Exploring therapeutic action in psychoanalytic psychotherapy753745/FULLTEXT02.pdf · explore...

Exploring therapeutic action in psychoanalytic psychotherapy Attachment to therapist and change Peter Lilliengren

“It has been a permissive and respectful atmosphere here and I have dared to open up. I have dared to talk about things that I have not talked about with many others before, even though I’m usually an open person.” (Patient, study I)

“It’s about daring to connect, to begin trusting someone. To begin to feel that she can come here and open up about something that feels frightening and threatening to her. It’s about daring to be dependent and to permit closeness, of which she has had little positive experience.” (Therapist, study II)

“[The therapist] is an attachment figure who provides the security needed for insecurity to be explored.” (Holmes, 2010, p. 6)

ABSTRACT There are several theoretical notions of how psychoanalytic psychotherapy work, but only limited empirical evidence to support any of them. The overall aim of this thesis was to explore therapeutic action in psychoanalytic psychotherapy from different perspectives (patient, therapist, and observer), using different methodological approaches (qualitative and quantitative). Study I explores 22 young adult patients’ views of therapeutic action with qualitative, grounded theory methodology. The analysis resulted in a conceptual model indicating that talking openly in the context of a safe therapeutic relationship led to new relational experiences and expanding self-awareness. Hindering factors included difficulties “opening up”, experiencing the therapist as too passive and that something was missing in therapy, leading to an experience of mismatch. Study II investigates 16 experienced therapists’ views of therapeutic action. The results indicated that the development of a close, safe and trusting relationship was perceived as the core curative factor. This interacted with the patient making positive experiences outside the therapy setting and the therapist challenging the patients’ thinking about the self. Patients’ fears about close relationships emerged as the sole hindering factor from the therapists’ perspective, perceived as creating distance in the therapeutic relationship and leading to maintenance of patients’ problems.

The results of study I and II suggest that the quality of patients’ attachment to their therapist may be important for treatment process and outcome. In order to examine attachment to therapist using quantitative methodology, a new observer-rating scale (Patient Attachment to Therapist Rating Scale; PAT-RS) was developed. Study III involves an initial examination of the psychometric properties of PAT-RS. Three raters rated a total of 70 interviews. The results indicated good inter-rater reliability for three of the subscales (Security, ICC = .74; Deactivation, ICC = .62; Disorganization, ICC = .74), while one had poor (Hyperactivation, ICC = .34). Correlations with measures of the therapeutic alliance, mental representations, and symptom distress were moderate and in the in the expected directions, suggesting construct validity for the reliable subscales. Study IV investigates the relationships between secure attachment to therapist, patient-rated alliance and outcome. A series of linear mixed-effects models, controlling for between-therapist variability and length of therapy, indicated that secure attachment to therapist at termination was related to improvements in symptoms, global functioning and interpersonal problems. The relationships with symptom change and increased functioning was maintained after the therapeutic alliance was accounted for. Further, a suppression effect was found, indicating that secure attachment to therapist at termination predicted continued improvement in functioning during follow-up, whereas the alliance predicted deterioration when both variables were modeled together.

The overall result of this thesis indicates that the development of a secure attachment to the therapist may be an important mechanism of change in psychoanalytic psychotherapy. This is discussed in relation to common and specific factors in psychotherapy, as well as in established theoretical notions of therapeutic action in the psychoanalytic discourse.

Implications for the measurement of secure attachment to therapist and its differentiation from the therapeutic alliance are considered. Based on the results, two tentative process models that may be useful for clinical practice and future research are proposed: a broaden-and-built cycle of attachment security development and change and a react-and-disconnect cycle of attachment insecurity maintenance. Future research should investigate the temporal development of attachment to therapist and its relation to the therapeutic alliance and outcome more closely. Specific strategies that foster a secure attachment to therapist, as well as interventions for dissolving insecure strategies, should be identified and integrated in the theory and practice of psychoanalytic psychotherapy. Keywords: Therapeutic action, mechanisms of change, psychoanalytic, psychodynamic, psychotherapy, young adults, attachment to therapist, therapeutic alliance, process, outcome, grounded theory, linear mixed models

LIST OF PUBLICATIONS This thesis is based on the following studies, which are referred to in the text by their Roman numerals. I. Lilliengren, P., & Werbart, A. (2005). A model of therapeutic action grounded in the

patients’ view of curative and hindering factors in psychoanalytic psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 42(3), 324–339. doi:10.1037/0033-3204.42.3.324

II. Lilliengren, P., & Werbart, A. (2010). Therapists’ view of therapeutic action in

psychoanalytic psychotherapy with young adults. Psychotherapy, 47(4), 570–585. doi:10.1037/a0021179

III. Lilliengren, P., Werbart, A., Mothander, P. R., Ekström, A., Sjögren, S., & Ögren, M-

L. (2014). Patient Attachment to Therapist Rating Scale: Development and psychometric properties. Psychotherapy Research, 24(2), 184–281. doi:10.1080/10503307.2013.867462

IV. Lilliengren, P., Falkenström, F., Sandell, R., Mothander, P. R., & Werbart, A. (2014).

Secure attachment to therapist, alliance, and outcome in psychoanalytic psychotherapy with young adults. Journal of Counseling Psychology. Advance online publication. doi:10.1037/cou0000044

The original papers are reproduced with permission from the publishers. Papers I, II and IV: © American Psychological Association (www.apa.org). Paper III: © Taylor & Francis (www.tandfonline.com).

LIST OF ABBREVIATIONS AAI Adult Attachment Interview CATS Client Attachment to Therapist Scale CAQ-T Components of Attachment Questionnaire – Therapist DIP-Q DSM-IV and ICD-10 Personality Questionnaire DRS Differentiation-Relatedness Scale DSM-IV Diagnostic and Statistical Manual of Mental Disorders GAF Global Assessment of Functioning Scale GSI Global Severity Index of the SCL-90 GT Grounded Theory HAq-II Helping Alliance Questionnaire ICD-10 International Statistical Classification of Diseases ICC Intraclass correlation coefficient IIP Inventory of Interpersonal Problems ORI Object Relations Inventory PACS Patient Attachment Coding System PAT-RS Patient Attachment to Therapist Rating Scale PT-AAI Patient-Therapist Adult Attachment Interview PTI Private Theories Interview REML Restricted maximum likelihood SASB Structural Analysis of Social Behavior SEM Structural Equation Modeling SCL-90 Symptom Checklist – 90 SPSS Statistical Package for the Social Sciences WAI Working Alliance Inventory YAPP Young Adult Psychotherapy Project

CONTENTS INTRODUCTION ...................................................................................................................... 1

Common and specific factors .......................................................................................................................... 1

Perspectives on therapeutic action .......................................................................................... 3 The patient’s perspective ................................................................................................................................. 3

The therapist’s perspective .............................................................................................................................. 4

Methodological perspectives ........................................................................................................................... 5

Therapeutic action in psychoanalytic psychotherapy ............................................................. 6 Insight .............................................................................................................................................................. 6

Relational impact ............................................................................................................................................. 8

Alliance and attachment ....................................................................................................... 10 The therapeutic alliance ................................................................................................................................. 11

Attachment to therapist ................................................................................................................................. 12

Assessing attachment to therapist .................................................................................................................. 13

The impact of patient-therapist attachment on psychotherapy process ......................................................... 14

Differentiating secure attachment to therapist from the therapeutic alliance ................................................ 15

Young adults in psychoanalytic psychotherapy ................................................................... 16 AIM OF THE THESIS ............................................................................................................. 19 METHOD ................................................................................................................................. 21

Participants and procedures .................................................................................................. 21 Research context ............................................................................................................................................ 21

Study samples ................................................................................................................................................ 22

Ethical considerations .................................................................................................................................... 23

Measures and methods .......................................................................................................... 23 Interviews ...................................................................................................................................................... 23

Qualitative methodology ............................................................................................................................... 24

Self-report measures ...................................................................................................................................... 25

Expert-rated measures ................................................................................................................................... 26

Statistical analyses ......................................................................................................................................... 27

Development of the Patient Attachment to Therapist Rating Scale (PAT-RS) .................... 28 Basic assumptions ......................................................................................................................................... 28

Theoretical structure and subscales ............................................................................................................... 29

Item construction ........................................................................................................................................... 31

Rating procedure and scoring ........................................................................................................................ 32

RESULTS ................................................................................................................................. 35 Study I: A model of therapeutic action grounded in the patients’ view of curative and hindering factors in psychoanalytic psychotherapy .............................................................. 35 Study II: Therapists’ view of therapeutic action in psychoanalytic psychotherapy with young adults .......................................................................................................................... 37 Study III: Patient Attachment to Therapist Rating Scale: Development and psychometric properties .............................................................................................................................. 38 Study IV: Secure attachment to therapist, alliance, and outcome in psychoanalytic psychotherapy with young adults ......................................................................................... 40

DISCUSSION .......................................................................................................................... 43 Reflections on main findings ................................................................................................ 43

Patients’ and therapists’ views of curative factors ........................................................................................ 43

Patients’ and therapists’ views of hindering factors ...................................................................................... 44

Assessing attachment to therapist with PAT-RS ........................................................................................... 46

Secure attachment to therapist and change .................................................................................................... 47

Secure attachment to therapist and the alliance constructs revisited ............................................................. 48

Psychoanalytic psychotherapy with young adults ......................................................................................... 49

Methodological considerations ............................................................................................. 51 Strengths ........................................................................................................................................................ 51

Limitations ..................................................................................................................................................... 52

Concluding thoughts ............................................................................................................. 53 Two tentative process models ....................................................................................................................... 53

Implications for practice ................................................................................................................................ 55

Suggestions for future research ..................................................................................................................... 56

ACKNOWLEDGEMENTS IN SWEDISH ............................................................................. 59 REFERENCES ......................................................................................................................... 61 APPENDIX: Patient Attachment to Therapist Rating Scale (PAT-RS) ORIGINAL PAPERS I-IV

1

INTRODUCTION Modern-day psychotherapy research has firmly established that “psychotherapy works” (Lambert, 2013a). Since the 1970-ties, several hundred randomized controlled trials and numerous meta-analyses have repeatedly shown that systematically applied psychotherapy leads to better outcomes than no treatment or usual care for a wide range of conditions (e.g., Budge et al., 2013; Cuijpers, Andersson, Donker, & van Straten, 2011; Koelen et al., 2013; Lipsey & Wilson, 1993; Smith, Glass, & Miller, 1980; Stoffers et al., 2013; Wampold, 2001). In general, there seems to be no or only small differences between various forms of bona-fide psychotherapy for the average patient (e.g., Baardseth et al., 2013; Benish, Imel, & Wampold, 2008; Lambert, 2013a; Leichsenring, 2001; Leichsenring & Leibing, 2003; Wampold et al., 2002). However, establishing that psychotherapy works in general does not tell us much about how it works. Further, psychotherapy does not produce the same outcomes in all patients. Research indicates that approximately 20% of patients drop out before experiencing any benefits (McMurran, Huband, & Overton, 2010; Roos & Werbart, 2013; Swift & Greenberg, 2012) and about 5-10% may actually deteriorate while in therapy (Gold & Stricker, 2011; Lambert, 2011, 2013a; Mohr, 1995). Thus, although the general outcome for the average patient is typically good, psychotherapy involves a complex set of interacting factors that might lead toward improvement, dropout or deterioration for any given patient.

The question of what promotes or hinders change in psychotherapy has been the central concern for over 60 years of process-outcome research (Orlinsky, Rønnerstad & Wilutzki, 2004). This area of research is vast and multifaceted, both in terms of its methods and findings. Yet, despite long-term attention from process researchers, we still know quite little about the fundamental “mechanisms of change” in different forms of psychotherapy (Johansson & Høglend, 2007; Kazdin, 2007, 2009; Kraemer, Wilson, Fairburn, & Agras, 2002). For example, Kazdin (2007) concludes that: “…it is remarkable that after decades of psychotherapy research, we cannot provide an evidence-based explanation for how or why even our most well studied interventions produce change” (p. 23). Thus, more research investigating the therapeutic action of psychotherapy is needed. A greater understanding of the underlying change mechanisms could aid in the development of new treatments and/or increase the effectiveness of the models currently practiced. Further, identifying the fundamental mechanisms and processes that lead to change might potentially also reduce the gap between different schools of psychotherapy, as well as bring researchers and clinicians closer to each other (Kazdin, 2008).

Common and specific factors The repeated finding that there are no or only small differences in outcome between bona-fide psychotherapy models has sparked a long-standing debate in the field regarding the relative importance of common and specific factors for treatment outcome. Common factors are

2

thought to be present to some degree in all forms of psychotherapy. These include the therapeutic alliance, patient expectations and therapist qualities such as empathy, positive regard, and genuineness, all of which have been shown to be significant predictors of outcome (Norcross, 2011). Since different forms of psychotherapy tend to produce similar results, some scholars have argued that the treatment effects are probably largely due to such common factors (Messer & Wampold, 2002).

However, it is also possible that specific factors drive the outcome in different forms of psychotherapy, generating similar results on average (DeRubeis, Brotman, & Gibbons, 2005). Specific factors may refer to specific techniques or procedures predominantly used in a particular form of treatment (e.g., exposure in behavior therapy or transference interpretations in psychoanalytic psychotherapy). Specific factors may also refer to the underlying processes through which change is proposed to occur based on a particular theoretical perspective (e.g., provision of insight in psychoanalytic psychotherapy or change in negative automatic thoughts in cognitive therapy). Although there is little evidence to suggest that certain specific factors are absolutely crucial for outcome with any given patient population, process-outcome studies do suggest that several specific factors are linked to change (Crits-Christoph, Connolly Gibbons, & Mukherjee, 2013). Further, some research even suggests that factors thought to be specific to one particular school of psychotherapy may unwittingly contribute to outcome in another form of psychotherapy (Ablon & Jones, 1998; Castonguay, Goldfried, Wiser, Raue, & Hayes, 1996; DeFife, Hilsenroth, & Gold, 2008; Hilsenroth, Blagys, Ackerman, Bonge, & Blais, 2005; Jones & Pulos, 1993).

Other researchers have argued that the distinction between common and specific factors is a false dichotomy (Wampold & Budge, 2012). According to a contextual view of psychotherapy (Frank & Frank, 1991; Wampold, 2001), specific factors are essential for outcome since without any specific activities (i.e., “healing rituals” promoting healthy actions) associated with a particular form of psychotherapy there would simply be no treatment for the patient to respond to. At the same time, specific factors can only be delivered in the context of common factors (including relationship factors, as well as patient expectations and cultural influences) and arguably derive much of their power from that context. Several attempts have been made to theoretically integrate both common and specific factors into generic, “meta-models” of psychotherapy (e.g., Anderson, Lunnen, & Ogles, 2010; Wampold & Budge, 2012).

However, there is probably still a long way to go before a generic meta-model of psychotherapy will receive wide acceptance in the field. All schools of psychotherapy have their own preferred theories and hypotheses regarding the therapeutic action of their particular form of treatment, and many of these notions have still not been empirically tested. Further, different theories of psychotherapy are based on different epistemological assumptions and “visions of reality” (Messer & Winokur, 1984; Safran & Messer, 1997), which makes it difficult to fully integrate or substitute them with a universal model. Still, one possible future prospect is that the major schools of psychotherapy (e.g., psychoanalytic, humanistic, cognitive, behavioral) will continue to assimilate findings from psychotherapy research, gradually leading to greater integration of the field as a whole (Lampropoulos, 2001; Messer, 2001). A starting point for this thesis project is that this process will likely be aided by the identification of both common and specific mechanisms of change in different forms of psychotherapy.

3

Perspectives on therapeutic action The study of therapeutic action may be approached from the perspective of the patient, the therapist or independent observers (Mintz, Auerbach, Luborsky, & Johnson, 1973; Strupp, Fox, & Lesser, 1969), as well as using different research methodologies (Lutz & Knox, 2014). As a background for the studies included in this thesis, some current issues and findings regarding the study of therapeutic action from these perspectives are introduced and summarized below. The patient’s perspective Although it may seem obvious that asking the patient what he or she experienced as helpful or hindering in treatment would be a valid approach to study therapeutic action, such efforts have been questioned historically (McLeod, 1990). In the psychoanalytic tradition, the patient’s conscious accounts of the therapy process have typically been interpreted as skewed by forces outside the patient’s awareness and therefore regarded with some suspicion. Similarly, behavioral researchers have tended to dismiss the patients’ subjective experiences as unreliable, preferring the use of objective measures of overt behaviors or psychophysical reactions. Scholars predominantly associated with the humanistic tradition, however, have long argued that patients’ perspective on their own process of change is a fundamental source of information (Howe, 1993; McLeod, 1990; Rogers, 1951; Strupp, Fox, & Lesser, 1969). Ultimately, it is the patient that implements change and therefore research should be directed towards what the patients find curative or hindering for them (Bohart, 2000; Bohart & Tallman, 1999).

The patients’ perspective on what is helpful or curative in psychotherapy (and, to a lesser extent, what is perceived as non-helpful or hindering) has been explored in a number of studies since the 1960-ties. Several reviews of this literature have also been published (see Bohart & Greaves Wade, 2013; Elliot & James, 1989; Rennie, 2002; Timulak, 2007, 2010). Overall, the findings suggest that patients typically experience factors that are common to different forms of psychotherapy as most helpful. For example, Elliott and James (1989) found that the most frequently reported helpful factors across therapies were (a) facilitative therapist characteristics, (b) self-expression permitted, (c) experiencing a supportive relationship, (d) gaining self-understanding, and (e) therapist encouraging extra-therapy practice. Similar findings were reported in a more recent review of 41 studies focusing on patient identified significant events in psychotherapy (Timulak, 2010).

Research further suggests that the patient’s view of what is curative or hindering in therapy may diverge considerably from their therapist’s view (Castonguay, et al., 2010; Hunsley, Aubry, Verstervelt, & Vito, 1999; Leuzinger-Bohleber, 2002; Timulak, 2010; Werbart & Levander, 2006, 2011) and that therapists often are unaware of such discrepancies (Hill, Nutt-Williams, Heaton, Thompson, & Rhodes, 1996; Rennie, 1994; Regan & Hill, 1992). Hindering factors reported by patients typically involve misunderstandings or negative reactions from the therapists, stressing the importance of actively inquiring into patients’ perspective of the process (Henkelman & Paulson, 2006; Hill & Knox, 2009).

