Outcomes generated by patients with rheumatoid arthritis: how important are they?

Copyright © 2005 John Wiley & Sons, Ltd 3: 131–142 (2005)

MUSCULOSKELETAL CARE3: 131–142 (2005)

DOI:10.1002/msc.3

Mai

n a

rtic

les

131

Main article

Outcomes generated by patients withrheumatoid arthritis:How important are they?Sarah Hewlett PhD MA RGN1, Maggie Carr MSc RGN2, Sarah Ryan RGN, PhD3,John Kirwan BSc MP FRCP1, Pam Richards1, Alison Carr PhD4, and Rod HughesMA MD FRCP2

1Academic Rheumatology Unit, University of Bristol, 2Rheumatology Department, StPeter's Hospital, Chertsey, 3Rheumatology Centre, Heywood Hospital, Stoke-on-Trent, 4Nottingham University, Nottingham

AbstractBackground: It has been shown previously that patients with rheumatoid arthritis (RA) cangenerate a wide range of outcomes that they consider important in treatment. It is notknown if these outcomes are generally important in the wider RA patient community. Objectives: (1) To examine whether recent patient-generated outcomes are generalizablewithin a wider RA population; (2) to assess the relative importance of each outcome; and(3) to explore whether any important outcomes have been omitted.Methods: A questionnaire, listing 23 outcomes previously generated by RA patients, wasdistributed through three rheumatology centres in the UK. Patients gave an importancescore to each outcome (0–3), selected their top three most important outcomes, and thenlisted any outcomes of personal importance that were missing.Results: 323 questionnaires were returned (65%). All outcomes were deemed important.Independence, pain, and mobility were most frequently selected by patients in their top threeoutcomes but were not chosen by 61–66% of patients. The next most commonly chosenoutcomes related to feeling well and fatigue. Factor analysis revealed six reasonably distinctgroupings: general well-being (11.9% explained variance), day-to-day functioning(10.6%), emotional and psychological well-being (10.6%), social role and confidence(10%), physical symptoms (9.5%) and medication issues (7.9%). Conclusion: Outcomes generated by patients as important in RA, are generalizable andinclusive. The most important (independence, pain and mobility) are routinely treated andmeasured. The next most important (feeling well, fatigue) are infrequently addressed anddeserve urgent consideration for measurement, treatment and research. Copyright © 2005

John Wiley & Sons, Ltd.

Key words: Outcomes, rheumatoid arthritis, values, patient-generated, fatigue

MSC 3(3) 9/1/05 2:20 PM 31

Introduction

The majority of clinical trials in rheumatoid arthritis (RA) routinely measure pain anddisability and these, along with joint symptoms and patient opinion of overall diseaseactivity, form the cornerstone of the American College of Rheumatology (ACR) coreset of internationally accepted outcomes (Felson et al., 1993). However, there isevidence that arthritis patients’ ratings of the severity of disability, pain andpsychological symptoms can differ from those of clinicians (Berkanovic et al., 1995;Suarez-Almazor et al., 2001; Kwoh et al., 1992). It is likely that patients assess theseverity of symptoms within the context of personal consequences or importance.

Patients’ beliefs about the importance of symptoms also differ from those ofclinicians. This is shown both for the importance of physical symptoms and for theimportance of psychological well-being, and occurs in RA and in other chronicillnesses (Hewlett et al., 2001; Kwoh and Ibrahim, 2001; Rothwell et al., 1997). Thisfinding has led to the exploration and development of measurement scales whichallow RA patients to express individual concerns about outcomes, for example, byprioritizing professionally-generated outcomes (Meenan et al., 1992), by identifyingimportant effects in previously-determined patient-generated outcomes (Carr,1996), or by giving personal importance values which are then used to weightoutcomes (Hewlett et al., 2002). Loss of activities considered important by womenwith RA has been shown to be a predictor of depression (Katz and Yellin, 1995) andit has been suggested that approaches to outcome measurement need to be moreopen to patient priorities, perceptions and values (Kvien and Heiberg, 2003).Asking patients to propose their own individual outcomes (Tugwell et al., 1987,Ruta et al., 1994) is potentially useful, but produces many different outcomes,making comparison between different trials difficult. The use of a core set ofoutcomes does allow comparison between different trials, but it is necessary toensure that outcomes considered important to patients are included.

