in remission or recovery at the moment they enrolled. In the future, it is important to make profiling and staging of participants, a routine part of conducting trials.

Implementation of S-PCT for recurrent depression in primary careInstead of replacing existing therapies, S-PCT could be part of a blended care approach. Offering blended care broadens the options when matching treatment strategies to treatment needs. For example, dependent on the results of profiling and staging, one could vary the amount and type of support. Face-to-face contacts might be added or reduced and other technologies, such as self-help manuals, the Internet, text messages and telephone support might be considered. Before implementation, the cost-effectiveness of this blended care approach should be evaluated to assesses the added (monetary) value of this approach.

According to Vanhaecht et al.101, the success of adding a new intervention to the current healthcare system largely depends on three items before the actual implementation can take place. 1) Commitment of the healthcare system (top-down)This thesis argues that the Dutch healthcare system may be insufficiently prepared for the high volume, high costs and high risk of chronic problems such as recurrent depression. In the Netherlands, the healthcare sector has started showing commitment to the problem of recurrent depression by several top-down initiatives. For example, the Department of Health, Welfare and Sports has given priority to the prevention of depression in its health policy in 2014. The introduction of ‘generalistic basic mental health care’ including several preventive treatments, and the introduction of a primary care mental health care nurse (POH-GGZ) in 2007 are another examples of attempts to make the mental health care


General discussion

163

8

system more accessible and more affordable. Also, ongoing research on relapse prevention (e.g. this thesis) is stimulated and sponsored by governments and related institutions. A final example is the gradual introduction of guideline-recommendations on relapse prevention which are now more part of the continued care for patients with recurrent depression. Unfortunately, guidelines on relapse prevention in primary care are still rather unspecific39 and deserve attention. Our trial results add to this knowledge base.2) Ownership of the problem is clear (bottom-up)In 2013, the Dutch King Willem-Alexander said in his yearly message to the Dutch people from the government that the Dutch ‘welfare’ state of the 20th century is on its way out. He said that in its place a “participation society” is emerging, in which people must take responsibility for their own future and create their own social and financial safety nets, with less help from the national government. The self-help type of intervention that we evaluated in our RCT, fits well within such a society. Still, for effectiveness reasons, self-help needs to be embedded in the current healthcare system51. Our meta-analysis of 25 studies (Chapter 3), did not provide clarification on the optimal provider or place of a preventive intervention; the setting of these studies was heterogeneous, varying from primary care and- secondary care to a community setting. In the Netherlands, as in most western countries, primary care professionals have regular contact with the vast majority of the population, learn about the patients’ social situation and provide continuous care102. Besides, the prevalence of patients with depression or depressive feelings in primary practice is around 21%103. Therefore, primary practice is the ideal institution to offer a preventive self-help for depression. However, primary care practitioners still refer more, and particularly the more severe, patients to a more expensive form of care in specialty care than to psychologists and social workers in primary care. Also, patients treated in secondary care may not return to primary care quickly enough103,104. Because mental health expenditures exceed budget in many European countries, a budgetary desire for secondary-primary care substitution is understandable. The policy incentive to strengthen the capacities of primary mental health care under the responsibility of the general practitioners by introducing a primary care mental health care nurse (POH-GGZ) so far has had limited effects103–105.

Unfortunately, in our trial, we notified that relapse prevention is currently not their key-priority.

Further work is needed to define their workload and priorities. A nurse guideline for relapse could help here. In sum, we conclude that, besides the patient him- or herself, primary care may be the best place to provide chronic care for people with recurrent depression. However, primary care might currently not be fully resourced (yet) to respond to this challenge.

R1R2R3R4R5R6R7R8R9


Chapter 8

164

3) Practical issues of integrating a clear concept in the current healthcare systemPrevious research showed that integration of depression care and primary care faces several barriers including patient factors and organizational issues106. For example, moving patients into a prevention treatment might be difficult (see Discussion/Enrolment) and referral to a counsellor in primary care requires a careful and timely coordination between health care groups. Still, we think that the addition of S-PCT to primary care seems a strategy that should be relatively easy implemented into current longitudinal primary care systems. Collaboration between primary and secondary care is in line with Dutch developments in mental health care. Besides, primary care psychologists and mental health nurses could act as case managers in a model for continued care for recurrent depression.

To end withThis thesis started with an overview of the burden of recurrent MDD, both for patients and society, and the challenges that decision makers in health care are confronted with these days. We suggested that decision makers should ask themselves how healthcare for major contributors of disease burden and healthcare costs, like recurrent MDD, is to be organised; how it is to be channelled to the right people, and how the right services can be delivered at the right time, at the right place and at bearable costs.

This thesis showed that recurrent MDD causes a major burden for society. It also showed that a minimally supported self-help intervention offered in primary care to patients with a history of depression can reduce this burden. Still, the prevention of relapse and recurrence of MDD proved complicated, both with regard to reaching high-risk patients as well as to offering personalised, cost-effective strategies to support these high-risk patients. By the ongoing studying of data on processes and outcomes, reflect on them, learn and act, it must be possible to improve the continuous care for people with recurrent MDD. For the well-being of these patients and the sound allocation of available budgets, this improvement process will be crucial for the years to come.


General discussion

165

8

RECOMMENDATIONS FOR FUTURE RESEARCH

In this thesis, we tried to contribute to the research area of relapse and recurrence in major depressive disorder. However, more research in this area is needed.

- The follow-up of trials, e.g. on S-PCT, should be extended to evaluate outcomes on the longer term.

- The continuation of ADM is the mostly used strategy in the prevention of relapse and recurrence. Therefore, an RCT evaluating supported self-help compared to ADM would be clinically relevant.

- To gain a better picture of the total burden of disease of recurrent depression, it would be interesting both to take the burden of disease due to relapse and recurrence and due to residual symptoms during remission of recovery into account.

- In order to move from the current ‘one size fits all’ approach to more tailored care, future research should identify which components of the intervention work best for whom.

- As in many trials, recruitment proved difficult in this prevention trial. Therefore, research should also focus on understanding and addressing the facilitators and barriers to participation of eligible patients in (depression) prevention interventions trials.

- In our RCT (Chapter 6), we included participants in either remission or recovery and we did not differentiate between depressive relapse or recurrence. As a consequence, it is not clear what part of our findings refers to relapse and what part to recurrence. As treatment strategies for prevention of relapse and recurrence of depression serve different goals, it is important to differentiate between relapse and recurrence whenever possible. Therefore, future scientific research should use appropriate terminology.

- Finally, the primary outcome in this study was relapse or recurrence over a 12-month follow-up period. In our CEA we saw that the majority of costs over 12 months were due to absenteeism. Therefore, it might be interesting to adjust depression treatment goals. Return to work and social activities despite residual depressive symptoms might be valuable alternatives. Also patient adjusted outcomes ‘What do patients themselves find important?’, might be relevant.

R1R2R3R4R5R6R7R8R9


Chapter 8

166

REFERENCES

1. De Graaf R, ten Have M, van Gool C, van Dorsselaer S. Prevalence of mental disorders and trends from 1996 to 2009. Results from the Netherlands Mental Health Survey and Incidence Study-2. Social psychiatry and psychiatric epidemiology 2012;47(2):203-13. doi:10.1007/s00127-010-0334-8.


3. Richards D. Prevalence and clinical course of depression: a review. Clinical psychology review 2011;31(7):1117-25. doi:10.1016/j.cpr.2011.07.004.

4. NICE. National Institute for Health and Clinical Excellence. In: National Institute for Health and Clinical Excellence. Depression: The Treatment and Management of Depression in Adults (update). Clinical Guidelines: National Institute for Health and Clinical Excellence. Depression: the treatment and management of depression in adults (update).; 2009. Available at: www.nice.org.uk/CG90.

5. LAREB. Bijwerkingen Centrum. Available at: http://www.lareb.nl/bijwerkingen.

6. Geddes JR, Carney SM, Davies C, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet 2003;361 (0140-6736 (Print)):653-661.

7. Kaymaz N, van OJ, Loonen AJ, Nolen WA. Evidence that patients with single versus recurrent depressive episodes are differentially sensitive to treatment discontinuation: a meta-analysis of placebo-controlled randomized trials. Journal of Clinical Psychiatry 2008;69 (1555-2101 (Electronic)):1423-1436.

8. Bockting CL, ten Doesschate MC, Spijker J, Spinhoven P, Koeter MW, Schene AH. Continuation and maintenance use of antidepressants in recurrent depression. Psychotherapy and Psychosomatics 2008;77 (1423-0348 (Electronic)):17-26.

9. Meijer WE, Heerdink ER, Leufkens HG, Herings RM, Egberts AC, Nolen WA. Incidence and determinants of long-term use of antidepressants. European Journal of Clinical Pharmacology 2004;60 (0031-6970 (Print)):57-61.

10. Ten Doesschate MC, Bockting CL, Schene AH. Adherence to continuation and maintenance antidepressant use in recurrent depression. J.Affect.Disord. 2009;115 (1573-2517 (Electronic)):167-170.

11. De Graaf R, Ten Have M, van Dorsselaer S. The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2): design and methods. International Journal of Methods in Psychiatric Research 2010;19 (1557-0657 (Electronic)):125-141.

12. Lokkerbol J, Adema D, Cuijpers P, et al. Improving the cost-effectiveness of a healthcare system for depressive disorders by implementing telemedicine: a health economic modeling study. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry 2014;22(3):253-62. Available at: http://www.sciencedirect.com/science/article/pii/S1064748113000638. Accessed February 3, 2015.

13. Biesheuvel-Leliefeld KEM, Kok GD, Bockting CLH, et al. Effectiveness of psychological interventions in preventing recurrence of depressive disorder: Meta-analysis and meta-regression. Journal of affective disorders 2015;174C:400-410. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25553400. Accessed March 2, 2015.


15. Ferrari AJ, Charlson FJ, Norman RE, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS medicine 2013;10(11):e1001547. doi:10.1371/journal.pmed.1001547.

16. Whiteford HA, Degenhardt L, Rehm J, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 2013;(1474-547X (Electronic)).

17. Kruijshaar ME, Hoeymans N, Bijl R V, Spijker J, Essink-Bot ML. Levels of disability in major depression: findings from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Journal of Affective Disorders 2003;77 (0165-0327 (Print)):53-64.



General discussion

167

8

19. Lokkerbol J, Adema D, de GR, et al. Non-fatal burden of disease due to mental disorders in the Netherlands. Social Psychiatry and Psychiatric Epidemiology 2013;(1433-9285 (Electronic)).