Some studies also indicate that while certain patients may experience the specific factors in a particular form of treatment as helpful, others may experience the same factors as

4

hindering (Gershefski, Arnkoff, Glass, & Elkin, 1996; Nilsson, Svensson, Sandell, & Clinton, 2007; Levy, Glass, Arnkoff, & Gershefski, 1996). For example, Nilsson and colleagues (2007) compared satisfied and dissatisfied patients’ views of process and outcome in psychoanalytic psychotherapy and cognitive-behavioral therapy, respectively. Patients who were satisfied with psychoanalytic psychotherapy typically valued the reflective nature of the psychotherapy process and described having gained self-understanding. In contrast, the satisfied patients in cognitive-behavioral therapy described that they had learned to manage anxiety and cope in difficult situations and appreciated the therapist activity in structuring the treatment process. Dissatisfied patients in psychoanalytic psychotherapy typically described experiencing the therapist as withdrawn, disengaged, not providing enough support or guidance, whereas dissatisfied patients in cognitive-behavioral therapy typically considered the therapist to be intrusive and oppressive, applying a rigid predetermined therapy scheme. Hence, what patients experience as curative or hindering in treatment may also be moderated by particular patient factors (Bohart & Greaves Wade, 2013).

Although previous research suggests that patients report common factors as most helpful in therapy, we know very little about how such factors may interact with each other or with specific treatment factors from the view of the patient. We know even less about how perceived curative factors might interact with the patient’s experience of negative or hindering aspects in treatment. Study I of the present thesis explores patients’ own views of therapeutic action in psychoanalytic psychotherapy with particular attention to how curative and hindering factors may interrelate.

The therapist’s perspective Historically, the therapist perspective has had a privileged position when it comes to the development of theories of therapeutic action (Bohart, 2000; Bohart & Tallman, 1999). In the psychoanalytic tradition, new ideas have typically grown out of clinical observations and experiences of practicing therapists rather than through formal research. Still, even in traditions that emphasize experimental research, theoretical and technical advancements have often been developed by clinicians in order to accommodate the problems they faced in clinical practice (e.g., Beck, 1976; Ellis, 1962; Linehan, 1993). Indeed, in any field as complex as psychotherapy, theoretical and methodological progress is not likely to be driven by research alone.

Few would probably question the idea that experienced therapists’ views of what works and what does not work in therapy may contain valuable information. For example, several authors have argued that experienced therapists develop “implicit theories” regarding how to conduct therapy that are based on a mixture of personal experiences, formal training and professional reflections (Kottler, 1986; Najavits, 1997; Schön, 1983). These private notions may include the therapist’s personal strategies of what to do during sessions or views about what processes are occurring in therapy, which may more or less overlap with the therapist’s official theoretical orientation (Sandler, 1983). Thus, experienced therapists’ implicit theorizing may contain novel ideas that, if elucidated and systematized, could improve our formal theories and practices.

Yet, surprisingly little research has been directed at exploring what practicing therapists view as curative and hindering in their everyday work (McLeod, 1990, 2013). One possible explanation for the lack of research in this area may be that, traditionally, therapists’

5

subjective experiences have been reserved for supervision and personal psychotherapy (McLeod, 1990). Therapists may also experience resistance against openly discussing private notions when these clash with their official theoretical orientation (Sandler, 1983). Further, the emphasis on adherence to treatment manuals in randomized controlled trials (which arguably limits the impact of therapists’ implicit theorizing) may also have reduced research interest in therapists’ everyday clinical thinking.

In recent years, however, the qualities of the therapist have come more into focus following the repeated finding that therapists differ in effectiveness (Baldwin & Imel, 2013). Further, in current definitions of evidence-based practice (American Psychological Association, 2005), clinical expertise is considered essential for the integration of research findings and clinical data in the context of patient characteristics, culture, and preferences. Given that experienced therapists develop private theories that may contain valuable information, one important area for further research concerns how practicing therapists conceptualize and implement notions of therapeutic action in different forms of psychotherapy and with different patient populations. Study II of this thesis explores experienced psychotherapists’ views of therapeutic action in psychoanalytic psychotherapy with young adults.

Methodological perspectives How to best approach the study of therapeutic action in psychotherapy has been discussed and debated among researchers since the dawning of modern process-outcome research (Orlinsky, Rønnerstad & Wilutzki, 2004). This touches on some fundamental epistemological and methodological issues involved in generating and testing clinical theories, such as the distinction between inductive or deductive reasoning and the use of qualitative or quantitative methods for analyzing data.

In the psychoanalytic tradition, theories of therapeutic action have historically been generated primarily through inductive reasoning based on clinical case studies. This approach has been heavily criticized by philosophical scholars such as Popper (1959) and Grünbaum (1984), but also by contemporary psychoanalytic researchers (Fonagy, 2003; Luyten, Blatt, & Corveleyn, 2006). For example, alluding to the epistemological distinction between “the context of discovery” and “the context of justification,” Fonagy (2003) argues that the psychoanalytic tradition may be rich in its “discoveries”, but has lagged far behind when it comes to the “justification” of these ideas.

While the traditional psychoanalytic case study may no longer be acceptable as a research method in and of it self, several more formal qualitative, inductive methods have been developed and applied in psychotherapy research (McLeod, 2013; Lutz & Knox, 2014). Typically, such methods involve the careful following of formalized steps and procedures for categorizing and interpreting data, as well as “bracketing” the researchers theoretical preconceptions. Some of these methods also include a second set of procedures based on deductive reasoning, which involve “testing hypotheses” within a qualitative dataset, presumably providing more robust and generalizable results (Strauss & Corbin, 1996). Kazdin (2007, 2009) points out that the strengths of inductive, qualitative research is that it can evaluate the richness and details of the psychotherapeutic process, including how and why change unfolds or why it does not unfold, from the particular perspective of the targeted respondents. Qualitative studies may be particularly fruitful in areas where there is a relative

6

lack of previous research or theory, illuminating novel ideas directly applicable in clinical practice, as well as providing grounded hypotheses for further research.

However, most research on curative and hindering factors in psychotherapy has been based on deductive reasoning and carried out with quantitative methods. Typically, such studies have correlated some variable of interest with outcome, sometimes statistically controlling for a particular confounding variable or variables (Crits-Christoph et al., 2013). Although the result of process-outcome research is sometimes discussed in terms of “cause and effect”, correlational designs do not allow causal inferences. While there exist methods and procedures that provide more thorough tests of causal relationships between process variables, very few psychotherapy studies have yet utilized them fully (Kazdin, 2007, 2009; Johansson et al., 2010). Consequently, most studies in this area can only be regarded as exploratory.

Several psychotherapy researchers have argued that, given the current state of the field, process-outcome research should move firmly in the direction of testing specific theoretical assumptions (Johansson & Høglend, 2007; Kraemer et al., 2002). However, others have pointed out that the field is still searching for potential mechanisms that may underlie the effects of different forms of psychotherapy (Kazdin, 2007). In this effort, both qualitative and quantitative studies may enhance our understanding of processes that promote or hinder change. Further, the establishment of any mechanism of change is likely to be reached only through the integration of several lines of research over time. Adopting a stance of methodological pluralism is probably still wise and such efforts could also help bridging the gap between academic research and clinical practice (Luyten et al., 2006; Kazdin, 2008). In the present thesis, qualitative methodology is employed in study I and II whereas study III and IV utilize quantitative methods in order to explore the main research questions.

Therapeutic action in psychoanalytic psychotherapy Turning to the tradition in focus for this thesis, a number of theoretical propositions regarding how psychoanalytic1 therapies work has been suggested since Joseph Breuer’s and Sigmund Freud’s first notion of the “cathartic cure” (Breuer & Freud, 1895/1955). Yet, almost 120 years after Breuer and Freud’s seminal work, there still exist no generally accepted theory of therapeutic action in the psychoanalytic discourse. Comparing psychoanalytic theories of change, Kernberg (2007) recognizes two fundamental mechanisms that are inherent (although differentially emphasized and conceptualized) in all contemporary psychoanalytic approaches – provision of insight and relational impact. These two notions are elaborated below in relation to empirical findings. Insight The idea that increasing the patient’s insight into his or her own unconscious psychological processes leads to change may be regarded the kernel of the psychoanalytic tradition. The

1 Throughout this thesis, the term “psychoanalytic” is used to refer to all therapeutic approaches that are theoretically grounded in the psychoanalytic tradition and is considered equivalent in meaning to “psychodynamic”.

7

notion can be traced back to Freud’s classic formulation “…where id was, there ego shall be” (Freud, 1933/1964, p. 80). From a psychoanalytic perspective, insight might be defined as the self-understanding of internal conflicts, intentions, anxieties, defenses and internalized expectations that underlie overt behaviors and symptoms (Fonagy & Kächele, 2009; Messer & McWilliams, 2007). In particular, increased understanding of recurrent patterns of maladaptive relating to self and others is often stressed. These patterns are assumed to occur in the therapeutic relationship (i.e., in the “transference”) and to reflect early learning experiences. Further, insight is typically thought to be achieved through the therapist’s skillful use of interpretations (Kernberg, 2007), particularly interpretations of the transference, linking the patients’ past and present experiences in the here and now. This may lead to the development of a new explanation or narrative regarding how and why symptoms occur, which makes them feel more manageable to the patient. Alternatively, insight may trigger a reappraisal of the usefulness of certain overt behaviors and provide the patient with an emotional release that frees him or her to act in new ways (Castonguay & Hill, 2007).

In terms of empirical support, there is some evidence indicating a positive link between increased insight during treatment and outcome (Grande, Rudolf, Oberbracht, & Pauli-Magnus, 2003; Grenyer & Luborsky, 1996; Høglend, Engelstad, Sørbye, Heyerdahl, & Amlo, 1994; Johansson et al., 2010; Kivlighan, Multon, & Patton, 2000). Some studies also suggest that increase in self-understanding precede symptom change (Grande et al, 2003; Kivlighan et al., 2000) and that insight may be a specific curative factor in psychoanalytic therapy compared to cognitive therapy (Connolly et al., 1999; Connolly Gibbons et al., 2009; Kallestad et al., 2010). In a recent review of eight studies investigating the insight-outcome correlation in psychoanalytic therapies, Crits-Christoph and colleagues (2013) found evidence for a large mean effect (r = .52). However, the number of studies was small and other reviews have found mixed results (Connolly Gibbons, Crits-Christoph, Barber, & Schamberger, 2007). Clearly, there is a need for more precise conceptualizations and measurements of the concept, as well as more research regarding how, when and with which patients, increased insight is essential for change (Barber, Muran, Crits-Christoph, McCarthy, & Keefe, 2013).

Since insight in the psychoanalytic sense of the term typically involves understanding of recurrent, maladaptive relationship patterns, it may be linked to the use of transference interpretations. The impact of transference work has been investigated in several studies (e.g., Connolly, Crits-Christoph, Shappel, Barber, Luborsky, & Shaffer, 1999; Crits-Christoph & Connolly Gibbons, 2001; Høglend, 1993, 2003; Ogrodniczuk, Piper, Joyce, & McCallum, 1999; Piper, Azim, Joyce, & McCallum, 1991; Ryum, Stiles, Svartberg, & McCullough, 2010; Schut et al., 2005). However, the results have been somewhat inconclusive, with some studies indicating a positive effect on treatment process and outcome whereas other studies suggest that the use of transference interpretations may have a negative impact, especially if used excessively (for a recent review see Høglend, 2014).

In a randomized controlled trial comparing psychoanalytic psychotherapy with or without transference work, Høglend and colleagues (2006, 2008) found that the effect of transference interpretations was moderated by the quality of patients’ object relations. Patients with lower quality of object relations, indicating more long-standing and severe interpersonal difficulties, made larger gains in the treatment format that included transference work. Further, this was particularly true in the context of weaker alliances (Høglend et al., 2011). Conversely, for patients with more mature object relations and high alliance, a negative

8

impact of transference work was observed. Thus, the inconsistent findings of previous studies might be explained by the moderation of particular patient variables. Further, a mediational analysis from the same trial showed that the positive effect of transference interpretations on long-term outcome was mediated by increased insight during therapy (Johansson et al., 2010). Thus, the moderate use of transference interpretations, leading to increased insight with patients who have poorer quality of object relations, may be the first evidenced-based mechanism of change established in psychoanalytic psychotherapy.

According to psychoanalytic theory, insight does not only refer to increased cognitive self-understanding but also involves a crucial emotional component (Fonagy & Kächele, 2009). The importance of so-called ”emotional insight” was recognized early by Freud (1914/1958) and Strachey (1934) as the most potent form of self-understanding. Emotional insight might be defined as the process by which the patient gains access to and integrates conflicted internal emotional states that were previously warded off through the use of maladaptive defense mechanisms (Davanloo, 1990; A. Freud, 1937; Malan, 1979). In line with this notion, exploration of affect has consistently been linked to more positive outcomes in psychoanalytic psychotherapy (for a meta-analysis, see Diener, Hilsenroth, & Weinberger, 2007). Although emotional processing is typically regarded as central change agent in several contemporary models of psychotherapy (Greenberg & Pascual-Leone, 2006, Whelton, 2004), some preliminary findings suggest that focusing on affect may have particular benefits in psychoanalytically oriented psychotherapy compared to cognitive therapy (Ulvenes et al., 2012).

Relational impact The second fundamental change mechanism in psychoanalytic discourse involves the impact of the therapeutic relationship. The idea that the experiential and transactional aspects of the relationship contain curative elements dates back to early psychoanalytic theorists such as Sandor Ferenczi and Otto Rank (Ferenczi & Rank, 1925). Typically, the underlying notion is that the interaction with the therapist can repair deficits, or correct skewed development, that originate in the patients’ experiences with family members and/or other important caregivers early in life. A developmental metaphor is often used (Mayes & Spencer, 1994) were the therapeutic relationship is seen as symbolically representing a parent-child relationship. As eloquently described by Loewald (1960), this position suggests that the process of change is: “…set in motion, not simply by the technical skill of the [therapist] but by the fact that the [therapist] makes himself available for the development of a new ‘object-relationship’ between the patient and the [therapist]” (p. 16). Typically, the acquisition of insight is regarded as an outcome of the therapeutic interaction rather than as a requisite for change in psychoanalytic models that stress the role of relational impact (Messer & McWilliams, 2007).

While there is ample evidence that the quality of the therapeutic relationship is linked with outcome across therapeutic modalities (Norcross, 2011), few studies have specifically tested psychoanalytically derived hypotheses regarding how the relationship leads to change. One psychoanalytic notion, which has influenced other schools of psychotherapy as well (see Castonguay & Hill, 2012), is the concept of corrective emotional experience (Alexander & French, 1946). This was originally defined as the patient “…re-experiencing the old, unsettled conflict but with a new ending” (p. 338), meaning that change occurs when the therapists act in ways that contradict the patients’ prior relational learning. Although the study of corrective

9

emotional experiences has been hampered by problems with definition and precision in measurement (Castonguay & Hill, 2012), some support for the notion might be inferred from different lines of research.

One such line of research concerns the processes involved when repairing ruptures in the therapeutic alliance (Christian, Safran, & Muran, 2012). Safran and Muran (2000) conceptualize the therapeutic alliance as a process of ongoing negotiation between patient and therapist. When ruptures in this negotiation occur, it is hypothesized that the mutual working through of the event towards re-establishing the alliance will typically involve corrective emotional experiences. In a small meta-analysis, Safran, Muran, and Eubanks-Carter (2011) found that the aggregated correlation of rupture-repair episodes to treatment outcome indicated a statistically significant medium effect (r = .24).

Another line of research is qualitative research focusing on patient perceived helpful events in treatment. For example, in a single-case study of a successful short-term dynamic treatment of panic disorder, Friedlander and colleagues (2012) found evidence for corrective emotional experiences in the patient’s own description of the psychotherapy process and the relationship with the therapist. Further, the patient narratives indicated that both specific (e.g., confrontation of defenses) and common factors (e.g., a positive alliance) interacted in producing corrective experiences. In another qualitative study involving 12 patients, corrective emotional experiences were typically related to repairing ruptures in the alliance as well as the therapist actively reaching out or normalizing patients’ concerns (Knox, Hess, Hill, Burkard, & Crook-Lyon, 2012).

An additional psychoanalytic notion of how the relationship contributes to change is through internalization (Gabbard & Westen, 2003). Internalization refers to the process by which the patient develops some attribute that originally belongs to the therapists. It may involve the patient developing a specific function, such as the ability to self-sooth through repeated experiences of being soothed by the therapists. The patient may also internalize the therapist’s affective attitudes toward him or her self or develop the ability to self-reflect by internalizing the therapists’ analytic capabilities (Gray, 1986; Kernberg, 1997; Kohut, 1984; Loewald, 1962). Specifically, the underlying idea of internalization is that the interaction with the therapist will gradually lead to changes in the patients’ implicit cognitive-affective schemas, or internal representations of self and others (Blatt & Behrends, 1987; Blatt, Auerbach, & Behrends, 2008; Blatt & Luyten, 2009).

Changes in internal representations during treatment have also been associated with long-term improvements in psychoanalytic therapies (Blatt, Zuroff, Hawley & Auerbach, 2010; Barber et al., 2013). There are also some studies supporting the idea that internalization might be involved in this process. For example, Quintana and Meara (1990) found that patients, over the course of therapy, tended to adopt attitudes towards themselves that were in line with the attitudes the therapists had towards them. Further, in contrast to merely “remembering” information about the therapists, the concept of internalization suggests that patients will bring forth representations of their therapists to regulate painful affects, moderate feelings of loneliness, and facilitate problem-solving and conflict resolution. Geller and Farber (1993) investigated the recall of internalized representation of the therapist in a sample of 206 patients in psychoanalytic psychotherapy or psychoanalysis. Supporting the notion of internalization, they found that representations of the therapist were most likely to be evoked when painful affects were experienced. Further, the ability to engage in internal dialogues

10

with one’s therapist, both during therapy and after termination, was significantly correlated with positive outcome. That the internal representation of the therapist might be used as a self-supportive strategy in the post-treatment process has been reported in other studies as well (Kantrowitz, Katz, & Paolitto, 1990; Falkenström, Grant, Broberg, & Sandell, 2007).