The ACR core set of outcome measures was designed to assess disease activityand some long-term outcomes (e.g. disability), but may not reflect other outcomesimportant to patients. When patients with RA were asked to prioritize outcomesusing a scale comprising outcomes determined by clinicians (Meenan et al., 1992),they selected pain as the most important outcome (Heiberg and Kvien, 2002;Minnock et al., 2003). However, in a recent qualitative study, RA patients wereinvited to generate their own outcomes of importance (Carr et al., 2003). This studycomprised focus groups of RA patients in five rheumatology centres across the UK,who were asked to discuss the issue, ‘What outcomes from your treatments areimportant to you?’. Systematic analysis (Colaizzi, 1978) by several researchers(reviewed by a patient research partner), showed that patients spontaneously raisedmany different outcomes, such as frustration, which are not in the current core set

Hewlett et al.M

ain

ar t

icle

s132


MSC 3(3) 9/1/05 2:20 PM Page 132

of painful joints, swollen joints, pain, disability and global opinion. It is possible thatpatients might consider some of these patient-generated outcomes to be as importantas those designated in the ACR core set, in which case a review of the core set mightbe indicated. However, it is not yet known if the outcomes generated by thesepatients are deemed important by a larger RA patient community.

The aims of this study were therefore (1) to examine whether recent patient-generated outcomes are important within a wider RA population; (2) to assess therelative importance of each outcome; and (3) to explore whether any importantoutcomes have been omitted.

Method

Participants

Participants consisted of consecutive RA patients attending outpatientappointments in the south, north and west of the UK, using three of therheumatology centres where the original qualitative study was performed (Bristol,Stoke-on-Trent, and Chertsey).

Materials

The 23 outcomes which emerged from the qualitative study in which patient-generated important outcomes were raised (Carr et al., 2003) were listed in aquestionnaire (Box 1). Patients were asked to give an importance score for eachoutcome listed (‘Please tick the column that shows how you feel about eachoutcome’, on a 4-point scale from 0 = not at all important to 4 = very important),

Importance of patient-generated outcomes in RA

Mai

n a

rtic

les

133


BOX 1. Outcomes generated by RA patients as important (Carr et al., 2003)

Outcome Item Outcome Item

Have less pain from my arthritis A Not feel tired or fatigued MHave less stiffness in my joints B Not feel lacking in energy NHave less swelling in my joints C Return to/maintain a normal lifestyle OKeep mobile (get around) D Feel able to join in social events PDo everyday tasks at home/work E Feel well in myself QStay independent F Not have such unpredictable disease RBe happy with my body image G Not have side-effects from my drugs SBe less anxious H Not have to take medication TNot feel frustrated by arthritis I Be able to fulfil my usual role at home UNot feel helpless about my arthritis J Enjoy my life VNot feel low or depressed K Be able to fulfil my usual role at work WHave a feeling of general ‘well-being’ L

MSC 3(3) 9/1/05 2:20 PM Page 133

a method of eliciting importance for outcomes that has previously been validated(Hewlett et al., 2001, 2002). Patients were then asked to select their top three mostimportant outcomes (‘Which are the most important three outcomes for you?’) andthen to rank those top three items in order of importance (‘What order of prioritywould you put these three in?’). Finally, patients were asked to record any additionaloutcomes which were personally important to them, but which had been omittedfrom the list. To reduce any order effect in the patients’ responses, three versions ofthe questionnaire were produced, each with the outcomes in a different order. Equalnumbers (55) of each version were sent to each of the three rheumatology centres,with the 165 questionnaires in a random order. Questionnaires were anonymousand information was only requested on age, gender and disease duration.

Procedures

It was anticipated that 100–150 patients from each centre would provide a reasonablerepresentation of views. Consecutive patients were approached either in person attheir appointment (Centre B), or by mail in the weeks following an appointment(Centres A and C). No attempt was made to contact non-responders, from whom nodemographic data were collected. Ethics approval was obtained in each centre.

Analysis

The importance scores (0–3) were summed for each outcome and data presented asa percentage of the maximum possible importance score. Spearman’s rankcorrelations were used to examine relationships between importance score and bothage and disease duration.

Relative importanceThe number of times an outcome was selected in the top three most importantoutcomes is presented as a percentage of the maximum possible occasions. Chi-Squaretests were used to test for differences in selection between centres and between genders.

Additional outcomesProposed new outcomes thought to be missing from the list were counted.