20. Rush, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry 2006;163(11):1905-1917.

21. Judd LL, Akiskal HS, Maser JD, et al. Major depressive disorder: a prospective study of residual subthreshold depressive symptoms as predictor of rapid relapse. J Affect.Disord. 1998;50 (0165-0327 (Print)):97-108.

22. Paykel ES, Ramana R, Cooper Z, Hayhurst H, Kerr J, Barocka A. Residual symptoms after partial remission: an important outcome in depression. Psychological medicine 1995;25(6):1171-80. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8637947. Accessed March 21, 2015.

23. Zimmerman M, Posternak MA, Chelminski I. Heterogeneity among depressed outpatients considered to be in remission. Comprehensive psychiatry 2007;48(2):113-7. doi:10.1016/j.comppsych.2006.10.005.

24. Trivedi MH, Corey-Lisle PK, Guo Z, Lennox RD, Pikalov A, Kim E. Remission, response without remission, and nonresponse in major depressive disorder: impact on functioning. International clinical psychopharmacology 2009;24(3):133-8. doi:10.1097/YIC.0b013e3283277614.

25. Guidi J, Fava GA, Fava M, Papakostas GI. Efficacy of the sequential integration of psychotherapy and pharmacotherapy in major depressive disorder: a preliminary meta-analysis. Psychol.Med. 2011;41 (2):321-331.

26. Kuyken W, Byford S, Taylor RS, et al. Mindfulness-based cognitive therapy to prevent relapse in recurrent depression. J Consult Clin Psychol. 2008;76 (6):966-978.

27. Piet J, Hougaard E. The effect of mindfulness-based cognitive therapy for prevention of relapse in recurrent major depressive disorder: A systematic review and meta-analysis. Clinical Psychology Review 2011;31 (1873-7811 (Electronic)):1032-1040.

28. Vittengl JR, Clark LA, Dunn TW, Jarrett RB. Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitive-behavioral therapy’s effects. J.Consult Clin.Psychol. 2007;75 (0022-006X (Print)):475-488.

29. Kuyken W, Hayes R, Barrett B, et al. Effectiveness and cost-effectiveness of mindfulness-based cognitive therapy compared with maintenance antidepressant treatment in the prevention of depressive relapse or recurrence (PREVENT): a randomised controlled trial. The Lancet 2015;386(9988):63-73. doi:10.1016/S0140-6736(14)62222-4.

30. Bockting CLH, Smid NH, Koeter MWJ, Spinhoven P, Beck AT, Schene AH. Enduring effects of Preventive Cognitive Therapy in adults remitted from recurrent depression: A 10 year follow-up of a randomized controlled trial. Journal of affective disorders 2015;185:188-194. doi:10.1016/j.jad.2015.06.048.



33. Ma SH, Teasdale JD. Mindfulness-based cognitive therapy for depression: replication and exploration of differential relapse prevention effects. J Consult Clin Psychol. 2004;72 (1):31-40.

34. Stangier U, Hilling C, Heidenreich T, et al. Maintenance cognitive-behavioral therapy and manualized psychoeducation in the treatment of recurrent depression: a multicenter prospective randomized controlled trial. American Journal of Psychiatry 2013;170 (1535-7228 (Electronic)):624-632.

35. Teasdale JD, Segal Z V, Williams JM, Ridgeway VA, Soulsby JM, Lau MA. Prevention of relapse/recurrence in major depression by mindfulness-based cognitive therapy. J Consult Clin Psychol. 2000;68 (4):615-623.


37. American Psychiatric Association. Practice Guidelines for the Treatment of Patients with Major Depressive Disorder, Third Edition.; 2010.

38. Spijker J, Bockting C, Meeuwissen J, et al. Multidisciplinaire Richtlijn Depressie (Derde Revisie). Richtlijn Voor de Diagnostiek, Behandeling En Begeleiding van Volwassen Patiënten Met Een Depressieve Stoornis. (Trimbos-instituut, ed.). Utrecht; 2013.

R1R2R3R4R5R6R7R8R9


Chapter 8

168

39. Van Weel-Baumgarten E, van Gelderen M, Grundmeijer H, et al. NHG-Standaard Depressie (tweede herziening). 2012. Available at: https://www.nhg.org/standaarden/volledig/nhg-standaard-depressie-tweede-herziening#idm225920.

40. Stant AD, TenVergert EM, Kluiter H, Conradi HJ, Smit A, Ormel J. Cost-effectiveness of a psychoeducational relapse prevention program for depression in primary care (Provisional abstract). Journal of Mental Health Policy and Economics 2009;12 (4):195-204. Available at: http://www.mrw.interscience.wiley.com/cochrane/cleed/articles/NHSEED-22010000564/frame.html.

41. Simon GE, Von KM, Ludman EJ, et al. Cost-effectiveness of a program to prevent depression relapse in primary care. Medical Care 2002;40 (0025-7079 (Print)):941-950.

42. Scott J, Palmer S, Paykel E, Teasdale J, Hayhurst H. Use of cognitive therapy for relapse prevention in chronic depression: cost-effectiveness study (Structured abstract). British Journal of Psychiatry 2003;182:221-227. Available at: http://www.mrw.interscience.wiley.com/cochrane/cleed/articles/NHSEED-22003008057/frame.html.

43. Vos T, Corry J, Haby MM, Carter R, Andrews G. Cost-effectiveness of cognitive–behavioural therapy and drug interventions for major depression. Australian and New Zealand Journal of Psychiatry 2005;39(8):683-692. Available at: http://anp.sagepub.com/content/39/8/683. Accessed January 13, 2015.





48. Spek V, Cuijpers P, Nyklicek I, Riper H, Keyzer J, Pop V. Internet-based cognitive behaviour therapy for symptoms of depression and anxiety: a meta-analysis. Psychological Medicine 2007;37 (0033-2917 (Print)):319-328.

49. Cuijpers P. Bibliotherapy in unipolar depression: a meta-analysis. J Behav.Ther.Exp.Psychiatry 1997;28 (0005-7916 (Print)):139-147.

50. Gregory RJ, Canning SS, Lee TW, Wise JC. Cognitive bibliotherapy for depression: A meta-analysis. Professional Psychology-Research and Practice 2004;35 (3):275-280. Available at: ISI:000221700500009.

51. Gellatly J, Bower P, Hennessy S, Richards D, Gilbody S, Lovell K. What makes self-help interventions effective in the management of depressive symptoms? Meta-analysis and meta-regression. Psychological Medicine 2007;37 (0033-2917 (Print)):1217-1228.

52. Cuijpers P, Donker T, van SA, Li J, Andersson G. Is guided self-help as effective as face-to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychological Medicine 2010;40 (1469-8978 (Electronic)):1943-1957.

53. Menchola M, Arkowitz H, Burke B. Efficacy of self-administered treatments for depression and anxiety. Professional Psychology, Research and Practice 2007;(38):421–429.

54. Den Boer PC, Wiersma D, Van den Bosch RJ. Why is self-help neglected in the treatment of emotional disorders? A meta-analysis. Psychological Medicine 2004;34 (0033-2917 (Print)):959-971.

55. Anderson L, Lewis G, Araya R, et al. Self-help books for depression: how can practitioners and patients make the right choice?. Br.J Gen.Pract. 2005;55 (0960-1643 (Print)):387-392.

56. Apodaca TR, Miller WR. A meta-analysis of the effectiveness of bibliotherapy for alcohol problems. Journal of clinical psychology 2003;59(3):289-304. doi:10.1002/jclp.10130.

57. Christensen H, Griffiths KM, Farrer L. Adherence in internet interventions for anxiety and depression. J Med Internet.Res. 2009;11 (1438-8871 (Electronic)):e13.

58. van’t Veer-Tazelaar PJ, van Marwijk HW, van OP, et al. Stepped-care prevention of anxiety and depression in late life: a randomized controlled trial. Archives of General Psychiatry 2009;66 (1538-3636 (Electronic)):297-304.


General discussion

169

8

59. Schreuders B, van MH, Smit J, Rijmen F, Stalman W, van OP. Primary care patients with mental health problems: outcome of a randomised clinical trial. British Journal of General Practice 2007;57 (0960-1643 (Print)):886-891.

60. Van Schaik A, van Marwijk H, Ader H, et al. Interpersonal psychotherapy for elderly patients in primary care. American Journal of Geriatric Psychiatry 2006;14(1064-7481 (Print)):777-786.

61. Younge, O. Boschma G. Debating the integration of psychiatric/mental health nursing content in undergraduate nursing programmes. In: Cutcliffe, J. Ward M, ed. Key Debates in Psychiatric/mental Health Nursing. Edinburgh: Churchill Livingstone; 2007:107–118.

62. Cutcliffe J, McKenna H. Generic Nurses: The Nemesis of Psychiatric/Mental Health Nursing? Key Debates in Psychiatric/Mental Health Nursing 2006:92 - 106. Available at: http://corescholar.libraries.wright.edu/nursing_faculty/21. Accessed July 20, 2015.

63. Wynaden G. Reflections on mental health nursing. Guest Editorial. Contemporary Nurse 2006;21:117-119.

64. Mataix-Cols D, Cameron R, Gega L, Kenwright M, Marks IM. Effect of referral source on outcome with cognitive-behavior therapy self-help. Comprehensive psychiatry 47(4):241-5. doi:10.1016/j.comppsych.2005.11.007.

65. Puffer S, Torgerson D. Recruitment difficulties in Randomised Controlled Trials. Controlled Clinical Trials. 2003;(24:):S214–15.

66. Hughes-Morley A, Young B, Waheed W, Small N, Bower P. Factors affecting recruitment into depression trials: Systematic review, meta-synthesis and conceptual framework. Journal of affective disorders 2014;172C:274-290. doi:10.1016/j.jad.2014.10.005.

67. Van der Weele GM, de Jong R, de Waal MWM, et al. Response to an unsolicited intervention offer to persons aged ≥ 75 years after screening positive for depressive symptoms: a qualitative study. International psychogeriatrics / IPA 2012;24(2):270-7. doi:10.1017/S1041610211001530.

68. Stein RE, Bauman LJ, Ireys HT. Who enrolls in prevention trials? Discordance in perception of risk by professionals and participants. American journal of community psychology 1991;19(4):603-17. Available at: http://www.ncbi.nlm.nih.gov/pubmed/1755438. Accessed August 22, 2015.