Lastly, one recent theoretical development that integrates the impact of the therapeutic relationship with both cognitive and emotional aspects of insight is the notion of mentalization (Allen, Fonagy & Bateman, 2008). This concept refers to the process through which an individual comes to understand self and others in terms of mental states and to use that information flexibly to explain and guide relationship behavior. Mentalization develops optimally in the context of secure attachment relationships where the caregiver uses marked mirroring to represent the child’s internal emotional states (Fonagy, Gergely, Jurist, & Target, 2002). Transferred to psychotherapy, development of patients’ capacity for mentalization is thought to be contingent on the therapist’s ability to represent and adequately mirror the patient’s shifting mental states in the therapeutic process (Allen & Fonagy, 2006; Allen et al., 2008). A few studies indicate that the capacity for mentalization may increase over the course of psychoanalytic psychotherapy and it has been found to relate to change in personality functioning (Levy et al., 2006), as well as to change in anxiety symptoms (Rudden, Milrod, Target, Ackerman, & Graf, 2006). However, other studies have not found mentalization to be clearly related to change (Karlsson & Kermott, 2006; Vermote et al., 2010) and more studies are needed before any conclusions can be drawn regarding its importance for different patient populations or in different treatment models (Barber et al., 2013).

In summary, there is some empirical support for the fundamental mechanisms of change proposed by contemporary psychoanalytic theories (i.e., increased insight, relational impact, improved mentalization). However, the available evidence is limited, sometimes contradictory, and the research is hampered by difficulties in defining and measuring the concepts. Few studies go beyond a basic correlational design and there is very limited empirical data linking specific interventions and strategies to any of these mechanisms. We also know very little about how insight and the relational impacts may interact with each other and with particular patient and therapist factors, or with hindering aspects in treatment. It is quite likely that the impact of both insight and relationship are moderated by several unknown variables. Although the evidence-base for psychoanalytic therapies is growing for a range of psychiatric conditions (Abbass et al., 2014; Barber et al., 2013; Leichsenring & Klein, 2014), we need to know more about the factors responsible for change.

Alliance and attachment As stated above, there is a large amount of research indicating that the quality of the therapeutic relationship is related to treatment outcome across therapeutic modalities (Norcross, 2011). This body of research suggests that several components (such as therapist empathy, congruence, positive regard and the therapeutic alliance) are linked to change. Two aspects of the therapeutic relationship that are central to this thesis are introduced and elaborated below – the therapeutic alliance and attachment to therapist.

11

The therapeutic alliance The concept of the therapeutic alliance has evolved from of the psychoanalytic tradition (Freud, 1912/1966; Sterba, 1934; Zetzel, 1956; Greenson, 1967) into its present-day pan-theoretical definition (Bordin, 1979; Muran & Barber, 2010). Bordin’s well-known conceptualization separates the alliance into three main components: agreement on goals; consensus on the therapeutic task; and the emotional bond between the patient and the therapist. In contrast to the original psychoanalytic formulation that emphasized unconscious distortions in the relationship between therapist and patient (i.e., Freud, 1912/1966), this definition highlights the conscious, collaborative aspects of the therapeutic relationship (Horvath, 2006).

The therapeutic alliance is beyond comparison the most studied process variable in psychotherapy research and the strength of the alliance has repeatedly been associated with good outcome. In a recent large-scale meta-analysis (involving 190 studies and over 14.000 treatments) the aggregated alliance-outcome correlation was found to be r = .275, indicating that the alliance accounts for about 7–8% of the variance in in outcome (Horvath, Del Re, Fückinger, & Symonds, 2011). The therapeutic alliance proved to be a robust predictor of outcome regardless of which particular alliance measure was used or if patients, therapists, or observers rated it. Further, the alliance-outcome relationship was consistent regardless of when it was measured in the treatment process (e.g., early, mid-treatment, late in therapy), which kind of problem the patient sought treatment for (e.g., depression, anxiety) or which therapeutic modality the patient received (e.g., cognitive-behavioral therapy, interpersonal therapy, psychoanalytic psychotherapy). No other currently known variable has been found to account for more of the variance in outcome across different forms of psychotherapy (Muran & Barber, 2010; Norcross, 2011).

However, although the alliance is associated with change, this does not imply that the relationship is causal. The idea that alliance leads to change has been questioned repeatedly (e.g., DeRubeis, Brotman, & Gibbons, 2005). Critics have pointed to studies showing that early symptom improvement predicts later alliances (e.g., Barber et al., 1999; Strunk, Brotman, & DeRubeis, 2010; Strunk, Cooper, Ryan, DeRubeis, & Hollon, 2012), suggesting that a strong alliance may be the result of specific treatment factors employed early in treatment. Recent research that address this issue indicates that although symptomatic improvement early in treatment predicts the alliance, the alliance also predicts subsequent change, pointing towards a reciprocal relationship between therapeutic gains and the alliance (Falkenström, Granström, & Holmqvist, 2013, 2014; Tasca & Lampard, 2012; Zilcha-Mano, Dinger, McCarthy, & Barber, 2013). Consequently, the alliance seems to be an essential ingredient in the change process and not merely a result of early improvement.

Still, the fundamental question regarding precisely how the alliance leads to change remains unanswered. Some researchers have raised concerns that this question will continue to be difficult to address since the concept is not distinctly defined (Horvath, 2006), nor is it anchored in any particular theory of therapeutic change (Castonguay, 2006). Some have even proposed that the alliance may have outlived its usefulness and that the field should move on to other constructs (Safran & Muran, 2006). Clearly, testing specific theoretical explanations of the alliance-outcome linkage is an important task for future research.

12

Attachment to therapist John Bowlby’s (1969, 1973, 1980, 1988) attachment theory has become one of the most influential theoretical frameworks in clinical practice, as well as psychotherapy research, in the last two decades (Obegi & Berant, 2009). In the clinical literature, several authors have elaborated on how attachment theory may inform individual psychotherapy (e.g., Eagle, 2013; Holmes, 2001, 2010; Fosha, 2000; Muller, 2010, Pearl, 2008; Wallin, 2007). A number of studies have also investigated how patients’ (as well as therapists’) global attachment orientations, or “attachment styles”, may influence treatment processes and outcome (for reviews, see Bernecker, Levy, & Ellison, 2014; Daniel, 2006; Diener & Monroe, 2011; Levy, Ellison, Scott, & Bernecker, 2011; Slade, 2008; Smith, Msetfi, & Golding, 2010). Based this body of research, it has been proposed that the therapy process may benefit if therapist use knowledge of their patients’ attachment style as a guide for understanding patient responses and attuning their interventions (Levy et al., 2011).

Bowlby (1988) also suggested that the therapeutic relationship is a form of attachment relationship in itself. Specifically, he proposed that: “…in providing his patient with a secure base from which to explore and express his thoughts and feelings, the therapist’s role is analogous to that of a mother who provides her child with a secure base from which to explore the world.” (p. 140). In recent years, several scholars have elaborated on the idea that the therapist is an “attachment figure” who optimally functions as a secure base for the patient (e.g., Farber, Lippert, & Nevas, 1995; Farber & Metzger, 2009; Obegi, 2008; Mallinckrodt, 2010). Clearly, Bowlby’s analogy does not imply that the therapist should strive to become a “new parent” to the patient. Rather, it suggests that the same psychobiological systems underlying human attachment are likely to be involved when establishing and maintaining a trustful and collaborate therapeutic relationship. Obegi (2008) points out that there are several structural similarities between therapeutic relationships and attachment relationships: both endure over time and involve seeking out a specific, non-interchangeable individual for help and support in a time of distress. Thus, besides the influence of patients’ (and therapists’) global attachment styles, the development and quality of the specific attachment formed between patient and therapist may be important for psychotherapy process and outcome.

Shaver and Mikulincer (2009; see also Mikulincer, Shaver, & Pereg, 2003) have integrated findings from attachment research into a model of adult attachment system functioning that may be useful for understanding the development of patient-therapist attachment and its impact in the psychotherapeutic process. According to this model, when a sufficiently intense threat is perceived (consciously or unconsciously), the attachment system is activated and prompts the individual to seek proximity to external or internalized attachment figures. Based on their developmental experiences, however, individuals will vary in their expectations regarding how attachment figures are likely to respond to their physical and emotional needs. If attachment figures are expected to be available and responsive, security-based strategies to seek comfort, emotional proximity, and support will typically be employed. In contrast, individuals who do not expect attachment figures to be responsive will experience a heightened sense of distress and use secondary attachment strategies. If it is anticipated that intensified proximity-seeking behaviors may result in some measure of responsiveness from others, hyperactivating strategies will likely be used. In contrast, if proximity seeking is expected to be useless or even lead to further emotional injury, deactivating strategies are more likely evoked. The model may also be extended to

13

individuals whom, due to developmental deficits and/or disruptive attachment experiences such as trauma or repeated physical or emotional abuse/neglect, have not developed an organized secondary strategy and may resort to disorganized strategies.

Since the attachment system will naturally activate when stressful emotional material is approached in the psychotherapy process, the patient’s internal working models of self and others (and the associated secondary attachment strategies) will start to color the relationship with the therapist at some point. The interaction with the therapist will then determine whether the relationship formed over time will have the basic qualities of a secure attachment relationship or if insecure strategies will dominate (Daly & Mallinckrodt, 2006; Mallinckrodt, 2010; Romano, Fitzpatrick, & Janzen, 2008). From an attachment perspective, the development of a sufficiently secure therapeutic relationship is fundamental for change since it enables the patient to explore, access, process and integrate distressing material (Bowlby, 1988; Eagle, 2013; Holmes, 2001, 2010; Fosha, 2000; Muller, 2010, Pearl, 2008; Wallin, 2007). In contrast, an insecure attachment to the therapist will inhibit the exploratory system, making integration and change less likely to occur. In line with the clinical literature, Shaver and Mikulincer’s (2009) model suggests that a sufficiently responsive therapist will promote a broaden-and-build cycle of attachment security which will lead to reduced distress and, over time, change the patients’ internal working models of self and others.

Assessing attachment to therapist The study of patient-therapist attachment is a new and still developing area in psychotherapy research. One important methodological advance has been the construction of specific measures for assessing the quality of patient attachment to therapist. Today, two self-report instruments exist. Mallinckrodt and colleagues (1995) developed the Client Attachment to Therapist Scale (CATS), which is a 36-item self-report questionnaire including three subscales: Secure, Preoccupied/Merger, and Avoidant/Fearful attachment to the therapist. CATS has good internal consistency and test-retest reliability, and the subscales correlate in expected ways with measures of adult attachment, working alliance, and object relations (Mallinckrodt et al., 1995, 2005). The Components of Attachment Questionnaire-Therapist (CAQ-T), developed by Parish and Eagle (Parish & Eagle, 2003a, 2003b), is a 34-item self-report measure that was designed to assess the impact of specific aspects of attachment in the therapeutic relationship: proximity seeking, separation protest, safe haven, particularity, secure base, and stronger/wiser. The mean of all 34 items in the scale is understood as a measure of the intensity of the patient’s overall attachment to the therapist. Psychometric data for CAQ-T suggest satisfying internally consistency and some support for its construct validity has also been reported (Parish & Eagle, 2003a, 2003b; Saypol & Farber, 2010).

Although the available data indicate that both CATS and CAQ-T are reliable and valid measures, the use of self-report to assess the patients’ attachment to their therapists may have specific limitations. For example, self-assessment might be problematic since attachment involves unconscious, implicit-procedural processes (Maier, Bernier, Pekrun, Zimmermann, & Grossmann, 2004). Observer ratings of interview material may be an appropriate alternative (de Haas, Bakermans-Kranenburg, & van Ijzendoorn, 1994; Jacobvits, Curran, & Moller, 2002). In one study, Diamond and colleagues (2003) adapted the interview protocol of the Adult Attachment Interview (AAI), forming the Patient–Therapist Adult Attachment Interview (PT-AAI), which was used to evaluate patients’ attachment states of mind in

14

relation to the therapist in a small sample of borderline patients (n = 10). The results indicated that therapist-specific attachment early in therapy was in high concordance with patients’ AAI classifications, indicating preliminary validity of the approach. Although promising, the use of the AAI coding scheme may require extensive training and the reliability of the coding procedure has not been evaluated in the context of the therapeutic relationship. Further, since the PT-AAI approach assesses attachment to therapist categorically rather than dimensionally, it might have limited utility for research purposes (Ravitz, Maunder, Hunter, Sthankiya, & Lancee, 2010). Other measures that aim to assess the quality of patient attachment to therapist from an observer perspective have recently been developed (Lilliengren et al., 2014; Talia et al., 2014). Study III of the present thesis involves an initial evaluation of one such approach (Patient Attachment to Therapist Rating Scale, PAT-RS; see below).

The impact of patient-therapist attachment on psychotherapy process A small but growing body of research indicates that the quality of patients’ attachment to their therapist relate to specific in-session processes in theoretically consistent ways. For example, patients who develop a more secure attachment to their therapist, as measured with the secure subscale of CATS, have been found to engage in deeper and smoother exploration in therapy (Mallinckrodt et al., 2005; Romano et al., 2008). Additionally, in a study applying CAQ-T, Saypol and Farber (2010) found that secure attachment to the therapist was associated with higher levels of patient self-disclosure in treatment. In a study investigating relationship-building episodes, Janzen and colleagues (2008) found that when patients disclosed distressing thoughts or feelings in therapy, the level of secure attachment to the therapist increased. Further, as the level of security increased so did the level of exploration, relief, and perceived therapist support, and the patients perceived their sessions as more impactful. Thus, the results indicate a reciprocal relationship between patient disclosure, increased secure attachment to therapist and session impact, which is consistent with the broaden-and-build cycle of attachment security (Shaver & Mikulincer, 2009).

Taken together, these studies support the idea that a secure attachment to the therapist will facilitate deeper emotional exploration in therapy, as hypothesized by Bowlby (1988). Further, insecure attachment to therapist (avoidant-fearful in particular) has consistently been associated with less exploration, depth, disclosure, and readiness for change in treatment, as well as a weaker alliance (Bachelor, Menunier, Laverdiére, & Gamache, 2010; Fuertes et al., 2007; Lunsford, 2010; Mallinckrodt et al., 2005; Romano et al., 2008). Hence, insecure attachment strategies that are evoked in the therapeutic relationship may form important obstacles to the therapeutic process.

However, a crucial question is whether the quality of patients’ attachment to their therapist is related to treatment outcome, as suggested by attachment theory. This question has only been addressed in a couple of studies so far and the results have been somewhat inconsistent. Sauer and colleagues (2010) found that higher levels of secure attachment to the therapists (measured at session three) predicted reductions in psychological distress over time in a sample of 95 moderately distressed patients treated with brief therapy at a university clinic. In contrast, Wiseman and Tishby (2014) recently found that secure attachment to therapist (measured at session five) was unrelated to change in a sample of young adults undergoing one year of psychodynamic psychotherapy. Nevertheless, lower attachment avoidance in relation to the therapist early in treatment predicted improvements, perhaps

15

indicating that the absence of insecure-avoidant attachment strategies (i.e., deactivation), which may block the treatment process (Muller, 2010), was more important than the presence of secure attachment to therapist. Clearly, more studies are needed to investigate the influence of both secure and insecure forms of attachment to therapist on therapy process and outcome. The association between secure attachment to therapist at termination and outcome is explored in study IV of this thesis.

Differentiating secure attachment to therapist from the therapeutic alliance There is a notable conceptual overlap between secure attachment to therapist and the therapeutic alliance, particularly regarding the bond-aspect of the alliance (Obegi, 2008). Bordin (1979) defined the alliance bond as the sense of mutual liking, trust and respect between patient and therapist. He also suggested that the strength of the alliance bond may become increasingly important as the patient approaches more threatening material in the treatment process: “…some basic level of trust surely marks all varieties of therapeutic relationships, but when attention is directed toward the more protected recesses of inner experience, deeper bonds of trust and attachment are required and developed” (p. 254). Hence, Bordin’s conceptualization of the bond part of the alliance seems to mirror the development of a secure attachment to the therapist to a large degree.

In line with this, studies using the CATS typically report strong correlations (.60 - .80) between secure attachment to therapist and patients’ self-rating of the alliance (Bachelor et al., 2010; Lunsford, 2010; Mallinckrodt et al., 1995, 2005; Romano et al., 2008; Sauer et al., 2010). Addressing this overlap, Mallinckrodt and colleagues (2005) concluded that: “…a high quality working alliance and a secure attachment to one’s therapist appear to have many features in common and are probably perceived relatively similarly by clients” (p. 97).

However, although the constructs overlap to some extent, there are also important conceptual differences (Mallinckrodt, Coble, & Gantt, 1995; Mallinckrodt, Porter, & Kivlighan, 2005; Obegi, 2008). For example, whereas the alliance is primarily concerned with the patient’s conscious perception of the collaboration with the therapist (Horvath, 2006), attachment to therapist reflects the function of the relationship, i.e., how the patient uses the bond and relates to the therapist on both conscious and unconscious emotional levels. Put differently, attachment to therapist specifies the quality of the bond (i.e., secure or insecure) the patient develops to his or her therapist.

Further, attachment theory suggests that the therapeutic attachment will contain several specific relational components (Mallinckrodt, 2010; Mallinckrodt et al., 1995, 2005; Obegi, 2008; Parish & Eagle, 2003a; see also Table 2 below). For example, attachment to therapist involves a “secure base”-component, indicating that the safety of the relationship enables the patient to approach and explore novel or frightening experiences. Other components include feeling comforted by the therapist (“safe haven”), turning to the therapist for knowledge and guidance (“stronger and wiser”), having a desire to seek out and maintain closeness to the therapist in times of distress (“proximity seeking”), viewing the therapist as a unique and non-interchangeable person (“particularity”), experiencing the therapist as sufficiently responsive (“responsiveness”), viewing the relationship as affectively charged (“strong emotions”), reacting with distress when separated (“separation anxiety”), and forming a representation of the therapist that may be evoked in times of distress (“mental representation”).