FactorsPrincipal component analysis using a rotated component matrix was used to explorethe data for factors or groupings of outcomes, which tended to be rated in similarways, and the factors were then assigned labels.

Hewlett et al.M

ain

art

icle

s134


MSC 3(3) 9/1/05 2:20 PM Page 134

Results

Of the 495 distributed questionnaires 323 (65%) were returned (Centre A 102/62%,Centre B 150/91%, Centre C 71/43%). There were 235 female and 88 male patientswith a mean age of 60.8 years (SD 13.2 years) and disease duration of 13.7 years (SD11.1 years). These were not different between centres.

Importance of patient-generated outcomes

Importance scores (0–3) were summed for each outcome, giving a possible maximumimportance range of 0–969 (323 patients x 3 points). All 23 outcomes were deemedto be important by a substantial proportion of patients, ranging from ‘Be able to fulfilmy usual role at work’ (546 points or 56.3% of possible importance total) to ‘Keepmobile’ (927 points, 95.7%) (Figure 1).

Relative importance of outcomes

The items most frequently selected by patients in their top three most importantoutcomes were ‘Stay independent’ (39% of patients), ‘Have less pain from my arthritis’(36%) and ‘Keep mobile’ (34%) (Figure 2). However, they were not universallyimportant and were not selected in the top three outcomes by the majority of patients(61%, 64% and 66%, respectively). After these three outcomes, the most commonlyselected in the top three were ‘Return to or maintain a normal lifestyle’ (26%), ‘Feelwell in myself’ (20%), ‘Enjoy my life’ (18%), and ‘Not feel tired or fatigued’ (15%).These were selected more frequently than outcomes relating to drugs (‘Not have side-effects from my drugs’) and joint swelling and stiffness (‘Have less stiffness in my joints’and ‘Have less swelling in my joints’) at 13%, 12% and 9%. A desire to feel less tiredwas selected as important more frequently than a desire to have more energy (15% vs.3%). When patients ranked their top three outcomes in order of importance theresults were no different. There did not appear to be an order effect in the listing ofitems as the top seven most frequently selected outcomes remained the same when thedifferent versions of the questionnaires were compared.

Importance and patient characteristics

The importance values (0–3) placed on individual outcomes were not related toeither age or disease duration for any of the 23 items (all rs < 0.26, notsignificant). In terms of the relative importance of outcomes, women selected‘Not feel tired or fatigued’ more frequently in their top three important


Mai

n a

rtic

les

135


MSC 3(3) 9/1/05 2:20 PM Page 135

Hewlett et al.M

ain

art

icle

s136


0

100

200

300

400

500

600

700

800

900

1000

D: Mob

ile

F: In

dep

V: Enjoy

Q: Fee

l well

A: Pain

B: Stif

fness

O:Norm

al life

L: Well

being

U: Hom

e role

S: S/ef

fects

N: Energ

y

M: T

ired

C: Swell

ing

E: Dail

y task

s

K: Low

moo

d

I: Frus

tratio

n

J: Help

less

R: Unpre

dictab

le

P: Soc

ial lif

e

H: Anxie

ty

T: Tak

e drug

s

G: Bod

y imag

e

W: W

ork ro

le

Sum

impo

rtan

ce s

core

s(0-

969)

0

5

10

15

20

25

30

35

40

F: In

dep

A: Pain

D: Mob

ile

O: Norm

al life

L: Fee

l well

V: Enjoy

M: T

ired

S: S/ef

fects

B: Stif

fness

U: Hom

e role

E: Dail

y task

s

L: Well

being

K: Low

moo

d

C: Swell

ing

W: W

ork ro

le

T: Tak

e drug

s

I: Frus

tratio

n

R: Unpre

dictab

le

J: Help

less

N: Energ

y

P: Soc

ial lif

e

G: Bod

y imag

e

H: Anxie

ty

% o

f pat

ient

s

FIGURE 1. Importance of 23 patient-generated outcomes (n=323).

FIGURE 2. Proportion of patients who selected each outcome in their top three mostimportant outcomes (n = 323).

MSC 3(3) 9/1/05 2:20 PM Page 136

outcomes than men (16.8% vs. 7.8% P= 0.05). Men selected ‘Be able to fulfil myusual role at work’ (19.5% vs. 10.3%, P = 0.036) and ‘Not have side-effects frommy drugs’ (18.2% vs. 4.8%, P < 0.0001) more frequently than women. Therewere no differences between men and women for the other 20 importantoutcomes (Table 1).