69. McCann S, Campbell M, Entwistle V. Recruitment to clinical trials: a meta-ethnographic synthesis of studies of reasons for participation. Journal of health services research & policy 2013;18(4):233-41. doi:10.1177/1355819613483126.

70. Godwin M, Ruhland L, Casson I, et al. Pragmatic controlled clinical trials in primary care: the struggle between external and internal validity. BMC medical research methodology 2003;3(1):28. doi:10.1186/1471-2288-3-28.

71. Hilvert-Bruce Z, Rossouw PJ, Wong N, Sunderland M, Andrews G. Adherence as a determinant of effectiveness of internet cognitive behavioural therapy for anxiety and depressive disorders. Behaviour research and therapy 2012;50(7-8):463-8. doi:10.1016/j.brat.2012.04.001.

72. Richards D, Richardson T. Computer-based psychological treatments for depression: a systematic review and meta-analysis. Clinical psychology review 2012;32(4):329-42. doi:10.1016/j.cpr.2012.02.004.

73. Hollandare F, Johnsson S, Randestad M, et al. Randomized trial of Internet-based relapse prevention for partially remitted depression. Acta Psychiatr Scand 2011;124 (4):285-294.

74. Proudfoot J, Ryden C, Everitt B, et al. Clinical efficacy of computerised cognitive-behavioural therapy for anxiety and depression in primary care: randomised controlled trial. Br.J Psychiatry 2004;185 (0007-1250 (Print)):46-54.

75. Marks IM, Mataix-Cols D, Kenwright M, Cameron R, Hirsch S, Gega L. Pragmatic evaluation of computer-aided self-help for anxiety and depression. The British journal of psychiatry : the journal of mental science 2003;183:57-65. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12835245. Accessed August 21, 2015.

76. Kok G, Bockting C, Burger H, Smit F, Riper H. Mobile Cognitive Therapy: Adherence and acceptability of an online intervention in remitted recurrently depressed patients. Internet Interventions 2014;1(2):65-73. doi:10.1016/j.invent.2014.05.002.

77. Nederlandse Zorgautoriteit. Advies Bekostiging POH-GGZ.; 2013.

R1R2R3R4R5R6R7R8R9


Chapter 8

170

78. Kok G, Burger H, Riper H, et al. The Three-Month Effect of Mobile Internet-Based Cognitive Therapy on the Course of Depressive Symptoms in Remitted Recurrently Depressed Patients: Results of a Randomized Controlled Trial. Psychotherapy and psychosomatics 2015;84(2):90-99. doi:10.1159/000369469.

79. Schindler A, Hiller W, Witthöft M. What predicts outcome, response, and drop-out in CBT of depressive adults? a naturalistic study. Behavioural and cognitive psychotherapy 2013;41(3):365-70. doi:10.1017/S1352465812001063.

80. Drummond M, Sculpher M, Torrance GW, O’Brien B, Stoddart G. Methods for the Economic Evaluation of Health Care Programmes. New York: Oxford University Press Inc.; 2005.

81. Judd LL, Paulus MJ, Schettler PJ, et al. Does incomplete recovery from first lifetime major depressive episode herald a chronic course of illness? The American journal of psychiatry 2000;157(9):1501-4. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10964869. Accessed July 27, 2015.

82. Van Londen L, Molenaar RP, Goekoop JG, Zwinderman AH, Rooijmans HG. Three- to 5-year prospective follow-up of outcome in major depression. Psychological medicine 1998;28(3):731-5. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9626729. Accessed March 9, 2015.

83. Jarrett RB, Kraft D, Doyle J, Foster BM, Eaves GG, Silver PC. Preventing recurrent depression using cognitive therapy with and without a continuation phase: a randomized clinical trial. Arch Gen.Psychiatry 2001;58 (4):381-388.

84. Thase ME, Simons AD, McGeary J, et al. Relapse after cognitive behavior therapy of depression: potential implications for longer courses of treatment. Am.J.Psychiatry 1992;149 (0002-953X (Print)):1046-1052.

85. Fava GA, Rafanelli C, Grandi S, Canestrari R, Morphy MA. Six-year outcome for cognitive behavioral treatment of residual symptoms in major depression. Am J Psychiatry 1998;155 (10):1443-1445.

86. Ramsey S, Willke R, Briggs A, et al. Good research practices for cost-effectiveness analysis alongside clinical trials: the ISPOR RCT-CEA Task Force report. Value in health : the journal of the International Society for Pharmacoeconomics and Outcomes Research 8(5):521-33. doi:10.1111/j.1524-4733.2005.00045.x.

87. Cuijpers P, Huibers M, Ebert DD, Koole SL, Andersson G. How much psychotherapy is needed to treat depression? A metaregression analysis. Journal of affective disorders 2013;149(1-3):1-13. doi:10.1016/j.jad.2013.02.030.

88. Goossens ME, Rutten-van Mölken MP, Vlaeyen JW, van der Linden SM. The cost diary: a method to measure direct and indirect costs in cost-effectiveness research. Journal of clinical epidemiology 2000;53(7):688-95. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10941945. Accessed August 6, 2015.

89. APA. Diagnostic and Statistical Manual of Mental Disorders : DSM-IV-TR.. Washington DC; 2000.

90. Ramsey SD, McIntosh M, Sullivan SD. Design issues for conducting cost-effectiveness analyses alongside clinical trials. Annual review of public health 2001;22:129-41. doi:10.1146/annurev.publhealth.22.1.129.

91. Severens JL, Mulder J, Laheij RJ, Verbeek AL. Precision and accuracy in measuring absence from work as a basis for calculating productivity costs in The Netherlands. Social science & medicine (1982) 2000;51(2):243-9. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10832571. Accessed August 6, 2015.

92. Chisholm D. Reducing the global burden of depression: Population-level analysis of intervention cost-effectiveness in 14 world regions. The British Journal of Psychiatry 2004;184(5):393-403. doi:10.1192/bjp.184.5.393.

93. Smit F, Beekman A, Cuijpers P, de Graaf R, Vollebergh W. Selecting key variables for depression prevention: results from a population-based prospective epidemiological study. Journal of affective disorders 2004;81(3):241-9. doi:10.1016/j.jad.2003.08.007.

94. Fava GA, Ruini C, Rafanelli C, Finos L, Conti S, Grandi S. Six-year outcome of cognitive behavior therapy for prevention of recurrent depression. Am J Psychiatry 2004;161 (10):1872-1876.

95. Mueller TI, Leon AC, Keller MB, et al. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am.J.Psychiatry 1999;156 (0002-953X (Print)):1000-1006.

96. Solomon DA, Keller MB, Leon AC, et al. Multiple recurrences of major depressive disorder. American Journal of Psychiatry 2000;157 (0002-953X (Print)):229-233.

97. Ten Doesschate MC, Bockting CL, Koeter MW, Schene AH. Prediction of recurrence in recurrent depression: a 5.5-year prospective study. Journal of Clinical Psychiatry 2010;71 (1555-2101 (Electronic)):984-991.


General discussion

171

8

98. Bockting C, Hollon SD, Jarrett RB, Kuyken W, Dobson K. A lifetime approach to major depressive disorder: The contributions of psychological interventions in preventing relapse and recurrence. Clinical psychology review 2015. doi:10.1016/j.cpr.2015.02.003.

99. Simon GE, Perlis RH. Personalized medicine for depression: can we match patients with treatments? The American journal of psychiatry 2010;167(12):1445-55. doi:10.1176/appi.ajp.2010.09111680.

100. Kendler KS, Eaves LJ, Loken EK, et al. The impact of environmental experiences on symptoms of anxiety and depression across the life span. Psychological science 2011;22(10):1343-52. doi:10.1177/0956797611417255.

101. Vanhaecht K, Sermeus W. Draaiboek voor de ontwikkeling, implementatie en evaluatie van een klinisch pad. 30 Stappenplan van het netwerk klinische paden. Acta hospitalia 2002;3.



104. Van Balkom A. Imago. Tijdschrift voor Psychiatrie 2005;(12):825-6.

105. KPMG. Substitutiemodel GGZ: Een Inschatting van de Effecten van de Invoering van de Generalistische Basis GGZ En Versterking van de Huisartsenzorg Voor de GGZ in Nederland. (VWS I opdracht van het M van, ed.). Plexus; 2014.

106. Grazier KL, Smith JE, Song J, Smiley ML. Integration of depression and primary care: barriers to adoption. Journal of primary care & community health 2014;5(1):67-73. doi:10.1177/2150131913491290.

Summary

R1R2R3R4R5R6R7R8R9


Summary

174


Summary

175

S

Major Depressive Disorder (MDD) is prevalent, has a high risk of relapse and recurrence and is therefore potentially, a chronic, lifelong illness for many people. MDD was the second leading cause of total burden of disease in 2010, accounting for 8.2% of the global burden of disease, behind low back pain. Within the group of mental and substance use disorders, 41% of the global burden of disease was caused by MDD. Due to the current economic down-turn and an expenditure for mental health care that is unlikely to be sustainable, decision-makers seem to have become increasingly aware of the disease burden and healthcare costs that are associated with MDD. In order to stop the rhythm of recurrent MDD, prevention of relapse and recurrence should play a major role. The proactive management that is currently mostly used, continuation of antidepressant medication (ADM), may not be the most optimal strategy. Research demonstrates that preventive psychological interventions are also effective in reducing the risk of relapse and recurrence.

There are several knowledge gaps regarding recurrent MDD and the prevention of relapse and recurrence by psychological interventions in people with recurrent MDD. In this thesis we try to close these knowledge gaps by answering following research questions: 1) What is the burden of disease of recurrent depression compared to single episode depression?2) What is the effectiveness of existing psychological interventions compared both to usual care and the continuation of ADM, to prevent relapse and recurrence in recurrent depression?3) What is the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care?4) What is the (cost-)effectiveness of supported self-help in primary care, for the prevention of relapse and recurrence in recurrent depression?

In Chapter 2 (Research question 1) we estimated the non-fatal disease burden of both single episode and recurrent depression, from both an individual and a population perspective. We used data from the first wave of the second Netherlands-Mental-Health-Survey-and-Incidence-Study (NEMESIS-2, n=6,646; single episode DSM-IV depression, n=115; recurrent depression, n=246). Results indicated that single depressive episodes emerged as a key driver of disease burden from an individual perspective. From a population perspective, recurrent depressions emerged as a key driver.