16

Although different measures of the alliance may capture some of these components, the alliance concept itself does not include any of them specifically (Obegi, 2008). Accordingly, a strong therapeutic alliance may not necessarily indicate that the patient and therapist have also developed a secure attachment relationship. Some patients may report a strong alliance even though they do not use the therapeutic relationship as a “secure base” or experience it as a “safe haven.” For example, Lunsford (2010) found a positive correlation between the Avoidant/Fearful subscale of the CATS and the Tasks subscale of the Working Alliance Inventory (WAI), suggesting that some patients may agree on and even carry out particular therapeutic tasks although they may not establish a trustful, emotional bond with the therapist. In the same study, the Preoccupied/Merger subscale of the CATS was unrelated to Goals and Tasks, but positively associated with the Bond subscale of the WAI, indicating that patients who formed an insecure-anxious attachment to their therapist also tended to report a strong alliance bond. Consequently, a strong alliance bond may sometimes reflect an insecure attachment strategy rather than a constructive use of the therapeutic relationship.

It is also possible that some patients who have developed a secure attachment to the therapist may report a temporary weak alliance, for example in connection with an alliance rupture (Safran & Muran, 2000) or when exploring negative transference reactions (Woodhouse, Schlosser, Crook, Ligiéro, & Gelso, 2003). Thus, one might hypothesize that the therapeutic alliance may vary during therapy while a secure attachment to the therapist probably develops more progressively throughout the psychotherapeutic process and is likely to stay relatively stable once established.

However, in order to investigate such hypotheses the alliance and secure attachment to therapist needs to be assessed distinctly; hence, the large overlap found between measures of these constructs is problematic. For example, Sauer and colleagues (2010) could not differentiate the impact of secure attachment to therapist from the alliance due to the strong co-variation between the measures. It is possible that observer-rated measures of secure attachment to therapist may be useful in this regard, which is considered in study IV of this thesis.

Young adults in psychoanalytic psychotherapy All four studies in this thesis were conducted on data from patients that were between the ages of 18-24 when entering treatment. This period in life is often referred to as young adulthood or emerging adulthood (Arnett, 2000, 2007; Jacobsson, 2005; Szajnberg & Massie, 2003; Tanner & Arnett, 2009). For most people in industrialized Western countries, this is typically a phase of profound external and internal change as various possibilities in life are explored, elaborated and consolidated. During this time, many move away from their family of origin, finish their education, become financially independent, move in with someone, perhaps also marry or become a parent. It is a time when society expects the individual to begin to make independent decisions and accept responsibility for them. By the end of this period most people will have made life choices that will have and enduring effect for the remainder of their adult lives. When later asked to consider the most important events in their lives, many adults name events that took place during this period (Martin & Smyer, 1990).

17

In the last couple of decades, young adulthood has also been associated with increased psychological distress in several Western countries (Evans, 2009; Grant & Potenza, 2009; McGorry, Purcell, Goldstone, & Amminger, 2011; Socialstyrelsen, 2013a; 2013b). This trend seems to be typical for societies where young people are given an extended time to explore their future possibilities (Arnett, 2000, 2007). When Erikson (1959, 1968) formulated his theory of psychosocial development he described adolescence as a time of identity crisis and inner turmoil, while the crisis in young adulthood was one of intimacy versus isolation. Nowadays, when young people have more time to explore their possibilities, this identity crisis seems to be delayed (Arnett, 2000, 2007; Côté, 2000; Robbins & Wilner, 2001). Accordingly, the increase in distress among young adults may indicate that the transition from adolescence to adulthood has become more complicated and conflicted in our time (Jacobsson, 2005).

From a psychoanalytic perspective, the developmental task of young adulthood concerns the consolidation of ego-capacities required for the life and career decisions at hand. When this consolidation is only partially or unevenly achieved, for example due to the activation of inner conflicts or deficits related to early developmental experiences, maladaptive coping strategies may be triggered, leading to increased symptoms and interpersonal distress (Adatto, 1980; Emde, 1985; Escoll, 1987). Although there are no randomized controlled trials of psychoanalytically oriented psychotherapy that specifically target young adults, several naturalistic studies indicate that such treatments are effective in this population (Baruch & Fearon, 2002; Blatt, Steyner, Auerbach, & Behrends, 1998; Falkenström, 2009; Harpaz-Rotem & Blatt, 2005; Jeanneau & Winzer, 2007; Philips, Wennberg, Werbart & Schubert, 2006; Werbart, Forsström, & Jeanneau, 2012). The effect sizes of psychoanalytic psychotherapy for young adults in routine care are comparable to the effects commonly reported in randomized controlled trials (Falkenström, 2009). Further, the effects tend to increase after treatment termination (Lindgren, Werbart & Philips, 2010; Werbart et al., 2012).

Yet, according to the clinical literature (e.g., Barnett, 1971; Escoll, 1987; Jacobs, 1988; Pearl, 2008; Perelberg, 1993), the developmental tasks of young adults may sometimes collide with the process of psychoanalytic psychotherapy. Jacobs (1988) suggests that, since young adults are in a transient life situation, they may be more focused on “real-world issues” and therefore lack motivation for reflecting upon themselves and their past. Further, young adults are typically still in the process of separating from internal representations of parents and occupied by conflicts regarding dependency and intimacy. Engaging in an intimate therapeutic relationship may evoke particular stress since young adult patients may not want to be caught up in transference feelings that pull them backwards in threatening ways (Jacobs, 1988). From an attachment perspective, providing a secure base for the young adult patient may also involve particular challenges for the therapist due to the conflict between patients’ attachment to their parents and their efforts to become autonomous (Pearl, 2008). In line with these clinical observations, research suggests that establishing a working alliance with young adults may pose particular difficulties and young adults have a higher risk of dropping out from treatment (Swift & Greenberg, 2012).

On the other hand, young adulthood has also been identified as a period in life where there is a strong potential for personality change and emotional growth (Tanner & Arnett, 2009). Change in personality may be considered a natural outcome related to young adults

18

activities in establishing careers, find their personal identity and committing to close relationships. The culture of Western societies typically promotes such activities during this period, which may act as a catalyst for change. Thus, young adulthood may be regarded both a “sensitive period” and as a “window of opportunity” when it comes to the risk for developing of psychological distress, as well as engaging in psychotherapy.

19

AIM OF THE THESIS The overall aim of this thesis is to explore therapeutic action in psychoanalytic psychotherapy from multiple perspectives (patient, therapist, and observer) using different methodological approaches (qualitative and quantitative). The primary objective is to contribute to the empirical literature regarding curative and hindering factors in psychotherapy in general, and therapeutic action in psychoanalytic psychotherapy in particular. Further, since the research project that provided the data for this thesis targeted young adults (aged 18-24), implications for psychoanalytic psychotherapy with this population have particular attention.

The thesis includes four separate studies that were carried out consecutively during the study period (2002–2013). Each study was designed in relation to the specific research questions that emerged from the findings of the previous study. An overview of the specific aim, sample size, material, data analytic strategy and main methodological approach of each study is presented in Table 1.

Table 1. Overview of the included studies

Aim n Material Data analytic strategy Methodology

Study I To explore patients’ view of therapeutic action in psychoanalytic psychotherapy

22 Patient interviews

Grounded theory Qualitative

Study II To explore experienced therapists’ view of therapeutic action in psychoanalytic psychotherapy

16 Therapist interviews

Grounded theory Qualitative

Study III To investigate the psychometric properties of a new rating scale for patient-therapist attachment

70 Self-report and expert-rated measures

Standard statistical methods

Quantitative

Study IV To investigate the relationships between secure attachment to therapist, alliance, and outcome

70 Self-report and expert-rated measures

Linear mixed-effects models

Quantitative

Study I and II were specifically aimed at exploring the patients’ and therapists’

subjective views of therapeutic action. Some of the research questions were: what do patients and therapists perceive as curative in psychoanalytic psychotherapy? What hinders change, according to patients’ and therapists’ views? How are these factors interrelated? Using an experience-near, inductive methodological approach, the objective was to construct tentative models of how curative and hindering factors interact from the view of the respective respondents.

20

Based on the main findings of study I and II, our attention was directed towards the quality of patient-therapist attachment for psychotherapy process and outcome. In order to study patient attachment to therapist using deductive reasoning and quantitative methods, a new instrument called the Patient Attachment to Therapist Rating Scale (PAT-RS) was developed. The specific aim of study III was to investigate the psychometric properties of this rating scale using standard statistical methods.

The aim of study IV was to explore the relationships between secure attachment to therapist, alliance and outcome. Specifically, we sought to investigate whether secure attachment to therapist was associated with changes during therapy and could predict any changes during the 1.5 year follow-up period, independent of patient-rated alliance. The study used deductive reasoning and quantitative, statistical methods to address these questions.

21

METHOD

Participants and procedures Research context The included studies are all based on data from the Young Adult Psychotherapy Project (YAPP). YAPP was a naturalistic, prospective, and longitudinal study of young adults (aged 18–25) in psychoanalytic psychotherapy conducted in Stockholm, Sweden. The overall aim of the project was to evaluate the effectiveness of psychoanalytic psychotherapy for young adults as routinely practiced at the Institute of Psychotherapy, where subsidized psychotherapy was provided for people with various psychological problems.

Inclusion of patients took place between 1998 and 2002. The patients applied through the Institute’s telephone service and were admitted as openings became available. A few patients were also referred to the project from psychiatric outpatient clinics in the greater Stockholm area. On the basis of clinical judgment of suitability and patient motivation, accepted patients were offered either individual or group therapy. A total of 134 patients were included (73% female; mean age = 22, SD = 2.2) of which 92 were allocated to individual psychotherapy and 42 to group therapy. No formal psychiatric diagnoses were assessed, but the main problem areas were identified and categorized from intake interviews (Wiman & Werbart, 2002). The most common complaints were low self-esteem (97%), depressed mood (66%), anxiety (55%), and conflicts in close relationships (66%). Further, about one-third had self-reported personality disturbance according to the DSM-IV and ICD-10 Personality Questionnaire (DIP-Q; Ottosson et al., 1995, 1998). All patients were assessed with standardized self-report measures and interview instruments administered at intake, at termination, and at two follow-up assessments: 1.5 and 3 years after termination, respectively.

The psychotherapies in YAPP were aimed at overcoming developmental arrests in young adulthood and improving the patients’ adaptive capacity. The goals, duration, and frequency of therapy were adjusted to each patient’s needs with the possibility of renegotiating during treatment. The project involved a total of 37 therapists who all shared a psychoanalytical frame of reference, even if they were working quite autonomously and had varying preferences regarding theory and technique. During the project, the therapists met weekly in clinical teams to discuss clinical experiences and treatment problems. However, no manual or treatment fidelity checks were used, rather the therapies were carried out in accordance with the naturalistic setting of the project.

The overall result of the project indicated that the patients made moderate (Cohen’s d > 0.50) to large (d > 0.80) improvements on primary outcome measures between intake and termination on average (Philips et al., 2006). These benefits were maintained at follow-up (Lindgren et al., 2010). The specific selection procedures and subsamples for each study are described below.

22

Study samples Study I. The subsample for the first study was selected on the basis of availability. At the start of the study (July 2002) 26 patients had terminated their individual psychotherapy in YAPP. However, interview data was missing in four of cases and, consequently, a total of 22 patients were included. Three were male (14%) and 19 female (86%) and their average age was 22.5 years (SD = 2.2, range 19–25) at the beginning of therapy. In terms of their demographic data, eight (36%) lived alone, 5 (23%) lived with parents, and 9 (41%) lived with a partner. None was married or had children. Many of the patients were working full time (n = 10, 45%), followed by full-time studies (n = 8, 36%) and working in combination with studies (n = 4, 18%). Sixteen of the patients (73%) were born in Sweden and had both parents of Swedish origin. In all, 17 patients (77%) had at least one parent with a university degree. Nine of the patients (41%) reported having had previous outpatient or inpatient psychiatric treatment. The average length of psychotherapy for this subsample was 18.6 months (range 7–32 months) with a frequency of one (12 cases) or two (10 cases) sessions weekly. The reported problems areas that brought the patient to therapy, as well as mean pre- and post levels of self-reported and expert rated symptoms and functioning, were very similar to the YAPP sample as a whole.

Study II. The subsample of the second study consisted of the 16 therapists who had treated the 22 patients included in study I. Four therapists were male and 12 were female, all of Scandinavian origin. As to their profession, two were physicians, seven were psychologists, and seven were social workers. The therapists were all highly educated, middle-aged specialists in psychoanalysis (n = 7) or psychoanalytic psychotherapy (n = 9) with extensive clinical experience. Besides their work as psychotherapists, they were all engaged as instructors and supervisors in the Advanced Psychotherapy Training Program at the Institute of Psychotherapy. The mean time in clinical practice after attaining their psychotherapy license was 11 years (range 3–16, SD = 3.97). Eleven therapists treated one patient (of the 22 included in study I) each, four had two patients, and one had three patients.

Study III and IV. All patients who received individual psychotherapy in YAPP and participated in the interview at termination (thereby enabling the assessment of attachment to therapist) were of interest for study III and IV. Of the 92 patients allocated to individual psychotherapy, nine never started treatment, one dropped out during the project, and 12 did not show up for the assessments at termination. Consequently, the available sample consisted of 70 patients. Fourteen of these (20%) were male and 56 (80%) female and the average age was 22 years (SD = 2.1, range 18–26) at start of therapy. Most (n = 23, 32.8%) lived alone, 17 (24.2%) with a partner, eight (11.4%) with a friend or in a student dormitory; 17 (24.2%) reported still living with one or both of their parents, and five (7.1%) with another close relative (i.e., sibling, cousin or grandparent). At intake, none of the patients was married or had children. Twenty-three (32.8%) were fulltime students, 16 (22.8%) worked full-time, and 14 (20%) reported combining part-time work with part-time studies. Nine (12.8%) studied part-time and four (5.7%) worked part-time without any other occupation. Three (4.2%) were unemployed or on longer-term sick leave; however, 18 (25.7%) reported being unable to work or study at the beginning of therapy due to poor mental and/or physical health. In terms of cultural and socioeconomic background, 65 (92.8%) of the patients were born in Sweden, 18

23

(25.7%) had at least one parent of foreign origin. Most came from highly educated families where at least one parent had an academic degree (n = 51, 72.8%). Half of the sample (n = 35, 50%) had experienced parental divorce during childhood.

A total of 32 therapists (21 female and 11 male) treated the 70 patients in this subsample. Fourteen of the therapists were psychoanalysts, 16 were licensed psychotherapists and two had basic psychotherapy training. Their professional background varied: six physicians/psychiatrists, 12 psychologists, 13 social workers, and one hand another academic degree. On average, the therapists had about 10 years of experience practicing psychotherapy. Twelve of the therapist treated one patient, 13 had two patients, and seven had three or more patients in the sample. The average length of treatment was 23 months (SD = 13.0, range 2–55) with a frequency of sessions equally distributed between once or twice weekly. The outcome for this sample was very similar to the reported results for the whole YAPP population (Lindgren et al., 2010) and indicated large (Cohen’s d > 0.80) to moderate (d > 0.50) within-group effects between intake and termination, but no or only small gains during follow-up (d < 0.20).

Ethical considerations YAPP was approved by the Regional Ethical Review Board at Karolinska Institute in Stockholm, Sweden. All patients were informed about the overall aims and procedures of the project before treatment started and gave their written consent to participate. Interviews and self-report measures were coded and all personal data was kept apart from the research databases throughout the project. In study I and II, particular efforts were made to preserve patient and therapist confidentiality when selecting and presenting quotations from the interviews in the results sections of the papers.

Measures and methods Interviews All four studies in this thesis utilize interview data that was collected shortly after treatment termination with two semi-structured interview protocols: The Private Theories Interview (PTI; manual published in Werbart & Levander, 2006, 2011) and the Object Relation Inventory (ORI; Auerbach & Blatt, 1996). The PTI is an in-depth interview that aims at collecting each patient’s subjective meaning making concerning his or her presenting problems, the background to those problems and his or her own ideas of cure, as well as descriptions of changes during and after psychotherapy. In line with a basic phenomenological approach, the interviewer starts with open-ended questions and then asks the patient to elaborate their answers in these main areas as well as give concrete examples and illustrative episodes when applicable.

In the ORI, the patient is requested to give a description of his/her mother, his/her father, him/ herself, and of the therapist. The spontaneous response was followed by an “inquiry” in which the interviewer enquiringly repeated adjectives or descriptive words mentioned by the patient, for example, “You said understanding?” The information obtained

24

with the ORI is used for rating the level of complexity and differentiation in mental representations of self- and others (see the Differentiation-Relatedness Scale below).

The therapists were interviewed with the same protocols shortly after termination with each patient they treated in YAPP. The protocols were adjusted so that the therapists were asked to elaborate their own views of the patients’ difficulties, as well as what contributed to, or hindered treatment, in each particular case.

Nineteen experienced psychotherapists and psychologists were trained in the specific techniques of each protocol and carried out the interviews in YAPP. The interviews lasted about 60 minutes together and were audio-recorded and transcribed verbatim.

Qualitative methodology In study I and II, interview material obtained with the PTI was analyzed with qualitative methodology. The specific data analytic strategy followed Grounded Theory (GT; Strauss & Corbin, 1998; Hartman, 2001). GT was originally developed within sociology (Glaser & Straus, 1967) but has increasingly been applied in other research areas (e.g., economics, education, health sciences and psychology). The aim of GT is to generate theory regarding social phenomena in a substantive area, based on the systematic analysis of qualitative data (although quantitative data may also be included in a GT-analysis). The method is based on the principles of constant comparative analysis and saturation of data, which involves examining the raw data closely to derive and organize categories from it until all data is accounted for. Thus, GT is an inductive and interpretative, “bottom-up” approach, and philosophically it has been regarded as a form of methodological hermeneutics (Rennie, 2006). The finished GT-analysis is typically presented as a conceptual model that centers on a core concept that, to mix methodological metaphors, “explains” as much of the “variance” in the qualitative data as possible (Hartman, 2001).