Centres differed in their selection of the top three important outcomes only for‘Return to or maintain a normal lifestyle’ which was selected by Centre B on 36.7%of occasions, Centre A 20.6%, and Centre C 11.6% (P < 0.0001) and ‘Not feelfrustrated by arthritis’ which was selected by Centre A on 10.8% of occasions,Centre B 2%, and Centre C 4.3% (P < 0.01).

Factor analysis

Factor analysis on the importance scores (0–3) revealed six reasonably distinctgroupings which together explained 60.47% of the variance (Table 2). After reviewby the steering group (including a patient, PR) these were labelled as ‘General well-being’ (11.88% explained variance), ‘Day-to-day functioning’ (10.61%), ‘Emotionaland psychological well-being’ (10.59%), ‘Social role and confidence’ (10.05%),


Mai

n a

rtic

les

137


TABLE 1. Outcomes selected in top three most important, comparison between genders

Outcome Female (%) Male (%) p value

Stay independent 96 (41.4) 23 (29.9) 0.072Have less pain from my arthritis 84 (36.2) 29 (37.7) 0.818Keep mobile (get around) 84 (36.2) 21 (27.3) 0.152Return to/maintain a normal lifestyle 58 (25.0) 22 (28.6) 0.535Feel well in myself 51 (22.0) 12 (15.6) 0.227Not feel tired or fatigued 39 (16.8) 6 (7.8) 0.05Enjoy my life 38 (16.4) 16 (20.8) 0.378Have less stiffness in my joints 28 (12.1) 9 (11.7) 0.929Have a feeling of general ‘well-being’ 27 (11.6) 4 (5.2) 0.103Do everyday tasks at home/work 26 (11.2) 7 (9.1) 0.621Be able to fulfil my usual role at home 24 (10.3) 9 (11.7) 0.741Not have side-effects from my drugs 24 (10.3) 15 (19.5) 0.036Not feel low or depressed 21 (9.1) 7 (9.1) 0.992Have less swelling in my joints 26 (8.6) 6 (7.8) 0.820Not have to take medication 18 (7.8) 6 (7.8) 0.992Not feel frustrated by arthritis 11 (4.7) 6 (7.8) 0.309Be able to fulfil my usual role at work 11 (4.7) 14 (18.2) 0.0001Not have such unpredictable disease 10 (4.3) 5 (6.5) 0.440Not feel helpless about my arthritis 9 (3.9) 6 (7.8) 0.166Not feel lacking in energy 8 (3.5) 2 (2.6) 0.715Feel able to join in social events 6 (2.6) 3 (3.9) 0.554Be happy with my body image 5 (2.2) 2 (2.6) 0.821Be less anxious 2 (0.9) 1 (1.3) Insufficient data

MSC 3(3) 9/1/05 2:20 PM Page 137

‘Physical symptoms’ (9.47%) and ‘Medication issues’ (7.88%). Three outcomescontributed to more than one factor: ‘Not feel depressed’ (Emotional andpsychological well-being; General well-being), ‘Feel well in myself’ (General well-being; Social role and confidence) and ‘Be happy with my body image’ (Social roleand confidence; Medication issues).

Additional outcomes

Ten new outcomes of personal importance were generated. The most frequentsuggestion, self-esteem, was raised by only 9/323 patients (2.8%) and described by

Hewlett et al.M

ain

art

icle

s138


TABLE 2. Factor analysis of 23 outcomes

Outcome Item Factor 11 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6

Not tired/fatigued M .792 .133 .168 –.019 .062 .130Not lacking in energy N .761 .130 .176 .160 .040 .074General ‘well-being’ L .672 .000 .089 .210 .129 .116Feel well in myself Q .470 .217 .093 .415 .157 .139

Stay independent F .093 .831 .126 .057 –.071 .108Everyday tasks E .235 .642 .118 .201 .212 –.068Keep mobile D –.082 .629 .163 –.009 .349 .031Fulfil role at home U .149 .603 –.025 .200 .134 .279Maintain normal life O .381 .467 .047 .268 .381 –.104

Not feel frustrated I .216 .086 .756 .073 .067 .131Not feel helpless J –.040 .170 .697 .259 .065 .178Be less anxious H .249 .191 .605 .142 .262 .203Not feel depressed K .416 .081 .569 –.074 .164 –.027Not unpredictable R .086 –.135 .477 .205 .396 .214