Our meta-analysis and meta-regression in Chapter 3 (Research question 2) provided more insight in the effectiveness of existing available psychological interventions, aimed at the prevention of relapse and recurrence in major depressive disorder. We systematically reviewed the pertinent trial literature until May 2014. A distinction was made between

R1R2R3R4R5R6R7R8R9


Summary

176

two comparator conditions: (1) treatment-as-usual and (TAU) (2) the use of antidepressants (ADM). Twenty-five randomised trials met inclusion criteria. Preventive psychological interventions were significantly better than TAU in reducing the risk of relapse or recurrence and also more successful than ADM over 2 years, with a relative risk of 0.64 and 0.83, respectively. Meta-regression showed that the preventive effect of psychological intervention was usually better when the intervention was preceded by treatment in the acute phase.

The aim of our study in Chapter 4 (Research question 3) was to investigate whether offering preventive psychological interventions would improve the cost-effectiveness of the Dutch health care system for recurrent MDD. A health economic model was used to assess return on investments (ROI). We compared a base-case scenario (enhanced TAU) with four scenarios in terms of cost-effectiveness: enhanced TAU plus A) cognitive therapy, CT; B) mindfulness-based CT, MBCT; C) interpersonal therapy, IPT and D) a hypothetical supported self-help based on PCT. Enhanced TAU is an evidence-based healthcare system for depressive disorder in full agreement with the Dutch clinical guidelines for the treatment of depression. The study showed that augmenting enhanced TAU (ROI=€1.30) with CT (ROI=€1.43) or MBCT (ROI=€1.45) might make the healthcare system for recurrent depression more cost-effective compared to enhanced TAU only. In order to reach the most competitive ROI (€1.45), the supported self-help needs to reach a moderate relative risk reduction (0.71).

In Chapter 5 we presented the design of the PARADE-study, a randomised controlled clinical trial (n=248) with the aim to prevent relapse and recurrence in MDD. Participants had a lifetime history of at least 2 depressive episodes and were remitted for at least 2 weeks at the start of the trial. Participants were randomised to receive either TAU (n=124) plus our intervention or TAU alone (n=124). The intervention consisted of a supported Self-help Preventive Cognitive Therapy (S-PCT) consisting of a printed self-help book with eight modules, minimally supported by a counsellor. Participants were contacted weekly for the course of 8 weeks to evaluate progress and understanding of the self-help. The follow-up was 12 months.

The effectiveness of S-PCT was presented in Chapter 6 (Research question 4). S-PCT significantly reduced relapse and recurrence over 12 months compared with TAU (risk-difference 14%). Compared to the TAU group, the S-PCT group showed a significant improvement in symptoms of depression and in quality of life (EQ-5D). The intervention had no effect on co-morbid psychopathology, self-efficacy and quality of life based on the SF12.


Summary

177

S

A cost-evaluation of S-PCT compared to TAU was presented in Chapter 7 (Research question 4). Clinical outcomes, costs and incremental cost-effectiveness ratios were calculated both from a societal perspective and from a healthcare perspective. Mean total societal costs in the S-PCT group were higher than in the TAU group. Willingness-to-pay values should be quite high to reach an acceptable probability that S-PCT is considered cost-effective in comparison with TAU. Sensitivity analyses showed similar results.

Finally, Chapter 8 provided an overview of the main findings in this thesis and answered the four research questions.

INTERPRETATION AND EXPLANATION OF MAIN FINDINGS

The burden of disease of recurrent depression compared to single episode depressionWe showed that single depressive episodes emerge as a key driver of disease burden from an individual perspective and that recurrent depressions emerge as a key driver from a population perspective. Both perspectives serve different purposes and may require careful alignment when being used jointly. Such an alignment may result in the optimal balance between an individual approach directed, for example, at the episodic treatment of acute single episode depressions, in combination with a public health care approach with an emphasis on the longer-term preventive management of recurrences.

Effectiveness of preventive psychological interventionsThe results of our meta-analysis are an extension to previous research which demonstrated that C(B)T, including MBCT, after remission might be equally effective in reducing the risk of depressive relapse and recurrence as ADM and more effective than TAU. However, previous research demonstrated that the effectiveness of MBCT and preventive C(B)T was limited to patients with a higher number of previous episodes. Our results, both from the meta-analysis and the RCT, suggest that prevention of relapse and recurrence can be advised to all patients with recurrent MDD, irrespective of their depression history.

Cost-effectiveness of preventive psychological interventionsAs far as we know, we assessed the first modelling article on the cost-effectiveness of preventive psychological interventions. We found that found that adding CT or MCT might make the healthcare system for recurrent MDD more cost-effective than enhanced TAU. Assumptions in the model were conservative and that the base-case scenario is likely to be more effective than the current Dutch healthcare system. This implies that results might be more optimistic. With regard to real-life effectiveness, only few RCTs on the cost-

R1R2R3R4R5R6R7R8R9


Summary

178

effectiveness of preventive psychological interventions versus TAU and ADM have been assessed and results of these studies are mixed.

Supported self-help for recurrent depression Effectiveness of S-PCTOur study is the first showing that a self-help for remitted patients in primary care is effective in preventing relapse and recurrence compared to TAU. The self-help in this study was supported by primary care mental health nurses and clinical psychologists. Studies already showed that mental health nurses are capable of providing high quality psychological interventions in primary care. For several reasons, we think the type of counsellor might not have impacted findings on a large scale.

Cost-effectiveness of S-PCTThe positive results of our modelling work did not correspond to the negative results of the cost-effectiveness analysis (CEA) alongside the RCT. We found two explanations for these different results. Pragmatic RCTs are designed to evaluate the effectiveness of interventions in real-life routine practice conditions, whereas modelling studies aim to test whether an intervention works under certain conditions. Another explanation may be the different cost-price of S-PCT between the studies. As an obvious explanation for the higher costs in the S-PCT group in the RCT is lacking, an explanation for the difference in results between the modelling study and the CEA is also lacking.

CLINICAL IMPLICATIONS

Current guidelines on the prevention of relapse and recurrence recommend to encourage a person who has benefited from taking ADM, to continue ADM for at least 6 months after remission of an episode of depression. With respect to psychological interventions, guidelines recommend to offer CBT to persons with a significant history of depression plus residual symptoms, and MBCT to patients with a history of at least three episodes of depression. Our studies confirm that C(B)T and MBCT should be offered to all remitted persons, however, irrespective of the type of previous acute-phase treatment, the previous numbers of depressive episodes (at least 2), and the level of residual symptoms. In addition, IPT can be advised as well. Also, it is recommended to offer C(BT), MBCT and IPT, directly after the acute-phase treatment to increase effectiveness. Further, a self-help intervention like S-PCT can be offered in primary care and might be an effective way for the prevention of relapse and recurrence in persons with at least 2 depressive episodes, with or without residual symptoms during remission or recovery. With respect to cost-effectiveness, CT and MBCT could be recommended when added to usual care.


Summary

179

S

FUTURE RESEARCH

Future trials on cost-effectiveness should extend follow-up, should evaluate what works for whom (by profiling and staging) and might adjust depression treatment goals to e.g. return to work or social activities despite residual depressive symptoms and should focus on understanding and addressing the facilitators and barriers to participation of eligible patients in relapse prevention interventions. Finally, as treatment strategies for prevention of relapse and recurrence of depression serve different goals, future studies should differentiate between relapse and recurrence whenever possible.

Samenvatting

R1R2R3R4R5R6R7R8R9


Samenvatting

182


Samenvatting

183

S

Depressie komt veel voor en is een van de meest invaliderende aandoeningen met een negatieve invloed op vele aspecten van het dagelijkse leven. Na herstel is de kans op terugval groot. De kans op het krijgen van een nieuwe depressie neemt bij elke episode toe; na 3 episodes is die kans maar liefst 90%. Ook tussen depressieve episodes door ervaren patiënten vaak depressieve klachten. Depressie kan daarom beter als een potentieel chronisch, levenslang probleem worden beschouwd in plaats van als een eenmalige gebeurtenis. Vanwege het chronische karakter is depressie, na lage rugpijn, wereldwijd zelfs de grootste veroorzaker van ziektelast (de totale hoeveelheid schade en/of ongemak als gevolg van een ziekte). Het behandelen van depressies in Nederland kost bijna een miljard euro per jaar. Alle spelers in de geestelijke gezondheidszorg (zoals zorgverleners, beleidsmakers, onderzoekers en zorgverzekeraars) zijn zich inmiddels bewust van de enorme ziektelast en de hoge kosten die met het chronische, terugkerende karakter van depressie gepaard gaan. Zij vragen zich voortdurend af hoe de zorg voor mensen met terugkerende depressies het beste kan worden ingericht; hoe de juiste zorg, op de juiste plek, op het juiste moment, aan de juiste patiënt en tegen aanvaardbare kosten kan worden aangeboden. Vast staat dat het onderbreken van het repeterende ritme van depressie cruciaal is om de ziektelast te beperken. Het voorkómen van terugval na herstel staat daarbij centraal. Op dit moment wordt in richtlijnen in eerste instantie geadviseerd om na herstel langdurig antidepressiva te gebruiken om terugval te voorkómen. Het langdurig gebruik van antidepressiva heeft echter vaak niet de voorkeur van patiënten, onder meer vanwege bijwerkingen en de daaruit voortvloeiende, gebrekkige therapietrouw. Onderzoek wijst uit dat psychologische interventies óók effectief kunnen zijn in het voorkómen van terugval.

Bij aanvang van dit onderzoek waren er belangrijke kennis-hiaten met betrekking tot depressie en het voorkómen van terugval door middel van psychologische interventies. In dit proefschrift proberen we een aantal van deze kennishiaten te dichten door antwoorden te formuleren op de volgende onderzoeksvragen:

1) Wat is de ziektelast van terugkerende depressies in vergelijking met eenmalige depressies?

2) Wat is de effectiviteit van bestaande psychologische interventies in vergelijking met gebruikelijke zorg en het continueren van antidepressiva, wat betreft het voorkómen van terugval in een depressie?

3) Wat is de kosteneffectiviteit van bestaande psychologische interventies in vergelijking met gebruikelijke zorg, wat betreft het voorkómen van terugval in een depressie?

4) Wat is de (kosten)effectiviteit van een minimaal begeleide zelfhulpinterventie in de eerste lijn in vergelijking met gebruikelijke zorg, wat betreft het voorkómen van terugval in een depressie?