There are no universally agreed upon rules regarding how to conduct a “proper” GT-analysis and even the originators of the method debate its basic procedures (see Rennie, 1998). Following Strauss and Corbin (1998), a set of common steps and procedures that typically characterizes GT were employed in study I and II. The analyses started with open coding which involved reading the selected interview material in detail, line by line, and provide an open code describing the main content of the narratives. Codes that seemed closely related in meaning, theme, or content were grouped into categories. As more and more categories emerged in open coding, a process of axial coding was initiated, which involved comparing and revising categories and identifying subthemes and relationships between categories. Finally, the categories and relationships were integrated into a conceptual model in selective coding. Figure 1 provides a graphical illustration of the coding process as implemented in study II.

Throughout the coding process, memos were written describing the properties and dimensions of the categories, as well as notes on the coder’s associations to established theoretical constructs. These memos were also used in efforts to “bracket” prior knowledge of the theoretical and empirical literature (Elliott, Fischer, & Rennie, 1999; Malterud, 2001). Further, in the later stages of the coding process, categories and memos were reviewed and discussed among the first and second author of study I and II. Questions raised were taken back to data in the original transcripts and differences in opinion between the researchers were discussed until agreement was reached. Also, in study II, the preliminary results were

25

presented to four of the respondents in the study (i.e., two male and two female therapists) who were asked to reflect if anything seemed odd or missing. Field notes from this 1.5-hour long meeting were then integrated into the coding process.

Figure 1. An illustration of the coding process in study II.

A specialized computer program for qualitative data analysis, known as ATALS.ti

(2000), was used in the coding process of studies I and II. ATLAS.ti retains the links between transcripts, codes, categories, and memos throughout the analysis, making it possible to move back and forth between coding, elaborating the categories, writing memos, and building the conceptual model. The networking function of ATLAS.ti was used to sort codes and to visually connect categories into diagrams that outlined their relations.

Self-report measures Symptom Checklist – 90 (SCL-90; Derogatis, 1994; Derogatis, Lipman, & Covi, 1973) is a self-report questionnaire that consists of 90 items referring to symptoms experienced over the last 7 days. The items are rated on 5-point Likert scales ranging from 0 (not at all) to 4 (very much). The Global Severity Index (GSI) of the SCL-90 was used as an indicator of total symptom load. The Swedish version of the measure has demonstrated adequate reliability and validity (Fridell, Zvonomir, Johansson, & Malling Thorsen, 2002). In study IV, excellent internal consistency (α > .90) was observed across all measurement points.

The Inventory of Interpersonal Problems (IIP; Horowitz, Rosenberg, Baer, Ureno, & Villasenor, 1988; Horowitz, Alden, Wiggins, & Pincus, 2002) is a 64-item self-report measure that assesses interpersonal problems in eight circumplex subscales. Each item is rated on 5-

26

point Likert scales ranging from 0 (not at all) to 4 (very much) and the sum of all items constitutes a measure of overall interpersonal distress. The Swedish translation of IIP has shown adequate psychometric properties (Weinryb et al., 1996). The internal consistency for the total score ranged between .93–.95 across the measurement points in study IV.

The Structural Analysis of Social Behavior Intrex Questionnaire (SASB; Benjamin, 2000) is a self-report measure aimed at assessing the patients’ internalized self-concept. The instrument consists of 36 items rated on 11-point Likert scales ranging from 0 (not true) to 100 (true). The test–retest reliability for the SASB has been shown to be high (r = .87) for both the American (Benjamin, 1984; Benjamin, Rothweiler, & Critchfield, 2006) and the Swedish version (Armelius, 2001). The items may also be grouped into eight clusters: Self-Autonomy, Self-Affirm, Self-Love, Self-Protect, Self-Control, Self-Ignore, Self-Hate, and Self-Blame with four or five items in each cluster. In study III, we followed Adamson and Lyxell (1996) and combined clusters 2, 3, and 4 to indicate a positive self-concept, while the combination of clusters 6, 7, and 8 indicated a negative self-concept. The internal consistency proved to be good for both the combined positive (α = .86) and negative (α = .85) clusters in our sample.

The Helping Alliance Questionnaire (HAq-II; Luborsky et al., 1996) is a well-established measure of the therapeutic alliance based on Luborsky’s (1976) and Bordin’s (1979) formulations of the alliance. The measure consists of 19 items that are rated on 6-point Likert scales ranging from 1 (I strongly feel it is not true) to 6 (I strongly feel it is true). Both patients and therapists rated the alliance every third month during treatment in YAPP. The HAq-II has demonstrated excellent reliability and good convergent validity with other measures of the alliance (Luborsky et al., 1996). Based on data from YAPP, the Swedish version of the HAq-II has been shown to have excellent internal consistency (Cronbach’s α = .91) for the patient questionnaire and good (α = .88) for the therapist questionnaire (Lindgren et al., 2010). Expert-rated measures The Global Assessment of Functioning Scale (GAF; American Psychiatric Association, 2000) is an expert-rated measure of overall symptomatic and social functioning. The scale ranges from 1 to 100 with 10-point intervals describing discrete levels of functioning. The ratings in YAPP were based on the interview material obtained with the PTI. A group of trained raters did all assessments and, to increase reliability, ratings were obtained through group discussion until consensus was reached. The therapists were not involved in the rating of their own patients.

The Differentiation-Relatedness Scale (DRS; Blatt & Auerbach, 2001; Diamond, Blatt, Stayner, & Kaslow, 1995) is a 10-point expert-rating scale used to evaluate the degree of complexity and differentiation in cognitive-affective schemas of self and significant others. The ratings are based on information obtained with the Object Relation Inventory (ORI; Auerbach & Blatt, 1996) and higher ratings indicate more mature and integrated mental representations. In the present studies, each patient’s representations of mother, father, him/herself, and the therapist were assessed by a group of trained raters. Inter-rater reliability was found to be in the good range (ICC = .71) for a subsample of patients in YAPP (Hjälmdahl, Claesson, Werbart, & Levander, 2001).

27

The Patient Attachment to Therapist Rating Scale (PAT-RS; Lilliengren, 2011; Lilliengren et al., 2014) is an observer-rating scale that aims at assessing the quality of patient-therapist attachment from patients’ description of the therapist as a person, their experience of the therapeutic process, and their reactions to attachment related issues in the therapeutic relationship (closeness, separation, etc.). PAT-RS was developed as a part of this thesis (see below) and the instrument includes four subscales: Security, Hyperactivation, Deactivation, and Disorganization. In study III and IV, PAT-RS was applied to interview material obtained with both PTI and ORI at termination of therapy. Three raters rated the material; the main developer of PAT-RS and two master-level students who received training in PAT-RS. Twelve of the 70 interviews were rated with consensus discussion among the raters during training and calibration and 37 interviews were rated independently for the assessment of inter-rater reliability (average scores was used in the final variable). A single rater (author and main developer of PAT-RS) rated the remaining 21 interviews. During the rating procedure, all raters were kept blind with regards to outcome measures in YAPP. The psychometric properties of the PAT-RS subscales were investigated in study III and are reported in the results section below. Statistical analyses Study III. We used standard statistical procedures to examine the psychometric properties of the PAT-RS subscales. Internal consistency was estimated with Cronbach’s α and inter-rater reliability with the intraclass correlation, two-way random effects model with absolute agreement (ICC [2, 1]; Shrout & Fleiss, 1979). In this context, the ICC estimate reflects rater similarity. According to criteria proposed by Cicchetti (1994), an ICC below .40 may be considered poor; between .40–.59 fair; between .60–.74 good, and above .75 excellent. The construct validity of the PAT-RS subscales was explored using bivariate Pearson r correlations with other measures. Missing data were handled with pairwise deletion. Finally, an exploratory factor analysis was performed to examine the underlying factor structure of all subscale ratings. Since we expected that data would not meet all criteria for normality we used the principal axis factoring extraction method (Fabrigar, Wegener, MacCallum, & Strahan, 1999). We also expected factors to be correlated and applied oblique rotation (direct oblinim with Keiser normalization) as suggested by Costello and Osborne (2005).

Study IV. Linear mixed-effects models procedures (Heck, Thomas, & Tabata, 2014) were used to investigate the relationships between secure attachment to therapist, patient-rated alliance and outcome. Because a considerable amount of the data was nested within therapists, we first examined the possible presence of therapist effects using the intraclass correlation estimate (ICC; see Wampold & Serlin, 2000). In this context, the ICC estimate may be interpreted as the percentage of the total variability in a given variable that may be attributable to between-therapists differences. No significant between-therapist variability was observed in either our independent variables or in terms of outcome. Still, since even small amounts of therapist variability may lead to biased estimates (Crits-Christoph & Mintz, 1991; Wampold & Serlin, 2000), we decided to include a random intercept at the therapist level in all analyses in order to provide a more conservative test of our final regression models.

The data was analyzed separately for two time periods: intake to termination and termination to 1.5 years follow-up. Because we were only able to assess secure attachment to

28

therapist at termination, the results for the first period are referred to as “associations”, whereas the results for the second period are regarded as “predictions”. We first modeled change between intake and termination using outcome at termination as dependent variable while controlling for score at intake, constituting our baseline model. To address the specific hypotheses in study IV, Security was then entered together with length of therapy as a covariate since, arguably, longer therapy may both lead to better outcomes and a more secure attachment to the therapist (Model 1). Next, the alliance was entered together with the other variables (Model 2) and, lastly, the procedures were repeated for the termination to follow-up time period.

Since we did not aim to examine the fit between different models, restricted maximum likelihood (REML) estimation was applied and since only one parameter was included at level 2 (i.e., the random intercept), a scaled identity matrix covariance structure (variance components) was assumed. In order to obtain more interpretable outputs we calculated the standardized β-coefficients for each independent variable in our final models, using the standard formula: β = B (SDx / SDy).

All statistical calculations in study III and IV were performed with the SPSS (v. 19) software package. Before the analyses, data was inspected following the recommendations of Tabachnick and Fidell (2013). A univariate outlier in IIP total score at termination was corrected to the next score within three standard deviations from the mean. Significantly skewed variables were either square root or log10 transformed. Primary significance level was set to p < .05 and, due to the primarily exploratory aims of the studies (Bender & Lange, 2001), all tests were performed without any correction for family-wise error rate (e.g., Bonferroni).

Development of the Patient Attachment to Therapist Rating Scale (PAT-RS)

A new rating scale for patient-therapist attachment, called the Patient Attachment to Therapist Rating Scale (PAT-RS), was developed as part of this thesis. The background for this effort was the results of study I and II, which directed our attention towards the possible influence of patient-therapist attachment quality on psychotherapy process and outcome. Given the rich interview material available in YAPP, the idea of using a rating scale to assess attachment to therapist from the patients’ narratives emerged. However, we could not find any available measure in the literature that fitted our needs and, consequently, we decided to construct our own instrument. The basic assumptions, theoretical structure, subscales, item construction and rating procedure of PAT-RS are described in detail below. This section may be regarded as a general introductory manual to the use of PAT-RS. The current English version of the rating sheet is provided in the Appendix of this thesis.

Basic assumptions In line with an attachment-based view of the therapeutic relationship (Bowlby, 1988; Farber & Metzger, 2009, Obegi, 2008; Mallinckrodt, 2010; Shaver & Mikulincer, 2009), PAT-RS is based on the assumption that engaging in a helping relationship with a therapist will

29

unavoidably activate the patient’s attachment system. This, in turn, will bring the patient’s internal working models and attachment strategies to the fore. The interaction with the therapist will then determine whether the relationship they form will have the basic qualities of a secure attachment relationship or if insecure strategies will dominate. Accordingly, PAT-RS is constructed as a relationship specific measure (Mikulincer & Shaver, 2007) and aims at assessing the quality of the relationship with the therapist rather than the patient’s “attachment style”. Of course, the patient’s global attachment orientation will likely influence the interaction with the therapist, but the particular quality of their attachment is regarded as the result of a complex interaction between the patient’s and the therapist’s personal characteristics (e.g., their respective attachment styles, as well as other factors), specific therapeutic technique (e.g., how the therapists responds to the patient’s attachment needs as well as the timing and focus of specific interventions) and contextual factors (e.g., how the relationship is structured over time in terms of frequency of meetings, etc.).

A second assumption underlying PAT-RS is that it is possible to assess the attachment quality of the therapeutic relationship from narratives of how patients view the therapist as a person, how they experience the therapeutic process, as well as their reactions to attachment related issues in the therapeutic relationship (i.e., closeness, separation, etc.). Since attachment involves unconscious processes and strategies, it is assumed that the quality of the attachment will be indicated by how the patient’s (explicitly or implicitly) describes the function of the therapeutic relationship. In contrast to attachment measures that assess the discourse quality in the interview (such as the AAI), the focus of PAT-RS is primarily on the content of the patient’s narratives and descriptions of how he or she experiences the therapist and uses the therapeutic relationship for exploration of anxiety-provoking inner experiences.

Lastly, PAT-RS is designed as a dimensional measure. Thus, it is assumed that it is both possible and meaningful to assess attachment quality in terms of levels on particular subscales (see below). These levels may vary between particular patient-therapist dyads, as well as within the same dyad at different assessment points. Consequently, the PAT-RS subscales are assumed to provide a nuanced picture of the quality of patient-therapist attachment at a particular point in time.

Theoretical structure and subscales PAT-RS is theoretically grounded in the two-dimensional model of adult attachment (Bartholomew, 1990; Bartholomew & Horowitz, 1991; Hazan & Shaver, 1987). This well-established model relates adult attachment-system function and interpersonal dynamics to the underlying dimensions of anxiety and avoidance (Mikulincer & Shaver, 2007). The anxiety dimension refers to the level of preoccupation with attachment-related issues such as separation and autonomy. High values in this dimension indicate the presence of hyperactivating attachment strategies, which are typically associated with high emotionality, strong reactions to separation, fear of abandonment, overdependence, and worry about not being accepted or loved. The avoidance dimension reflects strivings for independence, autonomy, and self-regulation and high values in this dimension indicates the presence of deactivating attachment strategies to avoid closeness, intimacy, vulnerability, dependence, and emotional openness in relationships. Following the two-dimensional model, the PAT- RS is designed to assess the quadrants of attachment relatedness in four subscales: Security, Hyperactivation, Deactivation, and Disorganization (see Figure 2).

30

Figure 2. The subscales of PAT-RS represented in the two-dimensional model of adult attachment. Subscale 1: Security. This subscale reflects the patient’s level of secure attachment to the therapist. High scores indicate low degree of anxiety and low degree of avoidance. This is theoretically desirable since a sufficient sense of security and trust is necessary for activation of the exploratory system, enabling the patient to be emotionally open and approach potentially threatening and painful material in the therapy process. The Security subscale also reflects to what extent there is a balance between autonomy and dependence in the relationship. Optimally, the patient can experience a certain degree of dependence without losing agency or the ability for mature self-regulation when appropriate. The therapeutic relationship thus functions as “a secure base,” enabling the patient to explore and make changes in his or her life outside sessions.

Subscale 2: Hyperactivation. This subscale reflects to what extent the patient relates to the therapist using hyperactivating strategies. High scores indicate a low degree of avoidance but a high degree of attachment anxiety. The patient might typically be preoccupied with worry about abandonment and lack of understanding and care from the therapist, which leads to increased emotional arousal to ensure attention and support. Theoretically, this is a suboptimal way of relating in therapy since high attachment anxiety inhibits the exploratory system. Although the patient might be very open with emotional material, there is typically a lack of appropriate cognitive elaboration and reflection (mentalization) necessary for integration and change in internal working models. Further, the patient will tend to depend too strongly on the therapist, leading to decreased agency and difficulties separating from therapy. Typically, the patient will fear making changes on his/her own outside therapy and might long for the perceived safety of the therapeutic relationship.

Subscale 3: Deactivation. This subscale reflects the patient’s use of deactivating strategies in relation to the therapist. High scores indicate low anxiety but a high degree of avoidance,

LOW AVOIDANCE

HIGH AVOIDANCE

LOW ANXIETY

HIGH ANXIETY

Subscale 3: Deactivation

Subscale 4: Disorganization

Subscale 2: Hyperactivation

Subscale 1: Security

31

leading the patient to keep an emotional distance in the therapeutic relationship. Patients yielding a high score on this subscale will typically value their independence and autonomy and have difficulties experiencing dependence and vulnerability and with receiving care. This will also lead to suboptimal relating in therapy since the avoidance of intimacy and openness limits the therapist’s ability to both understand and impact the patient on an emotional level. The avoidance of emotional arousal might also lead to an intellectualized therapeutic process, which does not have the same impact on the inner working models. The patient may be active outside therapy, but typically there is a lack of emotional connection between the therapeutic process and the patient’s activities.

Subscale 4: Disorganization. The last subscale reflects the patient’s level of fearfulness and disorganization in relation to the therapist. High scores indicate both a high degree of anxiety and a high degree of avoidance. Theoretically, this indicates that the patient is trapped in a state of “fright without a solution” (Hesse & Main, 2000) with the therapist. Fear evoked by the therapeutic process leads the patient to seek proximity to the therapist, but at the same time the patient experiences an increasing need for distance because of additional fear that is evoked by the closeness itself. Thus, the patient does not “succeed” in developing a coherent way of relating but rather oscillates between hyperactivation and deactivation and/or develops idiosyncratic, maladaptive ways of relating in therapy. Typically, the patient will have grave difficulties trusting the therapist although he or she might at the same time describe a strong need for closeness. The patient might be very reluctant to open up, fearing the therapist’s or his/her own reaction, or both. This might lead to “a passive neediness” or the patient might be openly hostile and demanding in an effort “to control” the therapist. The experience of the therapeutic relationship will typically be described in contradictory statements and might also involve “unusual” or “odd” ideas and perceptions. When it comes to the balance between autonomy and dependence the patient typically oscillates between extremes (does not need the therapist at all—cannot do anything without his/her therapist). This subscale is theoretically the opposite of attachment security and is therefore hypothetically associated with most difficulties, both within the therapeutic setting and outside therapy. Item construction To assess the subscales, PAT-RS is organized around nine components that have been suggested to define an attachment relationship (Parish & Eagle, 2003a, 2003b): 1. Secure base, 2. Stronger/wiser, 3. Safe haven, 4. Proximity seeking, 5. Particularity, 6. Responsiveness, 7. Strong feelings, 8. Separation anxiety and, 9. Mental representation (see Table 2).