Enjoy my life V –.012 .103 .032 .770 .156 .036Join in social events P .297 .093 .142 .691 .137 .092Body image G .078 .150 .197 .599 .012 .467Fulfil role at work W .348 .156 .298 .524 –.083 –.161

Less joint stiffness B .152 .152 .056 .079 .781 .245Less pain A .033 .173 .237 .136 .664 –.047Less joint swelling C .175 .229 .161 .040 .620 .385

No drug side-effects S .234 .097 .106 –.047 .171 .756No medication T .022 .051 .284 .200 .139 .691

Variance 60.47% 11.88% 10.61% 10.59% 10.05% 9.47% 7.88%

1Factor 1: General well-being; 2: Day-to-day functioning; 3: Emotional and psychological well-being; 4: Social role and confidence; 5: physical symptoms; 6: Medication issues

MSC 3(3) 9/1/05 2:20 PM Page 138

one patient as ‘Feeling less worthy’ (patient 48). A second outcome wasrelationships, raised by seven patients (e.g. ‘Have proper hugs without pain’, patient104) and six patients wanted a complete cure (‘A miracle cure’, patient 274). Othermissing outcomes were coping, long-term consequences, leisure activities, sleep,driving, not being a burden and no drug monitoring (each 2–5 patients). Fourpatients commented that many outcomes were inter-related or that improving onewould mean an automatic improvement in others. One patient commented on beingprepared to trade the importance of ‘Not have side-effects from my drugs’ against theimportance of ‘Return to or maintain a normal lifestyle’.

Discussion

This study shows that outcomes generated by a small number of patients in aprevious in-depth qualitative study on important outcomes in RA, aregeneralizable when examined in a larger, quantitative study in three UK centres,where no outcome was deemed unimportant. The data, which are based onpatient-generated outcomes of importance, provide further evidence that painand function are important outcomes for patients, as has been shown in studiesusing clinician-generated important outcomes. This supports their inclusionwithin the ACR core set, not only as measures of disease activity and long-termoutcome (the prime reason for which they were included) but also in terms ofpatient priority.

However, the data show that no outcome in RA is deemed universallyimportant and more than 60% of patients in this study did not select pain orfunction within their three most important outcomes. Patients more frequentlyselected feeling well and fatigue in their top three important outcomes (20%, 15%)than the commonly measured disease activity outcomes of side effects from drugs(13%), stiffness (12%) and swelling (9%). In previous studies of prioritizingoutcomes, 65–69% of RA patients reported their treatment priority as pain usingthe AIMS2 categories (Minnock et al., 2003; Heiberg and Kvien, 2002), comparedto 36% of patients selecting pain within their top three important outcomes in thisstudy. However, the AIMS 2 comprises 12 professionally-determined categories, ofwhich seven relate primarily to physical activities, and does not include widerranging patient-generated outcomes such as fatigue and well-being. WhenMinnock and Bresnihan re-surveyed their sample and included the additionaloption of selecting fatigue, it replaced pain as the most frequent patient priority(fatigue 65%, pain 48%) (Minnock and Bresnihan, 2004).

Data from the current study suggest a wide range of important outcomes andpriorities for patients. However, little is known about the relative importance of


Mai

n a

rtic

les

139


MSC 3(3) 9/1/05 2:20 PM Page 139

outcomes, nor about whether priorities change with disease status, duration, orprogression, or with personal lifestyle issues. Little is known about the trade-offpatients may be prepared to make between different outcomes (e.g. accepting sideeffects from medication to achieve pain relief). The variability in the importanceof outcomes may help explain some of the variability in adherence withtreatments or in patient perceptions of improvement in health status. Furtherwork on identifying factors influencing priorities would be helpful.

The current status or severity of the outcomes (e.g. level of pain or fatigue) werenot assessed in the patients taking part in this study, therefore it is not known whetherthe importance of an outcome is related to outcome status. While a relationshipbetween importance and severity of an outcome has been shown previously (Heibergand Kvien, 2002), it has also been demonstrated that the importance of someoutcomes such as disability, is independent of status (Hewlett et al., 2002).