R1R2R3R4R5R6R7R8R9


Samenvatting

184

In Hoofdstuk 2 (onderzoeksvraag 1) gaan we in op de ziektelast van eenmalige en terugkerende depressies in Nederland. We bekeken de ziektelast zowel vanuit een individueel perspectief als vanuit een populatie perspectief. Vanuit individueel perspectief werd de ziektelast voor de individuele depressieve patiënt vastgesteld (ziektelast= ziektegewicht * ziekteduur) terwijl vanuit populatieperspectief de ziektelast voor de totale groep mensen met depressies werd vastgesteld (ziektelast= ziektegewicht * ziekteduur* aantal patiënten met een terugkerende depressies). Voor deze studie hebben we gebruik gemaakt van de gegevens uit de eerste meting van de tweede ’Netherlands-Mental-Health-Survey-and-Incidence-Study’ (NEMESIS 2-studie; n=6.646; eenmalige depressieve episode n=115; terugkerende episode, n=246). We stelden vast dat vanuit individueel perspectief een eenmalige depressie de zwaarste ziektelast veroorzaakt. Echter, vanuit populatieperspectief veroorzaken de terugkerende depressies de grootste ziektelast. Deze bevindingen veranderden niet na correctie voor co-morbiditeit, d.w.z. voor aandoeningen die patiënten naast hun terugkerende depressie konden hebben zoals een angststoornis. Op basis van deze resultaten kunnen zorgverleners (die de belangen van een individuele patiënt centraal stellen) concluderen dat ze maximaal moeten inzetten op het voorkómen- of behandelen van een eenmalige depressieve episode. Echter, omdat terugval in een depressie zoveel voorkomt, kunnen beleidsmakers (die de belangen van grotere groepen binnen de gezondheidszorg centraal stellen) zich afvragen hoe ze de ziektelast voor terugkerende depressies het beste kunnen aanpakken. We concludeerden dat het belangrijk is om een verschil te maken tussen ziektelast vanuit individueel en populatieperspectief om misverstanden tussen zorgverleners en beleidsmakers te voorkómen met betrekking tot de allocatie van (schaarse) middelen.

In Hoofdstuk 3 (onderzoeksvraag 2) stellen we onszelf de vraag welke psychologische interventies zijn beschreven, die zich richten op het voorkómen van depressies bij mensen die minimaal 2 depressieve episodes hebben doorgemaakt. We geven een overzicht van de resultaten van bestaande gerandomiseerde en gecontroleerde studies tot en met mei 2014. De psychologische interventies werden vergeleken met gebruikelijke zorg en/of het gebruik van antidepressiva. We voerden een systematische zoekstrategie uit in de elektronische databases van Medline, Psychinfo, CINAHL, Embase en Cochrane. Studies werden geïncludeerd op basis van vooraf gedefinieerde selectiecriteria. De kwaliteit van de studies werd beoordeeld met behulp van een scoringslijst. Er werden 25 studies geïncludeerd met in totaal 2.055 deelnemers. De psychologische interventies waren gebaseerd op cognitieve (gedrags-) therapie (C(B)T), interpersoonlijke therapie (IPT) en op mindfullness gebaseerde cognitieve therapie (MBCT). We vonden dat preventieve psychologische interventies het risico op terugval in een depressie significant méér verlagen dan gebruikelijke zorg (RR=0.64) en het gebruik van antidepressiva (RR=0.83) over een periode van twee jaar.


Samenvatting

185

S

Bovendien stelden we vast dat het preventieve effect van een psychologische interventie groter is wanneer deze interventie direct volgt op een behandeling in de acute fase van de depressie.

In Hoofdstuk 4 (onderzoeksvraag 3) onderzoeken we of het additioneel aanbieden van preventieve psychologische interventies de kosteneffectiviteit van het bestaande Nederlandse gezondheidszorgsysteem voor terugkerende depressie verbetert. We gebruikten daarvoor een economisch model dat is ontwikkeld aan het Trimbos Instituut. Met het model is het mogelijk de Return on Investment (ROI) van een bepaalde interventie te bepalen. De ROI staat daarbij voor het rendement van de investering in een preventieve psychologische interventie. Voor de effectiviteit van de psychologische interventies maakten we gebruik van de resultaten uit Hoofdstuk 3. Andere aannames voor het model staan in Hoofdstuk 4 in detail beschreven. Als basisscenario gebruikten we gebruikelijke zorg die exact volgens de Nederlandse richtlijnen voor (terugkerende) depressie wordt toegepast, de ‘optimale gebruikelijke zorg’. Dit basisscenario vergeleken we met 4 andere scenario’s in termen van kosteneffectiviteit: 1) basisscenario pluscognitieve (gedrags-) therapie, 2) basisscenario plus op mindfullness gebaseerde cognitieve therapie, 3) basisscenario plus interpersoonlijke therapie en 4) basisscenario plus een hypothetische, minimaal begeleide zelfhulpinterventie. Omdat de effectiviteit van de hypothetische, minimaal begeleide zelfhulpinterventie niet bekend was, bepaalden we de minimaal benodigde risicoreductie die nodig is om te kunnen concurreren met de psychologische interventie die de hoogste ROI scoorde. Onze studie laat zien dat door uitbreiding van het basisscenario met cognitieve (gedrags-) therapie of met op mindfullness gebaseerde cognitieve therapie de kosteneffectiviteit van ons gezondheidszorgsysteem voor terugkerende depressie kan verbeteren. Voor het behalen van de meest competitieve ROI, zou de minimaal begeleide zelfhulpinterventie moeten leiden tot een relatieve risicoreductie van 0.71. In een latere fase van het onderzoek hebben wij in een gerandomiseerde studie binnen de setting van de Nederlandse gezondheidszorg, de risicoreductie van deze zelfhulpinterventie geëvalueerd (Hoofdstuk 6).

Op basis van de voorgaande hoofdstukken kunnen we concluderen dat preventieve psychologische interventies zoals C(B)T en MBCT en een hypothetische zelfhulpinterventie een interessante rol kunnen spelen bij het voorkómen van terugval in een depressie. Immers, deze interventies blijken effectief en in theorie kosteneffectief. In hoofdstuk 5 presenteren we het onderzoeksprotocol van een studie die we vervolgens uitvoerden om het effect van een psychologische interventie bij het voorkómen van terugval in een depressie te evalueren; de Parade-studie (Preventie-van-Recidiverende-Depressie in de Eerste Lijn).

De Parade-studie is een gerandomiseerd en gecontroleerd onderzoek met 2 groepen. Deelnemers in de controlegroep ontvingen gebruikelijke zorg (onder meer gebruik van

R1R2R3R4R5R6R7R8R9


Samenvatting

186

antidepressiva, psychologische interventie of geen behandeling). Deelnemers in de interventiegroep ontvingen gebruikelijke zorg uitgebreid met een Preventieve Cognitieve Therapie (PCT). Geschikte deelnemers hadden een voorgeschiedenis met twee of meer depressieve episodes en waren op het moment van inclusie hersteld, vastgesteld door middel van een telefonisch diagnostisch interview. Het protocol beschrijft een gedetailleerde beschrijving van de overige inclusiecriteria waaraan deelnemers moesten voldoen. Deelnemers werden door middel van een selectieprocedure uit de databases van zowel huisartsenpaktijken als GGZ-instellingen geselecteerd. Randomisatie naar de twee onderzoeksgroepen vond plaats op patiënt-niveau. De follow-up was 12 maanden. Tijdens de follow-up werd bij de deelnemers na 6 en 12 maanden een uitgebreid telefonisch interview afgenomen om eventuele terugval naar een depressie vast te stellen. De onderzoekers die het interview afnamen wisten niet welke behandeling de deelnemer kreeg en deelnemers werd gevraagd de uitkomst van hun randomisatie niet te bespreken met de onderzoeker. De primaire uitkomstmaat was terugval in een depressie na 1 jaar. Secundaire uitkomstmaten waren depressieve symptomen, kwaliteit van leven (gemeten met de SF-12 en EQ-5D), co-morbide psychopathologie (zoals angst en somatisatie) en zelfredzaamheid.

De Preventieve Cognitieve Interventie (PCT) bestond uit een zelfhulpinterventie met minimale begeleiding door een counsellor in de eerste lijn, genaamd ‘Tussen dip en droom’. Bij aanvang van de interventie ontvingen deelnemers een zelfhulpboek met 8 hoofdstukken. Elke week kon de deelnemer een hoofdstuk bestuderen en bijbehorende opdrachten maken. In het protocol is een beschrijving van de inhoud van de PCT terg te vinden. De deelnemer had wekelijks telefonisch contact met de counsellor gedurende maximaal 15 minuten. Dit kon zowel een praktijkondersteuner Geestelijke Gezondheidzorg (POH-GGZ) uit de huisartspraktijk zijn als een klinisch psycholoog, beiden getraind door een ervaren psycholoog die de interventie heeft ontwikkeld. De counsellor begeleidde de deelnemers door de 8 weken durende interventie door te sturen en te motiveren, zonder op de inhoud van de interventie in te gaan.

Het aanbieden van een zelfhulpinterventie met minimale begeleiding in de eerste lijn leek om meerdere redenen interessant. Uit onderzoek blijkt dat zelfhulp in de acute fase van een depressie even effectief is als een individuele face-to-face interventie. Omdat ook uit onderzoek blijkt dat enige begeleiding bij psychologische interventies voor depressie noodzakelijk is voor het slagen ervan, werd een counsellor uit de eerste lijn ingeschakeld. Op deze manier hoefde geen beroep te worden gedaan op [schaarse] tweedelijns therapeuten. Verder sluit de inbedding van de zelfhulpinterventie in de eerste lijn goed aan bij het terugkerende, chronische karakter van depressie. Tenslotte ziet de huisarts veel patiënten met een depressie of met depressieve gevoelens. Tot nu toe was onbekend of een minimaal begeleide zelfhulpinterventie in de eerste lijn effectiever is dan de gebruikelijke zorg die vandaag de dag in de eerste lijn wordt geboden aan patiënten met terugkerende depressies.


Samenvatting

187

S

In Hoofdstuk 6 (onderzoeksvraag 4) presenteren we de resultaten van de Parade-studie naar het effect van een minimaal begeleide psychologische zelfhulpinterventie. Er werden 248 patienten geïncludeerd waarvan de helft naar de interventiegroep werd gerandomiseerd (n=124) en de andere helft naar de controlegroep (n=124). De zelfhulpinterventie zorgde voor minder terugval in een depressie vergeleken met gebruikelijke zorg na 1 jaar (respectievelijk 36% en 50%). Tevens werd in de interventiegroep na 1 jaar een afname in depressieve symptomen en toename in kwaliteit van leven gevonden (gemeten met de EQ-5D) in vergelijking met de controlegroep. De interventie had geen effect op co-morbide psychopathologie, zelfredzaamheid en kwaliteit van leven gemeten met de SF-12.