For each of these components, the patient is expected to experience the therapeutic relationship and relate to the therapist in theoretically consistent ways, corresponding with the four quadrants in the two-dimensional model. Adopting a “prototype-matching” approach to attachment measurement (Bartholomew & Horowitz, 1991; Maunder & Hunter, 2012; Pilkonis, 1988; Strauss, Lobo-Drost, & Pilkonis, 1999), we constructed four prototypical descriptions (one for each subscale/quadrant) for each of the nine components listed above. These short descriptions may be considered “ideal examples” of what patients may typically report for each component (see the full rating sheet in the Appendix). The prototypical descriptions were derived from items of previously constructed self-report

32

measures (CATS; Mallinckrodt et al., 1995; CAQ-T; Parish & Eagle, 2003a, 2003b), as well as relationship-specific attachment coding schemes (PT-AAI; Diamond et al., 2003; AAI Q-sort; Koback, Cole, Ferenz-Gillies, Fleming, & Gamble, 1993). Further, the clinical attachment literature (Bowlby, 1988; Fosha, 2000; Holmes, 2001, 2010; Muller, 2010; Pearl, 2008; Wallin, 2007) was searched for typical examples of attachment-related responses in the therapeutic relationship. Before including the prototypical descriptions in the final rating sheet they were discussed with one clinical and one non-clinical attachment researcher.

Table 2. Components of an Attachment Relationship

Component Description

1. Secure base The attachment figure provides a sense of security and confidence in exploring new situations.

2. Stronger/wiser The attachment figure is someone who seems capable of providing protection and guidance.

3. Safe haven The attachment figure is someone to turn to for comfort, support and reassurance, especially in times of distress.

4. Proximity seeking The tendency of individuals to seek an optimal level of proximity and closeness with their attachment figure.

5. Particularity The attachment figure is a particular, unique other person, who cannot be easily replaced by another.

6. Responsiveness The attachment figure is perceived as optimally responsive in relation to the individual’s emotional needs.

7. Strong feelings The attachment figure is associated with positive (i.e. joy, love), as well as negative (i.e. anger, sadness), emotions during the formation, maintenance, disruption and renewal of the attachment bond.

8. Separation anxiety Separation from the attachment figure is associated with distress, which might evoke attempts to resist being separated.

9. Mental representation In adults, an internalized representation of the attachment figure can be evoked for comfort or guidance in times of distress.

Note. Adapted from “A New Measure of Components of Attachment” by M. Parish and M. N. Eagle, 2003, unpublished manuscript, Derner Institute of Advanced Psychological Studies, Adelphi University.

Rating procedure and scoring PAT-RS is primarily constructed for the rating of transcribed interviews. The rater is instructed to read the interview material as a whole and mark all paragraphs and sentences that contain information about how the patient perceived the therapist and experienced the therapeutic relationship. Since PAT-RS is a relationship-specific measure, the instruction is to disregard descriptions of relationships other than the one with the therapist. Further, it is the patient’s relationship with the therapist in the present that is to be rated. Descriptions of change in the therapeutic relationship are noted, but the rater should focus on how the patient experiences and relates to the therapist at the time of the interview.

After reading the whole interview transcript, the rater reads all four prototypical descriptions for the first component in PAT-RS (see Appendix) and rates each description on a 5-point Likert scale ranging from 1 (does not fit at all) to 5 (fits very well). The central idea with the prototypical descriptions is to direct the rater’s attention toward what is typically expected, but the rater is allowed to consider other information in the interview when it seem

33

relevant for the specific component and the underlying dimension being assessed. The procedure is then repeated for the next component. If in doubt regarding a particular component the rater is instructed to review marked paragraphs and try rating again. If there is still insufficient information for rating a particular component the rater indicates this and moves to the next component.

A mean index score for each subscale is calculated by dividing the total number of components rated for that particular interview transcript by the sum of ratings in each subscale. Consequently, for each subscale the final subject score ranges from 1.00 to 5.00.

35

RESULTS

Study I: A model of therapeutic action grounded in the patients’ view of curative and hindering factors in

psychoanalytic psychotherapy The aim of study I was to explore patients’ views of therapeutic action in psychoanalytic psychotherapy. Some of the specific research questions were: what do patients perceive as curative in treatment? What hinders change, according to patients’ views? How are these factors interrelated? PTI-interviews conducted with 22 patients who had recently terminated individual psychotherapy in YAPP were analyzed using grounded theory methodology.

The analysis resulted in a conceptual model consisting of 9 main categories with 16 subthemes (Figure 3). The first curative factor that emerged was Talking about oneself (1), which involved talking in two distinct modes: (a) expressing, reflecting, and labeling one’s own thoughts and feelings, and (b) remembering and revising one’s personal history. However, at the same time as talking about oneself was described as curative, the patients also expressed that talking about themselves in therapy could be anxiety provoking and energy consuming. Hence, Talking is difficult (2) emerged as a hindering factor, possibly obstructing the process of talking about oneself as a curative factor.

The second curative factor emerging was Having a special place and a special kind of relationship (3), which indicated that the patients viewed the therapy setting and the relationship with the therapist as a unique, curative experience in itself. This experience involved (a) an atmosphere of acceptance, respect and support, (b) the therapist being an “outside person” with no relation to the patient’s family or friends and, (c) having time and continuity in the treatment. Talking about oneself in the context of this special kind of relationship was associated with the therapeutic impact of New relational experiences (4). The narratives indicated that when patients were able to overcome the difficulties associated with talking, this was in itself experienced as an important new relational experience. Further, the feeling of safety in therapy provided them with a platform for trying out new ways of being and relating to other people, thus increasing the probability of new relational experiences between sessions.

The third curative factor that emerged was Exploring together (5), indicating that patients valued the mutual collaboration with the therapists’ and that this collaboration had a distinct “exploring” quality to it. Several subthemes emerged, including (a) analyzing causes to one’s problems and finding connections to one’s personal history, (b) discovering and challenging self-defeating thoughts and negative interpretations of the self and the world, (c) focusing on what the patient wanted to get out of life and, (d) defining problems and setting goals for the treatment which lead to a clearer picture of the difficulties and ways of coping

36

with them. The exploratory work in therapy was further connected to an experience of Expanding self-awareness (6). This included (a) having “sorted out” thoughts, feelings and experiences and memories of the past as well as (b) having discovered patterns in one’s personality and way of relating to self and others. Further, the patients described that the joint exploratory activity with the therapist helped them (c) gain new perspectives and (d) learn new ways of thinking, which could be implemented in their ordinary life and lead to reductions in distress.

Figure 3. A tentative theoretical model of therapeutic action grounded in the patients’ view of curative and hindering factors in psychoanalytic psychotherapy with young adults. Curative factors are indicated by solid line rectangles, hindering aspects by dashed-line rectangles, therapeutic impacts by dotted ellipses, and negative impacts by dashed-line stars.

However, expanding self-awareness was also connected with a negative impact,

labeled Self-knowledge is not always enough (7). This category indicated that some patients experienced expanding self-awareness as insufficient for change. This was further associated with disappointment in the treatment and connected to the hindering factor Something was missing (8). Typically, disappointed patients experienced the therapist as too passive and wanted more feedback, guidance, and advice. Some also expressed missing more structured and “action-oriented” interventions (such as homework assignments) linking therapy to the problems they faced in ordinary life.

Lastly, the experience that something was missing was further associated with the negative impact of Experiencing mismatch (9). Several disappointed patients expressed concerns that the therapeutic modality might not have been “right” for their problems or

37

wondered how they “fitted together” with their particular therapist. Experiencing mismatch had a negative influence on the experience of a special relationship with the therapist, possibly undermining the curative factors in the model. There was no connection between exploring together and the “something was missing” or “experiencing mismatch” categories, possibly indicating that such experiences were not often addressed or resolved in treatment.

Study II: Therapists’ view of therapeutic action in psychoanalytic psychotherapy with young adults

The aim of study II was to explore experienced therapists’ view of therapeutic action in psychoanalytic psychotherapy. Specific research questions were: What in treatment contributed to change and what hindered change, according to the therapists’ view? What kind of changes do the therapists perceive in their patients? How are these factors interrelated from the therapists’ point of view? PTI-interviews with 16 therapists (who treated the 22 patients included in study I) were analyzed using grounded theory methodology.

The analysis resulted in a conceptual model consisting of one core category with three subcategories, seven main categories, and seven linking concepts (Figure 4).

Figure 4. A tentative theoretical model of therapeutic action grounded in the therapists’ view of curative and hindering factors in psychoanalytic psychotherapy with young adults. Curative factors are indicated by solid line rectangles with rounded corners, the hindering factor by a dashed-line rectangle, positive outcome categories by ellipses, and negative outcome by a dashed-line star. The linking concepts are placed in italics directly on the lines between categories. Dashed lines indicate a negative influence between categories and direct lines indicate a positive influence.

38

The core curative factor in treatment from the therapists view was Developing a close, safe and trusting relationship (1). This involved the patient gradually becoming more attached to the therapist and, consequently, being able to talk about difficult and painful subjects, as well as express feelings more openly. When patients opened up this provided an opportunity to revise inner conceptions of self and others, either in direct dialogue with the therapist or implicitly through the interaction between patient and therapist. From the therapists’ view, it was central that the patients’ negative expectations were contradicted in the relationship.

Further, the therapists regarded the development of a trusting relationship to be dependent on their adopting and maintaining a stance of genuine interest, acceptance, flexibility and confidence in the therapy process. Further, the therapists also stressed the importance of having time and continuity, as well as the patients’ resources and commitment to the therapy process, sometimes viewed as conflicting with their “intrinsic developmental force” as young adults.

Other curative factors that emerged were the Patient making positive experiences outside the therapy setting (2) and the therapist Challenging and developing the patient’s thinking about the self (3). Here, the therapeutic process was described as a joint activity of “thinking together”, resulting in the patient Becoming a subject (4) and acquiring an Increasing capacity to think and process problems (5).

The sole hindering factor that emerged in the model was The patient’s fear about close relationships (6). The therapists described several pathways through which this fear may hinder the therapeutic process, including the patient keeping an emotional distance in sessions or reducing the intensity or length of treatment, as well as negative countertransference evoked in the therapist leading to difficulties maintaining a therapeutic stance. The result of such processes was typically viewed as hindering the development of a close, trusting relationship, leading to Core problems remaining (7). Finally, the therapists also tended to view therapeutic work as something that the patient continued after therapy had ended, indicated by the category The therapeutic process continues after termination (8).

Study III: Patient Attachment to Therapist Rating Scale: Development and psychometric properties

The aim of study III was to describe the development and theoretical structure of the Patient Attachment to Therapist Rating Scale (PAT-RS), as well as investigate its psychometric properties. Three raters (main author and two master level students trained by the main author) applied PAT-RS to interview material obtained from 70 patients who had terminated their individual therapy in YAPP.

Reliability was estimated on 37 of the interviews that were rated independently by the raters (Table 3). Strong internal consistency (α >.90) was found for all four subscales indicating that the ratings of the prototypical descriptions followed each other closely in each subscale. The inter-rater reliability was close to excellent for Security (ICC = .74) and Disorganization (ICC = .74), as well as in the good range for Deactivation (ICC = .62). However, the inter-rater reliability for the Hyperactivation subscale was poor (ICC = .34).

39

Table 3. Reliability of the PAT-RS Subscales (n = 37)

PAT-RS Subscale

Security Hyperactivation Deactivation Disorganization

Internal consistency Rater 1 Rater 2 Rater 3

.98 .99 .97

.97 .92 .94

.97 .97 .95

.99 .97 .98

Inter-rater reliability Single measures

95% CI Average measures

95% CI

.74 [.60, .85] .90 [.82, .94]

.34 [.14, .55] .61 [.33, .78]

.62 [.44, .78] .83 [.70, .91]

.74 [.60, .85] .90 [.82, .94]

Note. PAT-RS = Patient Attachment to Therapist Rating Scale; CI = confidence interval. Means, standard deviations and correlations are presented in Table 4. As expected,

Security was negatively correlated with all three insecure subscales. In line with the underlying theoretical model, Disorganization was positively correlated with both Hyperactivation and Deactivation and the association between the Hyperactivation and Deactivation was weak. Further, Security correlated strongly and negative (r = –.92) with an insecurity index based on the average score of the insecure subscales, suggesting that Security captures a general secure-insecure dimension of patients’ attachment to their therapist.

Table 4. Means, standard deviations and correlations

PAT-RS Subscale

n M SD Security Hyperactivation Deactivation Disorganization

PAT-RS Subscale Security 70 3.3 1.1 - Hyperactivation 70 1.9 0.8 –.43** - Deactivation 70 1.9 0.9 –.76** –.15 - Disorganization 70 1.4 0.8 –.63** .35** .55** - Insecurity indexa 70 1.7 0.6 –.92** .56** .70** .82**

Alliance Patient rated 61 5.0 0.5 .47** –.43** –.26* –.33* Therapist rated 66 4.7 0.4 .40** –.22* –.32** –.27*

Mental representations Mother 70 7.2 1.0 .25* –.10 –.18 –.19 Father 70 7.4 0.8 .28* –.07 –.13 .00 Self 70 7.2 1.1 .45** –.21* –.21* –.20* Therapist 68 7.8 1.4 .44** –.15 –.34** –.11

Self-concept Positive 65 52.4 18.2 .28* –.27* –.12 –.24* Negative 65 24.3 19.0 –.21* .31* .00 .21*

Interpersonal problems 65 1.2 0.6 –.13 .27* .04 .28* Symptom severity 65 0.9 0.7 –.31** .55** .04 .35** Global functioning 69 68.5 11.5 .46** –.35** –.29** –.52** Length of therapy 70 23.0 13.0 .19 –.08 –.05 .19

Note. PAT-RS = Patient Attachment to Therapist Rating Scale aThe insecurity index represents the average score of all three insecure subscales *p < .05. **p < .01.

40

The pattern of correlations with other measures supports construct validity of the Security subscale. As expected, there was a positive association between Security and the therapeutic alliance, more mature mental representations, a more positive and less negative self-concept, less symptoms and higher global functioning. The correlations are of moderate strength, suggesting that the subscale also captures something different from other measures. Contrary to expectations, however, Security was not significantly associated with lower levels of interpersonal problems, although the correlation found is in the predicted direction.

Overall, the pattern of correlations also supports the construct validity of the Deactivation subscale, which was moderately associated with lower alliance and less complex mental representations of the self and the therapist. The pattern of correlations with self-report measures indicates that the Deactivation subscale taps the use of avoidant strategies in comparison with the other insecure subscales.

However, the pattern of correlations for the Hyperactivation and Disorganization subscales were very similar, making it difficult to interpret their respective validity beyond the general conclusion that both subscales likely tap insecure attachment to therapist. Also, the Hyperactivation subscale had poor reliability, hence; any statements regarding its validity would be pre-mature. Notably, none of the subscales were significantly related to therapy length.

The exploratory factor analysis yielded a three-factor solution, explaining 82% of the variance. Security and Deactivation merged to form opposite poles in Factor 1. Factor 2 and 3 had high loadings in Hyperactivation and Disorganization, respectively, suggesting that the subscales capture different aspects although the correlations with other measures came out very similar.

Study IV: Secure attachment to therapist, alliance, and outcome in psychoanalytic psychotherapy with

young adults The aim of study IV was to explore the relationships between secure attachment to therapist (as measured with the Security subscale of PAT-RS), patient-rated alliance and outcome in a sample of 70 young adult patients treated with psychoanalytic psychotherapy. Specific research questions and hypotheses were: (1) Is secure attachment to therapist at termination associated with improvement? In line with attachment theory, we hypothesized that secure attachment to the therapist at termination would be related to change during therapy. (2) Does secure attachment to therapist relate to outcome once alliance is controlled for? Since secure attachment indicates that the patient uses the relationship for emotional exploration, we hypothesized that it would be more strongly associated with improvement. (3) Does secure attachment to the therapist or the alliance predict any changes after treatment termination? In line with attachment theory, we hypothesized that secure attachment to therapist would be more closely linked with sustained or even increased improvement post-treatment compared with the alliance. The main research questions were addressed in a series of linear mixed-effects models, controlling for length of therapy and between-therapist variability.

41

Table 5. Linear mixed-effects models for associations with outcome at termination

Note: GSI = Global Severity Index; GAF = Global Assessment of Functioning; IIP = Inventory of Interpersonal Problems; CI = confidence interval; ICC = Intraclass correlation for therapist effect. Unstandardized estimates are based on transformed variables. *p < .05. **p < .01. Table 6. Linear mixed-effects models for predictors of outcome at follow-up

Note: GSI = Global Severity Index; GAF = Global Assessment of Functioning; IIP = Inventory of Interpersonal Problems; CI = confidence interval; ICC = Intraclass correlation for therapist effect. Unstandardized estimates are based on transformed variables. *p < .05. **p < .01.

The results for associations between secure attachment to therapist and outcome are

displayed in Table 5. In support of our first hypothesis, Model 1 indicated that secure attachment to therapist was associated with pre-post improvement in symptoms (GSI, β = –.29, p = .007), global functioning (GAF, β = .45, p < .001) and interpersonal problems (IIP, β = –.22, p = .024). After controlling for the alliance in Model 2, the relationships were maintained in terms of symptoms (GSI, β = –.28, p = .033) and global functioning (GAF, β =

42

.45, p = .002), but the association with reduction in IIP was no longer significant (β = –.17, p = .160). The alliance was not significantly related to change in any outcome, supporting our second hypothesis. Of note, length of therapy proved unrelated to pre-post treatment change in all outcome variables.

The results for predictions of changes after treatment termination are presented in Table 6. Unexpectedly, we found that length of therapy predicted deterioration in terms of symptoms (GSI, β = .26, p = .002) as well as interpersonal problems (IIP, β = .21, p = .015) during follow-up. Security was not directly associated with changes after termination in Model 1. However, when entered together in Model 2, both secure attachment to therapist and the alliance were significantly related to change in global functioning, but in opposite directions (Security, β = .32, p = .030; Alliance, β = –.28, p = .035). That is, a more secure attachment to the therapist at termination predicted increase in functioning during follow-up, whereas a stronger alliance predicted decrease when both predictors were accounted for. No indication of significant between-therapist variability was observed in any of the models.