The three centres surveyed in this study were likely to reflect a range ofdifferent socio-economic groups in the north, south and west of the UK andshowed no demographic differences, suggesting they may provide a reasonablyrepresentative view of importance. Only one major difference was seen in outcomeselection (Centre B patients more frequently chose ‘Not feel frustrated byarthritis’) out of 23 options. Centre C invited all patients to participate by postrather than in person, which may account for their lower response rate.

Patients selected ‘Not feel tired or fatigued’ more frequently than ‘Not feellacking in energy’ which suggests that not only is fatigue perceived as a biggerproblem, but also that fatigue and energy may be different concepts and may notsimply be the two extremes of a single continuum. This has implications for themeasurement of fatigue and the design of scales.

Patients identify well-being issues and fatigue as being the most importantoutcomes after pain and function. Well-being issues are difficult to define, measureand treat. They are likely to be consequences of disease activity and may improve asdisease activity is reduced. In contrast, although fatigue may be a result of manycomplex interactions (such as pain, disability and anaemia), fatigue is in itself a causeof other outcomes. For example, fatigue has been shown to have physical, cognitiveand emotional elements that disrupt function, leisure activities, relationships andwork (Tack, 1990; Hewlett et al., 2004). Fatigue has been shown to be a factor indecisions about change in drug therapy and has been proposed as a key outcome forassessment (Wolfe et al., 2001). Fatigue in RA was the subject of much discussion atOMERACT 7 in 2004 (Outcome Measurement in Rheumatology Clinical Trials),and voting among international researchers and clinicians endorsed a mandate forfurther work on measuring fatigue with the potential for consideration as anadditional measure for the core set (Kirwan et al., 2005).

Hewlett et al.M

ain

ar t

icle

s140


MSC 3(3) 9/1/05 2:20 PM Page 140

Conclusions

This study supports the use of pain and function as core outcome measures, reflectingboth disease activity and patient priorities, but suggests that additional outcomessuch as feeling well and feeling less fatigued, are considered by patients to be moreimportant than traditional disease activity outcomes such as joint swelling andstiffness. This adds weight to the idea that a review of the recommended core set ofoutcome measures might be undertaken. The measurement and treatment of fatiguein particular should be considered as an urgent and important area of future research.

Acknowledgements

The authors would like to thank the patients for participating, clinical staff fordistributing questionnaires, Dr Paul Creamer for his collaboration, and RosemaryGreenwood for statistical advice. Dr Hewlett’s academic post is funded by theArthritis Research Campaign.

Sarah Hewlett initiated the study and was principle investigator. SarahHewlett, Maggie Carr and Sarah Ryan carried out the study and Sarah Hewlettanalysed the data. All authors contributed to the interpretation of the data analysisand critically reviewed and contributed to the drafts of this paper. Sarah Hewlettacts as guarantor for the paper and accepts full responsibility for the conduct of thestudy, had access to the data, and controlled the decision to publish.

ReferencesBerkanovic E, Hurwicz ML, Lachenbruch PA (1995). Concordant and discrepant views of patients’

physical functioning. Arthritis Care Research 8(2): 94–101.Carr AJ (1996). A patient-centred approach to evaluation and treatment in rheumatoid arthritis: The

development of a clinical tool to measure patient-perceived handicap. British Journal ofRheumatology 35: 921–32.

Carr A, Hewlett S, Hughes R, Mitchell H, Ryan S, Carr M, Kirwan J (2003). Rheumatologyoutcomes: The patient’s perspective. Journal of Rheumatology 30: 880–3.

Colaizzi PF (1978). Psychological review. In R Valle, M King (eds.) Existential PhenomenologicalAlterations for Psychology. Oxford: Oxford University Press.

Felson DT, Anderson JJ, Boers M, Bombardier C, Chernoff M, Fried B, Furst D, Goldsmith C, KieszakS, Lightfoot R, Paulus H, Tugwell P, Weinblatt M, Widmark R, Williams HJ, Wolfe F (1993).The American College of Rheumatology preliminary core set of disease activity measures forrheumatoid arthritis clinical trials. Arthritis and Rheumatism 36(6): 729–40.

Heiberg T, Kvien TK (2002). Preferences for improved health examined in 1,024 patients withrheumatoid arthritis: Pain has highest priority. Arthritis and Rheumatology (Arthritis CareResearch) 47: 391–7.

Hewlett S, Smith AP, Kirwan JR (2001). Values for function in rheumatoid arthritis: Patients,professionals and public. Annals of Rheumatic Disease 60: 928–33.