In Hoofdstuk 7 (onderzoeksvraag 4) rapporteren we de kosteneffectiviteit van de minimaal begeleide zelfhulpinterventie in vergelijking met gebruikelijke zorg over een periode van 1 jaar. Immers, wanneer een dergelijke interventie in de praktijk wordt geïmplementeerd, is het van belang om vast te stellen hoeveel schaarse middelen vrij moeten worden gemaakt die ook voor andere doeleinden zouden kunnen worden ingezet. We hebben de kosteneffectiviteit zowel vanuit maatschappelijk perspectief als vanuit gezondheidszorgperspectief berekend. Vanuit het perspectief van de gezondheidszorg worden alleen de gezondheidszorg-gerelateerde kosten meegenomen (zoals kosten van medicatie, bezoeken aan specialist of huisarts) terwijl het vanuit maatschappelijk perspectief relevant is om álle kosten mee te nemen in de analyses (zoals gezondheidszorg-gerelateerde kosten en kosten als gevolg van verminderde arbeidsproductiviteit). We vonden dat na 1 jaar de gemiddelde, totale maatschappelijke kosten in de interventiegroep €2.114,- euro hoger lagen dan in de controlegroep. De interventiegroep had met name meer kosten als gevolg van werkverzuim en bezoek aan specialisten. Wanneer we deze hogere kosten afzetten tegen het effect concluderen we dat de interventie niet kosteneffectief is. Deze conclusie gaat eveneens op wanneer de kosteneffectiviteit wordt vastgesteld vanuit het perspectief van de gezondheidszorg.

In Hoofdstuk 8 (Discussie) wordt tot slot een overzicht gegeven van de belangrijkste bevindingen van dit proefschrift, gevolgd door een reflectie op deze bevindingen, een bespreking van methodologische kwesties en aanbevelingen voor verder onderzoek.

De resultaten van de Parade-studie bevestigen dat het risico op terugval na herstel op dit moment erg groot is. Maar liefst de helft van de deelnemers die gebruikelijke zorg ontving, viel binnen 1 jaar terug in een depressie.

De resultaten van ons literatuuronderzoek naar de effectiviteit van psychologische interventies sluiten aan bij eerder onderzoek dat aantoont dat C(B)T en MBCT effectief zijn in het voorkomen van terugval. Echter, eerder onderzoek toont aan dat deze interventies alleen effectief zijn bij mensen die 4 of meer depressieve episodes hebben doorgemaakt.

R1R2R3R4R5R6R7R8R9


Samenvatting

188

Op basis van onze resultaten, zowel die van het literatuuronderzoek als die van de Parade-studie, kunnen C(B)T en MBCT worden geadviseerd aan alle patiënten, ongeacht het aantal eerder doorgemaakte episodes.

Het positieve resultaat met betrekking tot de kosteneffectiviteit van onze interventie dat we vonden met ons model (Hoofdstuk 4), komt niet overeen met het negatieve resultaat dat we vonden in de kosteneffectiviteitsanalyse in de Parade-studie (Hoofdstuk 7). Omdat onduidelijk is wat de oorzaak is van de hogere kosten in de interventiegroep ten opzichte van de controlegroep, is ook onduidelijk wat de oorzaak is van het verschil in resultaten tussen de twee studies.

Het bleek ingewikkeld om patiënten met 2 of meer depressie te includeren in onze studie. Zo is het onmogelijk om in elektronische patiëntendossiers patiënten te selecteren die 2 of meer depressies hebben gehad en bovendien op dat moment depressievrij zijn. Er moest daarom een veel grotere groep aangeschreven worden dan alleen de patiënten die daadwerkelijk aan deze criteria voldeden. Daarnaast bleek tot onze verrassing het animo om mee te doen aan een onderzoek naar terugvalpreventie onder de potentieel geschikte deelnemers laag.

Als aanbeveling voor verder onderzoek adviseren we om een langere follow-up duur te gebruiken dan de 12 maanden in onze studie. Alleen dan wordt duidelijk hoe het effect van de interventie zich ontwikkelt in de tijd en krijgen we meer grip op het verloop van de kosten. Ook zou het behandeldoel van de interventie kunnen worden aangepast; in plaats van het voorkómen van een terugval zouden doelen als werkhervatting of deelname aan maatschappelijke activiteiten (werk/sport) kunnen worden overwogen. Daarnaast zou er meer aandacht kunnen worden besteed aan de redenen voor potentiële deelnemers om wel of niet deel te nemen aan een onderzoek naar terugvalpreventie.

Tot slotDit proefschrift laat zien dat terugkerende depressies een grote ziektelast veroorzaken. Het trekt de belangrijke conclusie dat psychologische interventies een welkom alternatief kunnen zijn voor het langdurig gebruik van antidepressiva. Daarnaast blijkt een psychologische zelfhulpinterventie met minimale begeleiding in de eerste lijn effectief. Echter, de patiënten die in aanmerking kwamen voor terugvalpreventie konden moeilijk worden bereikt, bleken weinig gemotiveerd en de zelfhulpinterventie bleek in de praktijk niet kosteneffectief.

Het beantwoorden van de vraag ‘hoe de juiste zorg, op de juiste plek, op het juiste moment, aan de juiste patiënt en tegen aanvaardbare kosten’ kan worden aangeboden, is nog niet beantwoord. Het voorkómen van terugval in een depressie (‘Keeping the clouds away’) is daarom een onderwerp dat de komende jaren voortdurend hoog op de agenda van zorgverleners, beleidsmakers, onderzoekers en zorgverzekeraars moet staan.

Acknowledgement

R1R2R3R4R5R6R7R8R9


Acknowledgement

190


Acknowledgement

191

A

Als ik het zou mogen samenvatten zijn de afgelopen 4,5 jaar ‘dynamisch’ geweest. Ik ben verhuisd, getrouwd, heb twee zwangerschappen doorlopen en ben boven alles, moeder geworden. Hoewel geluk en plezier de boventoon hebben gevoerd is er ook een aantal moeilijke momenten geweest met als dieptepunt het overlijden van mijn zwager. Mijn werk is in deze jaren een stabiele, overzichtelijke en fijne factor geweest waar ik mijn denkwerk goed in kwijt kon. Bij de start van mijn promotieonderzoek kreeg ik een project in handen dat ik op mijn eigen manier en in mijn eigen tempo heb mogen invullen en afronden. Ik ben het VUmc, mijn promotoren en mijn copromotor heel dankbaar voor hun flexibiliteit en de ruimte die ze mij daarbij hebben gegeven. Die zijn voor mij van onschatbare waarde geweest.

Dan nu de rest van het dankwoord. Hold your horses, een bladzijde meer of minder maakt nu ook niet meer uit. Ik ben er eens uitgebreid voor gaan zitten.

Promotoren en co-promotorProfessor Harm van Marwijk, beste Harm, na een baan als medisch adviseur bij Pfizer kwam ik bij jou en de onderzoekswereld van het VUmc terecht. Ik had met jou als dagelijks begeleider veel contact. Zo geordend als mijn hoofd en bureau zijn, zo onoverzichtelijk leken soms jouw hoofd en bureau. Kwam ik met één vraag binnen, stond ik met vijf ‘crazy out of the box’ vragen weer buiten. Je bent een vrolijk en optimistisch persoon met wie het altijd leuk praten is. Zelfs op het inclusie-dieptepunt hield jij de moed erin, knap vind ik dat. Naast de Parade-perikelen hebben we het gehad over het ouderschap, dokter zijn en het inrichten van je leven op je eigen manier. Jouw stap naar Manchester past helemaal binnen dat laatste. Ik hoop dat daar alles voor je op z’n plek valt. Bedankt, en ik houd graag contact met je.

Dr. Anneke van Schaik, beste Anneke, hoe fijn was het dat jij in mijn promotiecommissie zat: jouw nuchtere en niet-oordelende blik heeft me op belangrijke momenten erg gesteund. Ik herinner me een goed gesprek met jou toen ik in de knel zat met het combineren van onderzoek, gezin en het eventueel starten van de huisartsenopleiding. Jij bent echt een mensenkenner. Jouw inhoudelijke inbreng, met name in de laatste schrijffase van het onderzoek was waardevol. Dankjewel voor alles!

Professor Henriëtte van der Horst, beste Henriëtte, in het begin van dit promotietraject hebben we vooral contact gehad tijdens het periodieke overleg rondom de Parade-studie. Als je jou een concept-artikel stuurt komt het op heel korte termijn weer terug met alleen maar bruikbaar commentaar. In de periode rond mijn zieke zwager, die samenviel met de laatste maanden van mijn proefschrift, hebben we meer persoonlijk contact gekregen. Het was fijn om te merken dat je zo betrokken was. Veel dank voor alles.

R1R2R3R4R5R6R7R8R9


Acknowledgement

192

Professor Filip Smit, beste Filip, ik zou willen zeggen: de ideale promotor. Een echte ‘Professor’ bij wie je goed voor de dag wil komen. Die geen genoegen neemt met ‘ok’ als het eigenlijk ‘perfect’ moet. Als je vond dat ik iets goed had gedaan strooide je met complimenten. Als ik effies had zitten slapen kreeg ik het ook te horen. Ik houd ervan! Dankjewel voor al je tijd en moeite die je hebt gestoken in mijn professionele ontwikkeling en mijn proefschrift. Als jij er niet was geweest was, was ik niet trots geweest op het resultaat.

Leescommissie Professor Cuijpers, professor Evers, professor Beekman, dr. Blankenstein en dr. Jeurissen, graag wil ik jullie bedanken voor de bereidheid om het manuscript te beoordelen en deel te nemen aan de oppositie. ‘Zonder oppositie geen promotie’, zeg ik altijd maar. Het is dankbaar om te merken dat er mensen zijn die het onderwerp net zo leuk en belangrijk vinden als ik.

Dear professor Tylee, we met in London after my presentation about recurrent depression and the Parade-study at King’s College in 2015. Thank you for reviewing my thesis and making my public examination an international happening.