43

DISCUSSION

Reflections on main findings Patients’ and therapists’ views of curative factors The models that emerged in studies I and II indicate that therapeutic action in psychoanalytic psychotherapy involves a complex set of interacting factors from both patients’ and therapists’ perspectives. The model based on the therapists’ views is somewhat more elaborated, likely due to the therapists’ training and experience in reflecting on these issues. Looking at the models together, there is an interesting overlap in terms of the curative factors that emerged. From the patients’ view, therapeutic action basically involved talking openly about themselves and exploring personal experiences in the context of a secure therapeutic relationship. Mirroring this, the therapists stressed the development of a close, safe and trusting relationship, which enabled them to “think together” with the patient. Thus, on a general level, both patients and therapists had similar views of the most fundamental curative factors, which may also be considered common factors in any exploratory model of psychotherapy (Lambert, 2013b; Wachtel & Messer, 1997).

In terms of more specific psychoanalytic mechanisms of change, both models include therapeutic impacts that correspond quite well with the notion of corrective emotional experiences (Alexander & French, 1946; Castonguay & Hill, 2012). For patients, experiencing the relationship with the therapist as “special” helped them overcome the difficulties associated with opening up. This was described as a new relational experience, which also encouraged them to try out new ways of relating to significant others. The therapists explicitly stressed the therapeutic impact of patients’ negative expectations being contradicted in the therapeutic relationship, but they also emphasized the importance of patients’ approaching their core problems and making corrective experiences outside the therapy setting. Although psychoanalytic psychotherapy is often focused on within-session interactions, Alexander and French (1946) did not restrict the term corrective emotional experiences to the therapeutic relationship: “…the actual experience of a new solution in the transference situation or in his everyday life gives the patient a conviction that change is possible and induces him to give up the old neurotic patterns” (p. 338, italics added). Hence, both patients’ and therapists’ views of corrective experiences seem to be in line with the original definition of this term.

It is also notable that both patients and therapists typically describe corrective emotional experiences as involving shifts toward more emotional openness and trust in relationships. This may be interpreted as reflecting change towards greater attachment security, particularly in the context of the therapeutic relationship, and seem consistent with the broaden-and-build cycle of security development (Shaver & Mikulincer, 2009). Interestingly, however, both models also involve the therapist helping the patient move

44

toward greater self-definition, self-acceptance and self-care, as well as a sense of personal agency and responsibility. Hence, the formation of a secure attachment to the therapist seems to function as a vehicle for helping the patient develop more mature forms of both relatedness and self-definition (Blatt, 2008; Blatt et al., 2008; Luyten & Blatt, 2013; Pearl, 2008).

Additionally, both patients’ and therapists’ views of therapeutic action involve the acquisition of “insight”. In line with a classic psychoanalytic view of this concept, some patients explicitly described having discovered patterns in their way of relating that were linked to their past and that this prompted them to relate to others in new ways. However, insight emerged as part of broader categories in both models, involving the development of several mental capacities. These included the ability to see things from different perspectives, as well as the capacity to contain and process problems rather than act out in destructive ways, perhaps indicating internalization of the therapist’s ability to contain and reflect.

From a contemporary attachment-informed psychoanalytic perspective, one possible interpretation is that these broad gains in “mental capacities” reflect improved mentalization (Allen & Fonagy, 2006; Allen et al., 2008). In line with this view, it is interesting that the therapists rarely mentioned using specific “insight-promoting” psychoanalytic techniques such as defense or transference interpretations, but rather seemed to focus on helping the patients “think about the self” in a more general sense. Hence, therapeutic action appeared to involve the reflective process in itself rather than the specific mental content resulting from that process (i.e., particular “deep” psychoanalytic insights). Further, it is also notable that both patients and therapists experienced the therapeutic impact of insight as a result of their “mutual exploration”, perhaps reflecting that the capacity for mentalization is optimally developed in the context of secure attachment relationships involving joint attention and “moments of meeting” (Fonagy et al., 2002; Stern et al., 1998).

Patients’ and therapists’ views of hindering factors In contrast to curative factors, patients’ and therapists’ views of hindering factors seem more divergent. From the patients’ view, hindering factors included difficulties “opening up” and expressing thoughts and emotions freely in sessions. Further, some patients experienced increased self-knowledge as insufficient for change, which was also related to experiencing that something was missing in treatment. Typically, this included experiencing the therapist as too passive and wanting more advice, direction and guidance, as well as more focus on between-session activities. Disappointed patients experienced a “mismatch” which seemed to undermine the relationship factors that were identified as curative in the model. In essence, the hindering aspects that emerged from the patients’ view appear to represent negative reactions to treatment elements that are quite specific to psychoanalytic psychotherapy (i.e., the invitation to form a close relationship and talk openly about anxiety-provoking inner experiences, exploratory work toward greater self-understanding, the therapist’s relative passivity, non-directiveness and non-intrusiveness). This result mirrors other studies that have indicated that certain patients may experience specific factors in different treatment modalities as obstructing rather than helpful (Gershefski, Arnkoff, Glass, & Elkin, 1996; Nilsson, Svensson, Sandell, & Clinton, 2007; Levy, Glass, Arnkoff, & Gershefski, 1996). Further, patients who reacted negatively typically had their own ideas of what could have been different. However, it seems that that these ideas were seldom or never discussed, nor integrated in the therapeutic process.

45

Turning to the therapists’ view, it was striking that the patients’ fear of closeness emerged as the sole hindering factor. The therapists described several pathways through which this fear could hinder the therapeutic process, including the patient keeping an emotional distance in sessions or reducing the intensity or length of treatment. Some therapists also mentioned that the patient’s distancing and disconnection from the therapeutic work could evoke negative reactions in them (i.e., “countertransference”), which then led to difficulties in maintaining an adequate therapeutic stance. The result of such processes was typically viewed as hindering the development of a close, trusting relationship, leading to a treatment that lacked in emotional depth, became stalled or was terminated prematurely.

The idea that patients’ fear of closeness hinders treatment may be interpreted as reflecting the psychoanalytic notion of “resistance” (Breuer & Freud, 1895/1955; Freud, 1926/1959). Increased patient resistance may signal that important emotional conflicts are being activated in the therapeutic relationship, which then needs the therapist’s careful attention (Frederickson, 2013; Coughlin Della Selva, 1996; Muller, 2010). However, neither the model based on the patients’ or the therapists’ view included a link between the hindering factors and the joint exploratory work, possibly indicating that such experiences were not often addressed, worked-through or resolved in treatment.

Looking at the results of study I and II together, it seems sadly evident that therapists and patients may sometimes get caught up in a “vicious circle” of “blaming the other” for the perceived obstacles in treatment. Perhaps some therapists responded to patients’ negative reactions by holding tighter to the specific factors in their treatment model, which then provoked even more negative reactions from the patient. The lack of connection between hindering factors and the joint exploratory work could also indicate that the therapists were simply unaware of their patients’ dissatisfaction and that disappointed patients preferred to keep such experiences to themselves. In a study that specifically focused on dissatisfied patients’ views of hindering factors in the YAPP sample, von Below and Werbart (2012) identified an underlying theme of abandonment, which reflected these patients’ profound lack of confidence in the relationship with the therapist whom they perceived as lacking in direction, responsiveness and flexibility. Perhaps some patients had trouble discussing negative experiences openly in fear of eliciting negative responses from the therapist, which would only risk increasing their sense of isolation and abandonment.

From an attachment perspective, it is notable that the hindering factors that emerged in both models involved distancing (which was typically attributed to the other) and distrust. This may indicate that insecure attachment strategies (perhaps deactivation in particular) were evoked and maintained in some therapeutic dyads. In line with a cyclical view of attachment pattern maintenance (Wachtel, 2014), one interpretation is that the therapists’ strategies for managing patients’ insecure attachment strategies (perhaps deactivation in particular) sometimes led to their preservation rather than their resolution. It is also possible that negative reactions from patients activated insecure strategies in some therapists, which may have led them to withdraw and become less flexible. Thus, the lack of links between negative experiences and the joint therapeutic work in both models may reflect that insecure attachment strategies inhibited the exploratory systems. One hypothesis is that the therapeutic process in such cases may become disconnected and characterized by “pseudo-mentalizing” (Allen & Fonagy, 2006). The patient and therapist may look like they are engaged in a therapeutic dialogue, but the reflective process lack genuine cognitive-affective integration.

46

This will have limited impact on the patient, perhaps indicated by some patients’ experience that self-knowledge was insufficient for change.

Assessing attachment to therapist with PAT-RS A new assessment instrument for patient-therapist attachment, PAT-RS, was developed as part of this thesis. In contrast to previously constructed self-report measures (i.e., CATS; Mallinckrodt et al., 1995; CAQ-T; Parish & Eagle, 2003a, 2003b), PAT-RS utilizes observer-ratings of patients’ spontaneous descriptions of the therapeutic process, as well as the therapist as a person, typically collected in research interviews. This makes the instrument less susceptible to self-report biases and at the same time more receptive to implicit indicators of attachment quality, provided that independent observers are able to detect them reliably. PAT-RS also differs from the AAI-based procedure proposed by Diamond and colleagues (2003), which evaluates the patient’s “attachment state of mind” in relation to the therapist. Rather than focusing on linguistic aspects, the scoring procedure in PAT-RS involves comparing patient narratives with prototypical descriptions of an attachment relationship, adapted to the therapy context. Consequently, in comparison to an AAI-based procedure, PAT-RS may require less specialized training to use.

PAT-RS is designed as a relationship-specific measure. That is, rather than assessing the patient’s attachment style, the aim is to assess the attachment quality of his or her specific relationship with the therapist. However, one might question if this is really what the instrument captures. Since the assessment is based on only the patient’s descriptions of the therapeutic process, it may be argued that what is really assessed is the patient’s mental representation of the relationship, not the relationship “in itself”. Since we do not have any data on the actual interaction in therapy, we cannot know how that representation relates to what transpired in the relationship. On the other hand, similar to findings for the alliance (Horvath et al., 2011), it may also be argued that the patient’s experience of the relationship is likely what is most important for process and outcome.

The initial examination of the psychometric properties of PAT-RS in study III encourages further development and refinement of the instrument. The Security subscale was rated with satisfactory inter-rater reliability and the pattern of correlations with other measures support its construct validity. The strong negative correlation with the insecurity index (r = −.92) further suggests that the Security subscale captures a general secure-insecure dimension of patients’ attachment to their therapists and, consequently, may be useful for assessing the overall quality of patient-therapist attachment. In line with adult attachment theory, the results of study IV also indicated that Security relate to change during treatment and that its unique variance may predict improvement in functioning during follow-up, which further support for the validity of the subscale.

Regarding the insecure subscales, the results of study III also suggest that the Deactivation subscale may be useful for its purposes. The inter-rater reliability was somewhat low, but still within the “good range”, and associations with other measures indicate that the subscale likely captures deactivating strategies. More caution should be taken regarding the Hyperactivation and Disorganization subscales. Although both subscales likely tap insecure forms of attachment to therapist, the Hyperactivation subscale had poor inter-rater reliability and could not be distinguished from the Disorganization subscale regarding the patterns of

47

correlations with other measures. Hence, the Hyperactivation subscale needs particular attention and refinement before further use2.

Our exploratory factor analysis resulted in a three-factor solution, which does not correspond with the underlying theoretical model (see figure 2). However, this result likely reflects sample characteristics and should be interpreted with caution given the small sample size. In terms of future studies, there is a need for testing PAT-RS in larger samples that also include patients from clinical populations that are more likely to develop an insecure attachment to the therapist (such as borderline or psychotic patients). Further, in order to examine the convergent validity of the PAT-RS subscales in more detail, future studies should optimally also include different methods for assessing attachment to therapist (i.e., self-report such as the CATS or observer-rated measures such as PT-AAI). Recently, Talia and colleagues (2014) developed the Patient Attachment Coding System (PACS), which may be used to assess patient attachment to therapist directly from session transcript. While this approach is new and may also need further refinement, one prospect for the future would be to cross-validate the instruments by comparing PAT-RS-ratings of research interviews with PACS-codings of session transcripts.

One unexplored issue concerns what kind of interview data is required for reliable use of the PAT-RS. In study III we applied PAT-RS to data obtained with the PTI and ORI, neither of which was specifically designed to tap patients’ attachment to their therapist. The interview protocols involved just a few open-ended questions regarding the patient’s retrospective view of the therapy process as well as asking the patient for a detailed description of the therapist as a person. At present, this should be considered the minimum requirements for patient narratives to be measured with PAT-RS. Until further studies have clarified the optimal data for using the PAT-RS we recommend the use of multiple raters and that inter-rater reliability is estimated independently for the particular material at hand.

Secure attachment to therapist and change In line with an attachment perspective on therapeutic process and change, study IV indicated that patients’ level of secure attachment to their therapist at termination was associated with outcome. However, since we were only able to measure attachment to therapist at termination, the associations cannot be interpreted as causal, or even predictive. For example, it is quite possible that changes in distress led to a more secure attachment rather than the other way around. Still, the results mirror the patients’ and therapists’ subjective views of therapeutic action that emerged in study I and II. Further, given that secure attachment to therapist has previously been linked to important in-session processes (i.e., Janzen et al., 2008; Mallinckrodt et al., 2005; Romano et al., 2008; Saypol & Farber, 2010), as well as to reduction of distress over time (Sauer et al., 2010), the results do suggest that the development of a secure attachment to the therapist during therapy may be an important mechanism of change underlying psychoanalytic psychotherapies.

Further in line with this notion, we found that the level of secure attachment to therapist at termination predicted continued improvement in functioning during follow-up. 2 The current English version of PAT-RS is included in the appendix of this thesis. The prototypical descriptions were refined during translation with particular attention to the issue of differentiating the insecure subscales. Thus, the English version of PAT-RS differs somewhat from the original scale that was used in study III.

48

However, this was only true after patient-rated alliance had been accounted for, indicating the presence of a “classic” suppression effect (Pandey & Elliott, 2010). This suggests that the weight of Security increased when patient-rated alliance (with which it is correlated) was entered in the model since the alliance removed irrelevant predictive variance from Security. Thus, the association between Security and continued improvement was limited to the unique variance of Security (i.e., not shared with patient-rated alliance), which may have implications for the differentiation of the constructs (discussed further below).

Interestingly, the suppression effect was only present in terms of global functioning and not for change in symptoms or interpersonal problems. This may indicate that the finding was simply coincidental or perhaps an artifact of other variables not accounted for. Another possible interpretation is that the establishment of a secure attachment to the therapist is particularly important in terms of improving patients’ overall functioning. In their proposed meta-model of the therapeutic relationship, Wampold and Budge (2012) suggest that the “real relationship” (of which secure attachment to therapist may be considered a part; see Moore & Gelso, 2011) is particularly related to improvements in functioning and quality of life. In contrast, agreements on tasks and goals (i.e., the alliance), as well as specific therapeutic techniques, may be more important for symptom change. The results of study IV seem to be in line with this model.

One unexpected finding was that length of therapy was unrelated to the level of secure attachment to therapist at termination. This was surprising since “having time” emerged as an important factor related to the establishment of a safe therapeutic relationship from both the patients’ and the therapists’ view in studies I and II. Also, duration of treatment has been associated with a more secure attachment to the therapist in some studies (Bachelor et al., 2010; Mallinckrodt et al., 1995), although the timeframe of the treatments were much shorter than in the present sample. Our results suggest that treatment duration (at least beyond 2 months which was the shortest treatment course in YAPP) is not automatically associated with the development of a more secure attachment to the therapist. Certainly, the establishment of a secure therapeutic relationship is a complex process, involving the interaction of several patient and therapist factors, particular in-session processes, as well as the adaptation and timing of specific interventions (Frederickson, 2013; Janzen et al., 2008; Muller, 2010), and should not be considered a simple matter of time.

We were also surprised that we did not find any indication of between-therapist variability in terms of the level of secure attachment to therapist at termination, especially since there was indication (albeit non-significant) of such variability in terms of the alliance. This may have been due to low power, but another interpretation could be that therapist differences in terms of strategies for fostering a secure therapeutic attachment were small in this sample. Although no manual was used in YAPP, perhaps the therapists worked quite similar in this regard. Another possibility is that the development of a secure attachment in therapy was more dependent on patient factors that were not measured in the study (such as, for example, the patients’ pre-treatment attachment status).

Secure attachment to therapist and the alliance constructs revisited The results of study IV may have important implications when it comes to differentiating secure attachment to therapist from the therapeutic alliance. First, we found that the correlation between Security and patient-rated alliance (r = .47) was notably weaker

49

compared with studies that used self-report to measure both constructs (Bachelor et al., 2010; Lunsford, 2010; Mallinckrodt et al., 1995, 2005; Romano et al., 2008; Sauer et al., 2010). This indicates that an observer-based approach for assessing secure attachment to therapist may be useful for reducing common method bias associated with the use of self-reports only. Further, Security remained significantly associated with changes in both symptoms and global functioning when the alliance was accounted for in our models. Thus, trained raters may capture the specific attachment-related functional qualities of the relationship that relate to outcome beyond the patient’s experience of a positive alliance.

The idea that secure attachment to therapist and the therapeutic alliance comprise qualitatively different aspects of the therapeutic relationship related to outcome is further supported by the suppression effect found in study IV. This indicated that the unique variance of patient-rated alliance was associated with deterioration in functioning during follow-up, while secure attachment to therapist was related to further improvement. One possible interpretation is that patient-rated alliance in later stages of therapy includes a “good-bye effect”, leading some patients to overestimate the quality of the collaboration with the therapist when approaching termination. Another possibility is that, over the course of therapy some patient-therapist dyads may have formed alliances that had insecure qualities from the view of attachment theory. Such alliances may be considered “strong” by the patient but the therapeutic process may not have had an optimal impact on an emotional level, leading to less robust results in the long run. Our result may indicate that an observer-rated approach for assessing attachment to therapist is more sensitive to detecting such nuances, leading to better differentiation of the attachment to therapist and alliance constructs.