Mai

n a

rtic

les

141


MSC 3(3) 9/1/05 2:20 PM Page 141

Hewlett S, Smith AP, Kirwan JR (2002). Measuring the meaning of disability in rheumatoid arthritis:The Personal Impact Health Assessment Questionnaire (PI HAQ). Annals of RheumaticDisease 61: 986–93.

Hewlett S, Hehir M, Cockshott Z, Byron M, Kitchen K, Tipler S, Pope D (2004). Patients’perceptions of fatigue in rheumatoid arthritis: Overwhelming, uncontrollable, ignored. Arthritisand Rheumatism 50(9): 1817.

Katz PP, Yelin EH (1995). The development of depressive symptoms among women with rheumatoidarthritis. Arthritis Care Research 38: 49–56.

Kirwan JR, Hewlett S, Heiberg T, Hughes R, Carr M, Hehir M, Kvein T, Minnock P, Newman S,Quest EM, Taal E, Wale J (2005). Incorporating the patient perspective into outcome assessmentin rheumatoid arthritis – progress at OMERACT 7. Journal of Rheumatology (in press).

Kwoh CK, O’Connor GT, Regan-Smith MG, Olmstead EM, Brown LA, Burnett JB, Hochman RF,King K, Morgan GJ (1992). Concordance between clinician and patient assessment of physicaland mental health status. Journal of Rheumatology 19(7): 1031–7.

Kwoh KC, Ibrahim SA (2001). Rheumatology patient and physician concordance with respect toimportant health and symptom status outcomes. Arthritis Care Research 45: 372–7.

Kvien TK, Heiberg T (2003). Patient perspective in outcome assessments – perceptions or somethingmore? Journal of Rheumatology 30: 873–6.

Meenan RF, Mason JH, Anderson JJ, Guccione AA, Kazis LE (1992). AIMS 2. The content andproperties of a revised and expanded Arthritis Impact Measurement Scales health statusquestionnaire. Arthritis and Rheumatism 35(1): 1–10.

Minnock P, Fitzgerald O, Bresnihan B (2003). Quality of life, social support and knowledge of diseasein women with rheumatoid arthritis. Arthritis and Rheumatism (Arthritis Care Research) 49(2):221–7.

Minnock P, Bresnihan B (2004). Pain outcome and fatigue levels reported by women with establishedrheumatoid arthritis. Arthritis and Rheumatism 50(9): 1197.

Rothwell PM, McDowell Z, Wong CK, Dorman PJ (1997). Doctors and patients don’t agree: Cross-sectional study of patients’ and doctors’ perceptions and assessments of disability in multiplesclerosis. British Medical Journal 314: 1580–3.

Ruta DA, Garratt AM, Leng M, Russell IT, MacDonald LM (1994). A new approach to themeasurement of quality of life – the Patient Generated Index. Medical Care 32(1): 1109–26.

Suarez-Almazor ME, Conner-Spady B, Kendall CJ, Russell AS, Skeith K (2001). Lack of congruencein the ratings of patients’ health status by patients and physicians. Medical Decision Making 21:113–21.

Tack BB (1990). Fatigue in rheumatoid arthritis: Conditions, strategies and consequences. ArthritisCare and Research 2: 65–70.

Tugwell P, Bombardier C, Buchanan WW, Goldsmith CH, Grace E, Hanna B (1987). The MACTARpatient preference disability questionnaire – an individualized functional priority approach forassessing improvement in physical disability in clinical trials in rheumatoid arthritis. Journal ofRheumatology 14(3): 446–51.

Wolfe F, Pincus T, O’Dell J (2001). Evaluation and documentation of rheumatoid arthritis diseasestatus in the clinic: Which variables best predict change in therapy. Journal of Rheumatology 28:1712–17.

Correspondence should be sent to Sarah Hewlett, arc Senior Lecturer, Rheumatology HealthProfessions, University of Bristol Academic Rheumatology, Bristol Royal Infirmary, Bristol BS28HW. Tel: 44 (0)117 928 2903, Fax:44 (0)117 928 3841. Email: [email protected]

Received December 2005Accepted March 2005

Hewlett et al.M

ain

ar t

icle

s142


MSC 3(3) 9/1/05 2:20 PM Page 142

Outcomes generated by patients with rheumatoid arthritis: how important are they?

Documents

Transcript of Outcomes generated by patients with rheumatoid arthritis: how important are they?