Medewerkers Parade-studieOnderzoeksassistente Sandra Kersten, lieve Sandra, toen je kwam solliciteren voor de functie van onderzoeksassistente zag ik het meteen helemaal zitten met je. Om huisartsen te vragen/overtuigen/smeken mee te doen aan de Parade-studie hebben we samen door de sneeuw gereden van Enschede tot aan Hoofddorp. We hebben gebeld, gemaild, ge-appt, gevouwen, geprint, gestickerd, ge-etiketteerd, gebrainstormd en heel veel overlegd. We hebben het echt samen gedaan en dat ging top. Je bent natuurlijk veel te lief voor deze wereld en gelukkig technisch veel beter dan ik. Je verdient het helemaal om nu te gaan promoveren. Jouw toekomstige zoon zal trots zijn! Als jij gaat promoveren ben ik er graag bij.

Dr. Judith Bosmans, beste Judith, het was supertof om jou in het projectteam te hebben, goeroe op het gebied van kosteneffectiviteit als jij bent. Daarnaast hebben we het gehad over onze gezinnen, de opvoeding en wat een mens op allemaal op z’n bordje kan krijgen. Ik bewonder jouw arbeidsethos en hoop je nog een keertje tegen te komen. Dankjewel voor je hulp!

Professor Claudi Bockting, beste Claudi, als ontwikkelaar van de interventie ben je nauw betrokken geweest bij de Parade-studie. Bedankt voor je directe betrokkenheid en het leuke contact dat we hebben gehad. De interventie als zelfhulp blijkt daadwerkelijk effectief, mooier kan het niet.


Acknowledgement

193

A

Dr. Evelien van Valen, beste Evelien, dankjewel voor het verzorgen van de trainingen. Mede door jouw enthousiasme was de inzet van onze therapeuten groot.

Co-auteursDr. Gemma Kok, dr. Joran Lokkerbol, dr. Margreet ten Have, dr. ir. Ron de Graaf, prof. Pim Cuijpers; bedankt voor jullie bijdrage aan de artikelen in dit proefschrift.

Deelnemers Ik bedank de 248 deelnemers die hebben deelgenomen aan de Parade-studie. Ieder van jullie heeft telefonisch zijn/haar verhaal verteld. Ik heb ervaren dat het leven, dat voor mij een mooi en vanzelfsprekend gegeven is, voor anderen een dagelijks gevecht kan zijn. Van de verhalen van mensen krijg ik nooit genoeg. Het heeft me gesterkt in de keuze voor het doktersvak.

Deelnemende zorgverlenersHet zijn uiteindelijk de zorgverleners die met hun expertise en betrokkenheid de depressieve patiënten kennen en behandelen. Dank voor al jullie inzet en de gezamenlijke inspanning om de zorg voor deze groep te verbeteren: Arts en Zorg Den Haag, huisartsen Welling en van Lingen, GZC Steenpoort, HAP Ilpendam (Esther Broekhuizen), GGZ Leiden (Emke Osinga), Prezens (Annemieke van Straten), GGZ Noord Holland Noord (Marty Dijkstra), Bosgroep (Christine Brouwer-Dudok de Wit), Indigo Brabant (Co Wildeboer Schut), GZC Waterlinie (Norbert Wijtenburg), HAP Spijkenisse (Kees Vos), GZC Haarlemmermeer (Teun Preijde).

Therapeuten Graag dank ik de enthousiaste therapeuten die de zelfhulpinterventie van de Parade-studie (minimaal) hebben begeleid; Denise Ras, Eline Eigenhuis, Etoes Maggé, Katja Boot, Lieke van den Reijen, Maryse Cnossen, Lisette Bos-Overdijk, Liselotte de Mooij, Carlijn Bult, Klazien van der Sloot, Leonie van Loosbroek, Marije Uffen, Lilian Zwaneveld, Cilia van Hal, Tonny van Savoyen - Oskamp, Sandra van Burk, Rachel Janssen, Kristina Hagels, Rozemarijn van de Kolk, Ignas Straatman, Eveline Paauw, Els Dozeman en Helen Volmer. Thuur Smet, dankjewel voor het maken van bestanden met namen en gegevens van potentiële deelnemers in de GGZ. Wij hebben maar tweemaal email-contact gehad maar jij hebt met jouw inspanningen mijn onderzoek, dus mijn proefschrift, dus mijn leven gered.

R1R2R3R4R5R6R7R8R9


Acknowledgement

194

AFDELING D5, MEDISCHE FACULTEITDr. Hans van der Wouden, beste Hans, zoals jij zie je ze zelden. Altijd aanwezig, altijd de deur open, meedenken met m’n vragen, opgewekt en nooit denken in je eigen belang. Je bent een enorme aanwinst voor de gang en voor mij een schoolvoorbeeld van integer werken. Ik zou bijna zeggen dat er van jouw soort wel meer mochten zijn maar dan zitten we met een explosie aan dochters...

Secretaresse Loes Haan, beste Loes, ouwe grapjas die je er bent, dankjewel voor al je hulp als die nodig was. Je hebt alles helemaal op orde en dat werkte top.

Kamergenoten, mijn mannen Jeroen en Pim, beste Jeroen, 3 jaar hebben we samen op 1 kamer gezeten. De perfecte combi van praten, grapjes, inhoudelijke overleggen, stiltes en bespreken van belangrijke en minder belangrijke zaken in het leven. En dat allemaal niet teveel maar zeker ook niet te weinig. De D-527 symbiose was buitenaards. Je sportprestaties zijn indrukwekkend en het vaderschap staat je goed.Beste Pim, het was goed dat jij pas het laatste half jaar in beeld kwam. Met jou wilde de balans wel eens wat doorslaan naar het bespreken van de minder belangrijke zaken in het leven. Ik heb er de vinger nog niet op kunnen leggen maar op een of andere manier lokt jouw verschijning bij mij vreemde geluiden en opmerkelijke performances op. Als jij nou een huisartsenpraktijk in Amsterdam-Noord overneemt, dan neem ik daar over 4 jaar alle oudjes voor m’n rekening. Win-win!

Gang D-5, in het bijzonder Nikki, Madelon, Annemarie, Marloes, Floor, Hanneke, Kate, Daniëlle, Daniëlle, Anne, Celine en Lidy. Bij het Amerikaans georiënteerde Pfizer werd me geleerd dat je op z’n minst ‘one best friend at work’ moet hebben. In die zin was D-5 een walhalla! Het is gewoon erg fijn om je verhalen te delen met meisjes die dezelfde bizarre werkplek (lees: de D-5-gang) hebben. Ik viel er natuurlijk wel een beetje buiten zo voorbij de klapdeuren. Het onthaal ter hoogte van de wc’s was desondanks altijd hartelijk. Er wordt lekker gemopperd en ook veel gelachen. De dagelijkse grote vraag was natuurlijk: waar is iedereen?? Ik kijk terug op een bijzondere tijd met jullie.

OOK BELANGRIJKPfizer, lieve oud-collega’s, Carlita, Linda, Dorien en Gregory. Ik denk af en toe terug aan die geweldige tijd met jullie. Keihard werken maar wel onder nogal fijne omstandigheden. Superleuk dat we af en toe nog contact hebben.


Acknowledgement

195

A

Buurvrouwen, lieve Heleen, Vanessa, Paulien, Nynke, Annemarie en Yvonne. Bedankt voor de gezelligheid en interesse. Het is fijn thuiskomen in de Mr. Six!

Vriendinnen, Karianne, Marieke, Camilla, Kirsten. Er zijn nogal wat cruciale momenten in het leven van ons, jonge vrouwen. Het is fijn om zowel deze, als de volslagen onbelangrijke momenten te bespreken. En nog eens te bespreken, en nog eens… En alsof het niet genoeg is; gewoon nog eens! Zoals alleen wij jonge vrouwen dat kunnen. Ik vind jullie de allerleukste! In dit boekje ter info wat informatie over m’n onderzoek.

Lieve paranimf Karianne, wij hebben eigenlijk weinig woorden nodig maar gebruiken er desondanks heel veel. Thanks dat je zo dicht betrokken wilt zijn bij dit hele gebeuren. Vriend Stephan, de 4e broer, gewoon altijd leuk met jou. Niks meer aan doen.

Vriendinnen uit Roermond, Marieke, Kim, Anka, Saskia, Nynke, Heike, Haikje en Eefje. Onze halfjaarlijkse reünie is een hoogtepunt in mijn agenda. Samen met jullie heb ik de humor ontdekt. Aan mijn jeugd in Roermond zit een gouden randje.

FAMILIENichtje Suus, wat jammer dat je zo ver weg zit en mijn promotie gaat missen. We missen veel van elkaars leven, het is niet anders. Ik haat bellen maar met zo af en toe een mega-email houden we elkaar toch een beetje op de hoogte.

Schoonzussen, lieve 1) Angeline, 2) Anke, 3) Dieuwke, 4) Eva, 5) Fleur en 6) Jody. Stuk voor stuk bloedeigen, bloedjes van schoonzussen. Kom daar nog maar eens om tegenwoordig! Druk is het zeker. Er wordt altijd wel verhuisd, van baan gewisseld, een kinderverjaardag gevierd, of een nieuw kapsel aangemeten. Het is dus hard werken om te feliciteren, te troosten, te bewonderen, te steunen of succes te wensen. Fijn dat jullie er op 15 april ook allemaal zijn. Het is bijzonder!

Zwagers, lieve Jurgen en Bas, het woord zwager klinkt voor mij altijd als een soort ver-weg-aangetrouwde-neef-Henk maar jullie zijn gewoon twee lieve, grappige en hardwerkende jongens. Grote ontbrekende in mijn rijtje zwagers is natuurlijk de GVR Jaap. Het blijft onwerkelijk dat hij er niet meer is, hij wordt enorm gemist.

Schoonouders, lieve Pieter Jan en Jacqueline. We hebben elkaar in de afgelopen 2 jaar in goede maar ook hele verdrietige situaties nog beter leren kennen. Het is fijn om van heel dichtbij mee te maken hoe jullie gezin in elkaar zit. Het is gezellig op de Sweelinckstraat en alles is goed. Dank voor jullie interesse in mijn werkzaamheden en jullie liefde voor onze kindjes.

R1R2R3R4R5R6R7R8R9


Acknowledgement

196

Broers, lieve Erik, Paul en Pieter. Ik vind het een interessant gegeven dat ik voor mijn allergrootste trots – drie broers- niks heb hoeven doen. Het wonen onder één dak en het hebben van dezelfde ouders schept een onvoorwaardelijke band. Door gesprekken met jullie en een nauwkeurige observatie van de acties die jullie uithaalden heb ik veel inzicht gekregen in het mannelijk brein. Als ik echt iets belangrijks met jullie wil bespreken is daar maximaal 4 min 47 seconden ruimte voor. Daarna vliegen de grappen en grollen me alweer om de oren. Ik weet nu: dat is geen desinteresse, dat is onkunde! Het gevoel dat ik jullie altijd kan bellen is heerlijk en ik ben trots op jullie (uiteindelijke) keuze voor jullie vrouwen/vriendinnen.

Pieter was de getuige op mijn bruiloft. Pistolen-Paultje mijn logische paranimf op 15 april. Voor Erik verzin ik t.z.t. nog een leuk event. De humor staat altijd centraal, dankjewel. Ik hou heel veel van jullie!

Ouders, lieve papa en mama, jullie hebben het voor elkaar gekregen om vier kinderen groot te brengen die met zelfvertrouwen en veel plezier in het leven staan. Jullie werkten allebei hard en stelden alles in het teken van ons gezin. Jullie stonden achter mijn keuze voor economie in Amsterdam en vier jaar later achter mijn keuze voor geneeskunde. Het stellen van grenzen en het geven van ruimte is de ultieme uitdaging in een goeie opvoeding lijkt me zo. Lieve mama, 1 maand voordat ik in Amsterdam op kamers ging zijn wij als twee padvinders op verkenning gegaan. Het Rembrandtplein, de VU, de wallen... Die rondgang en jouw interesse in alles wat ik daarna in Amsterdam en omstreken heb meegemaakt zijn voor mij zo belangrijk geweest. Jouw Duracell-batterijtje raakt echt nooit leeg! Ik ben nu al jaloers op de gezondheid, schoonheid en energie waarmee jij dit jaar 70 gaat worden.

Lieve papa en mama, de laatste tijd gaat het vaker over ouder worden. Als het zover is, zal ik in mijn (verpleeg)huis een gezellig kamertje voor jullie vrijhouden en de gebitjes dagelijks poetsen. Ik doe het hoor! Kon ik jullie maar voor altijd vasthouden. Ik hou heel veel van jullie, bedankt voor alles.

DE KERNEr komt een moment waarbij je je aan iemand voorstelt en niet langer vertelt uit welk gezin je komt maar welk gezin je zelf gestart bent. Het verleggen van die focus vond ik aanvankelijk bizar. Toch wende het snel en draait mijn leven nu om mijn lieve, eigen gezin.Hugo; mijn vrolijke, slimme kleuter. Vesper; mijn ondeugend, knuffelend Bond-girl peutertje. Lieve Quirijn: vanaf 2005 delen we ons leven en is het feest, we hebben samen al veel geweldige momenten meegemaakt. In 2015 hebben we veel voor onze kiezen gekregen. Na zo’n jaar krijgt de belofte om voor altijd bij elkaar te blijven pas echt betekenis. Bedankt


Acknowledgement

197

A

voor jouw eeuwige support, met name bij het afmaken van dit proefschrift en de start van mijn opleiding. Je gaf me maanden lang alle ruimte. Het geeft mij het vertrouwen dat ik de voortdurende strijd tussen mijn ambities en mijn moederkloekgevoel goed aankan. Ik hou heel veel van jou en ben apetrots op hoe wij het leven met z’n tweeën inrichten. Als jij thuis komt is het feest en is er rust. Jij, Hugo en Vesper zijn alles voor mij.

Nu ik zoveel heb gelezen over depressie, met mensen heb gesproken over depressie en de levensverhalen van depressieve mensen heb gehoord, zou ik bijna willen zeggen: ik kan er een boek over schrijven. Dat treft! Over en uit.

List of publications

R1R2R3R4R5R6R7R8R9



200



201

L

INTERNATIONAL PUBLICATIONSBiesheuvel-Leliefeld KE, Kersten SM, van der Horst HE, van Schaik A, Bockting CL, Bosmans JE, Smit F, van Marwijk HW. Cost-effectiveness of nurse-led self-help for recurrent depression in the primary care setting: design of a pragmatic randomised controlled trial. BMC Psychiatry. 2012 Jun 7;12:59.

Biesheuvel-Leliefeld KE, Kok GD, Bockting CL, Cuijpers P, Hollon SD, van Marwijk HW, Smit F. Effectiveness of psychological interventions in preventing recurrence of depressive disorder: meta-analysis and meta-regression. J Affect Disord. 2015 Mar 15;174:400-10.

Dijkstra-Kersten SM, Biesheuvel-Leliefeld KE, van der Wouden JC, Penninx BW, van Marwijk HW. Associations of financial strain and income with depressive and anxiety disorders. Journal of Epidemioly and Community Health. 2015 Jul;69(7):660-5.

Biesheuvel- Leliefeld KEM, Kok G, Bockting CLH, de Graaf R, ten Have M, van der Horst HE, van Schaik A, van Marwijk HWJ, Smit F. Non-fatal disease burden for subtypes of depressive disorder: population-based epidemiological study. Accepted for publication in BMC Psychiatry.

NATIONAL PUBLICATIONSKarolien Biesheuvel-Leliefeld, Harm van Marwijk. Huisarts en Wetenschap, jaargang 2013, nummer 2:95-95 Screenen op depressie voorlopig niet aan de orde.

SUBMITTED FOR PUBLICATIONBiesheuvel- Leliefeld KEM, Lokkerbol J, Smit F. Cost-effectiveness of preventing depressive recurrences by psychological interventions; a population health economic modelling study.

Biesheuvel- Leliefeld KEM, Dijkstra-Kersten SM, van Schaik A, van Marwijk HWJ, Smit F, van der Horst H, Bockting CLH. Effectiveness of a supported self-help for recurrent depression: a randomised controlled trial in primary care.

Biesheuvel- Leliefeld KEM, Bosmans JE, Smit F, Bockting CLH, van Schaik A, van Marwijk HWJ, van der Horst HE. Cost-effectiveness of a supported self-help for recurrent depression in primary care.

R1R2R3R4R5R6R7R8R9



202

ADDITIONAL PUBLICATIONSLeliefeld K, Van Der Sluijs H, Van Der Haven I. Isolated congenital anterolateral bowing of the fibula: a case report with 24 years follow-up. Acta Orthop Belg. 2009 Dec;75(6):842-6.

Klooker TK, Leliefeld KE, Van Den Wijngaard RM, Boeckxstaens GE. The cannabinoid receptor agonist delta-9-tetrahydrocannabinol does not affect visceral sensitivity to rectal distension in healthy volunteers and IBS patients. Neurogastroenterol Motil. 2011 Jan;23(1):30-5.

About the author

R1R2R3R4R5R6R7R8R9


About the author

204


About the author

205

A

Karolien Biesheuvel-Leliefeld was born on September 14th, 1981 in Heerlen, The Netherlands. She is third in a family with three brothers (Erik ‘78, Paul ’79 and Pieter ’86). After graduating secondary school at Bisschoppelijk College Schöndeln in Roermond (Gymnasium 1999), she studied Economics at the Faculty of Economics and Business Administration of VU University Amsterdam. In the meantime she worked as a student-assistant at the Economic and Social Institute (ESI). She obtained her master’s degree in Business Administration in 2003, after which she travelled through South-America for four months. Upon return she began her medical training at the Academic Medical Centre of the University of Amsterdam. During this study she worked part-time as an associate in the Marketing Intelligence Team (MIT) of Delta Lloyd where she worked on a project aimed at the efficient gathering of data of the Dutch insurance-market. She moved to New York for three months where she worked in a lab at the Department of Infectious Diseases of the Presbyterian Hospital of Columbia University. In December 2007 she started her internships and completed these in December 2009 at the Department of General Medicine of the Slotervaart Hospital in Amsterdam. In January 2010 she started working as Medical Advisor for neuropathic pain at pharmaceutical company Pfizer. After 1,5 years she started a PhD project at the department of General Practice and Elderly Care medicine of the VU University medical centre in Amsterdam named ‘Keeping the clouds away; prevention of relapse and recurrence in major depressive disorder’. During her PhD she teached and supervised medical students and followed epidemiological- and teaching courses. She entered into a three-year specialist medical training to become an elderly care physician at the VU University Medical Centre Amsterdam in September 2015.

Karolien married Quirijn Biesheuvel in February 2012. She has a son (Hugo, 2012) and a daughter (Vesper, 2014).


Prevention of relapse and recurrencein major depressive disorder

Karolien Biesheuvel - Leliefeld

Uitnodigingvoor het bijwonen van de openbare verdediging van

mijn proefschrift getiteld

Keeping the cloUds awayPrevention of relapse and

recurrence in major depressive disorder

Vrijdag 15 april 2016, 11.45 uur Auditorium

Vrije Universiteit De Boelelaan 1105, Amsterdam

Na afloop van de verdediging bent u van harte welkom in

café The Basket op de campus van de VU

Karolien BiesheuvelMeester Sixlaan 261181 PK Amstelveen

06 - [email protected]

ParanimfenPaul Leliefeld

[email protected] - 24884706

Karianne van der Weijden [email protected]

06 - 24594099

Karolien Biesheuvel - LeliefeldKeeping the clouds away

Depression is among the most disabling disorders and negatively affects many aspects of life. It is associated with a high risk of recurrence. Of all people with a first episode, more than half experiences such a recurrence. Treating depression in the Netherlands costs almost a billion euro (€966 million) per year. These substantial economic consequences of depression are mainly due to its recurrent nature. An important potential area of improvement in care for people with depression is the prevention of recurrence. The most commonly used strategy is the continuation of antidepressant medication. However, due to possible side effects and non-adherence issues, continuing antidepressant medication may not always be the preferred option. Psychological interventions could be a valuable alternative. In order to improve clinical outcomes it is highly relevant to study the prevention of recurrent depression using psychological interventions in primary care. Therefore, Karolien Biesheuvel and her colleagues carried out several studies, including a randomised controlled trial in The Netherlands. The research questions were:

1) What is the burden of disease of recurrent depression compared to single episode depression?

2) What is the effectiveness of existing psychological interventions compared both to usual care and the continuation of antidepressant medication, to prevent relapse and recurrence in recurrent depression?

3) What is the cost-effectiveness of existing psychological interventions to prevent relapse and recurrence in recurrent depression, compared to enhanced usual care?

4) What is the (cost-)effectiveness of a psychological self-help intervention in primary care, for the prevention of relapse and recurrence in recurrent depression?

Proefschrift Biesheuvel

Documents

Transcript of Proefschrift Biesheuvel