Importantly, however, it should be kept in mind that patient-rated alliance is not the same as the alliance construct “in itself”, but rather reflects one particular perspective on the alliance (Muran & Barber, 2010). Hence, the relationships between alliance and secure attachment may differ depending on the perspective used when assessing each construct. In study III we included therapist-rated alliance and it related to secure attachment with a similar moderate strength (r = .40) as patient-rated alliance, indicating that our observer-approach also differentiates secure attachment from the therapist perspective on the alliance. Yet, perhaps the alliance would relate more strongly to Security if also rated by an independent observer (although assessing both constructs with observers would of course re-introduce the issue of common method bias). Future studies should include both self-report and observer-rated measures of attachment to therapist, as well as the alliance, in order to clarify their respective relations and associations with outcome.

Psychoanalytic psychotherapy with young adults Since all patients included in the studies were young adults, some of the results may be specific for psychoanalytic work with this age group. For example, Jacobs (1988) suggests that since young adults are in a transient life situation, they may be very focused on “real-world issues” and therefore lack motivation for reflecting upon themselves and past experiences. Perhaps the findings of study I, indicating that some patients experienced self-knowledge as insufficient for change and wished for a more active therapist, is typical of patients in this age group. However, not all patients in our sample described such experiences and the curative factors that emerged in the same model contrast the view that young adults are not motivated for exploratory work. Further, similar findings regarding negative

50

experiences to specific factors in psychoanalytic psychotherapy were also reported in the study by Nilsson and colleagues (2007), in which the mean patient age was 43. Consequently, negative reactions of this kind seem likely to be related to factors other than age, such as patient personality characteristics and/or the therapists’ specific way of relating and intervening in the therapeutic process.

Further, since young adults are in the process of separating from internal and external parental figures, it has been suggested that engaging in a close therapeutic relationship with a therapist roughly of the same age as their parents may be a particular challenge for young adults (e.g., Barnett, 1971; Escoll, 1987; Jacobs, 1988; Pearl, 2008; Perelberg, 1993). Contrary to our expectations, however, patients rarely mentioned the therapists’ age as an issue in the interviews. Rather, the age and experience of the therapist was typically viewed as a positive factor, contributing to the experience of a “special kind of relationship” and likely also to the “stronger wiser” component of the attachment.

In connection to this, it is notable that the only age-specific category that emerged in the qualitative analyses was “young adults intrinsic developmental force” in study II. Here, the therapists described the transitional period of young adulthood as a positive, as well as a potentially negative, contributor to the therapeutic process. On the positive side, the young adults were described as having an inherent curiosity in themselves and others, which enhanced motivation and engagement in treatment. Further, the therapists noted that the young adult patients typically had to face new situations outside the therapy office (i.e., starting a new job, engaging in romantic relationship, etc.), which provided natural opportunities for corrective experiences. On the negative side, the therapists also experienced that the young adults mobility could interfere with the establishment of a close therapeutic relationship. Sometimes therapists interpreted the practical problems that typically emerged when patients switched jobs or moved away for studying as signs of resistance related to patients’ fear of engaging in a close relationship. However, considering that young adulthood involves a general striving toward individuation, avoidance of closeness and dependency may be a natural tendency in this age and should not automatically be interpreted as resistance. Thus, psychoanalytic work with young adults may require more flexibility than usual on the therapist’s part as well as an openness to explore when such practical problems may be signs of resistance and when they are not in a non-defensive and non-controlling manner.

The therapists’ generally positive view of young adults “intrinsic developmental force” seem in line with the notion that young adulthood involves a strong potential for personality change and emotional growth (Tanner & Arnett, 2009). In this regard, it is also interesting to note that we did not find any relationship between the length of treatment and outcome, nor between treatment length and a more secure attachment to the therapist, in study IV. Perhaps this reflects young adults potential for rapid engagement and change, which also challenges the commonly held belief (among lay persons, as well as some experienced therapists alike) that psychoanalytic psychotherapy must necessarily be “long-term” in order to produce lasting change. In fact, contrary to our expectations, longer therapy predicted deterioration in symptoms and interpersonal problems during follow-up in study IV. Since length of treatment was associated with more symptoms and interpersonal problems at intake, this may reflect that patients who were more troubled tended to stay longer in treatment (perhaps due to difficulties in establishing a secure attachment to the therapist) and were more likely to relapse after termination.

51

Methodological considerations Strengths All four studies in this thesis are based on data collected in the context of a naturalistic, longitudinal project, evaluating psychoanalytic psychotherapy as it was routinely practiced at the Institute of Psychotherapy in Stockholm. The overall design of the project contains several features that may be considered strengths with regards to external validity (Kazdin, 2003; Leichsenring, 2004; Seligman, 1995). For example, just as in ordinary clinical practice, the treatments were self-corrective, meaning that the therapists were allowed to adjust the treatment according to their clinical judgment of each patient’s needs. Further, the treatments were not fixed in duration but rather continued until the therapist and patient agreed upon termination or the patient dropped out. The patients’ were also self-selected for psychotherapy and sought treatment for a mixture of psychological complaints, including low self-esteem, depression, anxiety and relational difficulties. Further, inclusion in the project, as well as the allocation to either individual or group therapy, was based on clinical judgment of suitability in conjunction with the patients’ own preferences. Thus, the data was collected under conditions that are highly representative of clinical practice.

Another strength of the project involves the extensive qualitative data collection. This data provided an opportunity to explore the patients’ own meaning construction in study I, as well as highly educated and experienced therapists clinical thinking in study II. The qualitative methodology followed formalized steps and two researchers collaborated in the later stages of the coding process. Efforts were made to “bracket” existing theory and own values (Elliott et al., 1999; Malterud, 2001). In order to validate the results and reduce researcher subjectivity, feedback from four respondents was integrated in the coding process of study II. The amount of interview data was sufficient to reach saturation points in the qualitative analyses.

The rich qualitative material also allowed us to explore the utility of a new observer-rated instrument for patient-therapist attachment in study III. Before using PAT-RS, the raters were trained by the developer of instrument. During the rating procedure all raters were kept blind with regards to patient outcome in terms of the quantitative measures. Further, the standardized measures used in study III and IV tapped multiple domains of distress and functioning from both observer and patients’ self-reported perspectives, providing a broad picture of outcome. Comparing with the few studies that have investigated secure attachment to therapist and outcome so far (i.e., Sauer et al., 2010; Wiseman & Tishby, 2014), study IV has several novel features that may add to the literature: we assessed secure attachment to therapist at termination of long-term therapy, used a new observer-rated measure to assess patient-therapist attachment quality, and included change during follow-up in the analyses.

An overall strength of this thesis is that therapeutic action is explored from different perspectives (patient, therapist, observer) and that both qualitative and quantitative methods were used. The general correspondence of the results across the four studies strengthens the conclusions drawn.

52

Limitations In contrast to the strength in terms of external validity, the naturalistic design of YAPP was typically weak in terms of its internal validity (i.e., no control group, no randomization to treatments, no manual or treatment fidelity checks, etc.). By design, naturalistic studies do not permit any strong causal conclusions (Kazdin, 2007, 2009). Therefore, it should be kept in mind that the aim of this thesis was exploratory and the results must be evaluated in relation to other studies that may have stronger designs in terms of internal validity.

Further, it should also be kept in mind that the material in all four studies was analyzed at the group level. Hence, we cannot say if our results may differ between different subgroups of patients. For example, perhaps the conceptual models that emerged in study I and II would have been different if the studies were focused on patients that primarily sought treatment for a particular anxiety disorder, depression or maladaptive personality traits. Also, the associations between secure attachment to therapist and outcome found in study IV might be stronger or weaker in different subgroups (e.g., patients who differ in terms of their pre-treatment attachment status or other personality characteristics).

Also, the patients included in the studies should not be considered as representative of all young adults in distress. The gender distribution in the sample was uneven and most came from highly educated families, residing in the greater Stockholm area. Likewise, the therapist involved in YAPP should not be considered a representative sample of all psychoanalytically oriented therapists. The therapists were highly educated with long clinical experience. Further, the therapists were also involved as teachers and supervisors at the Institute of psychotherapy, which was a specialized unit for treatment, training and research in psychoanalytic psychotherapy.

As already mentioned, no manual was used. However, the therapists shared a psychoanalytic frame of reference and attended regular clinical meetings during the study, ensuring that the treatments delivered were psychoanalytic in orientation. Still, it is not clear to what extent the results may generalize to different forms of psychoanalytic psychotherapy. For example, it is possible that patients’ and therapists’ views of therapeutic action, as well as the impact of a secure attachment to therapist, differ in focused short-term psychodynamic psychotherapies compared with long-term treatments or psychoanalysis. Further, although the qualitative data was rich in terms of patients’ and therapists’ overall views of curative and hindering factors and processes, the interviews did not contain much information at the level of specific therapist interventions. Thus, the issue of when and how particular interventions may aid or thwart the development of a secure therapeutic attachment remains.

Some study-specific limitations also warrant mentioning. In study I and II, the interview material was selected based on availability. Although the patients and therapists seemed representative of their respective samples in YAPP, we cannot rule out the possibility that they differed in some respect. Further, in terms of the data analytic strategies employed in study I and II, some deviations from strict grounded theory methodology had to be accepted. The semi-structured interview manual was constructed in advance and interviewers other than the researchers themselves conducted the interviews, which may have limited the researchers sensitivity to the data. Further, there was no collection of new data (i.e., “theoretical sampling”) to deepen categories, as recommended by Strauss and Corbin (1998). This may have limited the specificity of the results, as it is possible that important experiences and processes were not captured in the interviews.

53

A major limitation in study III was that the interviews protocols were not specifically focused on assessing the patients’ attachment to their therapist. This led to large variation in terms of how much specific data was available for rating each component in PAT-RS. The inter-rater reliability of the PAT-RS subscales may be better if PAT-RS is applied to interviews were the interviewer probes more for attachment specific material. Another limitation in study III was the lack of other measures of attachment to therapist or of patients’ attachment styles, with which we could explore the convergent validity of PAT-RS subscales. Also, characteristics of the sample, as well as the use of termination as assessment point, likely limited the variability of the insecure subscales in PAT-RS.

As already mentioned above, a major limitation of study IV was that we were only able to assess attachment to therapist from the interviews at termination. Although the assessment involved the patients’ retrospective account of the psychotherapy process, we have no control over the temporal relationship between Security and outcome. Consequently, we cannot say if secure attachment to therapist preceded improvements or was the result of them or of other factors. Further, we did not have access to patients’ pre-treatment attachment characteristics. It is possible that patients who had a predominantly secure attachment style at the start of therapy mainly accounted for the association between secure attachment to therapist and outcome in this study. However, previous studies have found mixed results regarding the relationship between patients’ global attachment status and the quality of patients’ specific attachment to their therapist (Mallinckrodt et al., 1995, 2005; Moore & Gelso, 2011; Romano et al., 2008; Sauer et al., 2010; Wiseman & Tishby, 2014). Hence, the relationships between secure attachment style and relationship-specific secure attachment to the therapist should not be assumed to be strong.

Concluding thoughts Two tentative process models The aim of this thesis was to explore therapeutic action in psychoanalytic psychotherapy. Clearly, the included studies do not provide any definite answer to this complex issue and much more research is needed before (if ever) a comprehensive model, accepted by all strands of psychoanalytic thought, will be reached (Gabbard & Westen, 2003; Fonagy & Kächele, 2009; Kernberg, 2007). Still, the overall result of this thesis suggests that the establishment of a secure attachment to therapist may be an important mechanism of therapeutic change. Further, although the studies do not provide information at the level of specific therapist interventions, the results indicate that certain interactions between patient and therapist may facilitate or obstruct the development of a secure therapeutic relationship. Integrating the results of all four studies, two tentative process models are proposed below that may be useful for clinical practice and further research.

The first model integrates the curative factors that emerged in the studies and suggests a broaden-and-build cycle of attachment security development and change (Figure 5). The model indicates that the process of opening up in treatment may typically lead to corrective emotional experiences for the patient, leading to shifts toward greater attachment security in the therapeutic relationship. This will enable mutual exploration, which may foster

54

Figure 5. A broaden-and-built cycle of attachment security development and change.

an increased capacity for mentalization. In turn, increased ability to mentalize may strengthen the patient’s sense of agency, self-definition and self-efficacy and promote further corrective experiences outside the therapeutic setting. Reductions in symptoms and interpersonal distress will increase the patient’s confidence in the treatment and willingness to open up and engage even more.

Figure 6. A react-and-disconnect cycle of attachment insecurity maintenance.

The second model integrates the hindering factors that emerged in a react-and-

disconnect cycle of attachment insecurity maintenance that may lead to the preservation of patients’ symptoms and distress (see Figure 6). This model suggests that patients’ negative reactions to specific elements in treatment may activate insecure attachment strategies (deactivation in particular). This, in turn, may evoke therapist countertransference and prompt the therapist to hold even tighter to the specific factors in their treatment model. Consequently, the therapeutic process becomes characterized by disconnection and/or pseudo-

Therapeutic process characterized by disconnection and/or pseudo-‐

mentalizing

Therapist countertransference

Therapist holding tighter to specific treatment factors

Mutual exploration inhibited

Activation of insecure attachment strategies

Negative reaction to specific treatment

element

Preservation of symptoms and interpersonal distress

Increased sense of agency, self-‐definition

and self-‐efficacy

Reduction of symptoms and interpersonal distress

Corrective emotional experiences in sessions

Mutual exploration

Increased capacity for mentalization

Shift towards greater attachment security

“Opening up”

Corrective emotional experiences outside

sessions

55

mentalizing, blocking genuine exploration, integration and change. The preservation of symptoms and distress may trigger even further negative reactions and disconnection; however, in some cases the therapeutic alliance may still be rated strong from the view of the patient (and possibly also by the therapist) due to the relative emotional “safety” provided by the insecure attachment strategies.

The underlying arrows in the models suggest that the included factors follow each other in a step-by-step fashion, but they are more likely to interact with each other in reciprocal ways. Based on these models, some general implications for practice are suggested below.

Implications for practice The results of this thesis suggest that therapists should actively strive toward fostering a secure attachment relationship in treatment. A sufficient sense of security is necessary for activation of the exploratory system, which will enable the patient to approach painful thoughts and feelings in the therapeutic process. A sense of security will also facilitate a mutual exploration process that promotes mentalization (Allen et al., 2008). Further, following Bowlby’s (1988) statement that: “…unless a therapist can enable his patient to feel some measure of security, therapy cannot even begin” (p. 140), therapists should be attentive to the process of attachment formation from the very first session. The first step of “opening up” may be experienced as stressful by some patients’ and, consequently, insecure attachment strategies may be evoked from the very beginning of treatment. Therapists should be attentive to indicators of insecure strategies being activated in the relationship and specifically inquire into the patient’s experiences when such indicators emerge. This may open the opportunity to test and revise the patient’s expectations, potentially leading to corrective emotional experiences and shifts towards greater attachment security.

Negative reactions to specific treatment elements may also signal that there are important differences between patients’ and therapists’ implicit theories of cure that needs to be elucidated and discussed (Philips, Werbart, & Schubert, 2005; Philips, Werbart, Wennberg, Schubert, 2007; Werbart & Levander, 2006, 2011). If left unattended, such discrepancies may otherwise lead to an experience of mismatch, which will obstruct the formation of a collaborative alliance, as well as the establishment of a secure therapeutic attachment. Further, perhaps due to the activation of insecure attachment strategies that inhibit mutual exploration, therapists may not be aware of the negative experiences of their patients. Therefore, the therapist needs to actively invite the patient’s view through out treatment in order to and check his or her ideas with those of the patient (Hill & Knox, 2009; Satran, 1995).

Our results also indicate that patients who experience mismatch often have their own ideas of what could improve the therapeutic work. The therapists should make efforts to explore these ideas; however, a naive acceptance of the patient’s view is probably not curative in itself. Research on the alliance ruptures (Safran & Muran, 2000; Safran et al., 2011) indicates that the negotiation of divergences between patients’ and therapists’ views may restore a collaborative alliance, as well as provide corrective emotional experiences that may deepen the patients’ sense of security in treatment. On the other hand, if the patient’s and the therapist’s ideas of the goals and tasks in treatment are incompatible and non-negotiable, therapists should consider referring the patient to another therapeutic modality rather than prolonging treatment in the hope that “time will do its work”.

56

Further, besides facilitating corrective emotional experiences in the therapeutic relationship, therapists should also pay close attention to patients’ extra-therapy activities and help promote corrective experiences outside the sessions. One of the main common factors found in previous research is “encouragement of gradual practice” (Elliott & James, 1989), which seemed to be absent from the view of disappointed patients in study I. From the position of providing a “secure base” for the patient, therapists should actively encourage patients to use what they have acquired in therapy for adaptive action in their lives. According to an integrative relational psychoanalytic view (Frank, 1999; Gold & Stricker, 2001; Wachtel, 2014) selective incorporation of “action-oriented” interventions may be considered at some points in the therapeutic process. Focusing on the establishment of a secure therapeutic relationship at the same time as attending to patients’ adaptive actions will also facilitate mature forms of both relatedness and self-definition (Blatt, 2008).

One typical negative patient experience that emerged was related to perceived therapist passivity. Although the passive psychoanalytic therapist is a caricature of the past in many respects, some practicing therapists may still have been trained in a “less is more“ approach. Such therapists may feel uncomfortable with being active due to their training or they may simply lack the skills for intervening actively in a manner that is consistent with psychoanalytic theory (Katzman & Coughlin, 2013). Additionally, some therapists may also resort to a more passive and/or inflexible stance due to countertransference and/or insecure attachment strategies being evoked in them. From an attachment perspective, therapist passivity may undermine mentalization processes (Allen, Fonagy, & Bateman, 2008). Thus, therapists should be attentive to their own level of activity. Increased passivity and disengagement may signal to the therapist that a react-and-disconnect cycle has been triggered that needs particular attention.

As suggested above, patients’ avoidance of closeness may indicate that insecure attachment strategies have been evoked in the treatment process and/or that patients’ and therapists’ ideas of cure differ. In both instances, such experiences need the therapists’ active attention. However, perhaps particularly in the context of psychotherapeutic work with young adults, avoidance of closeness and dependency should not automatically be interpreted as signs of resistance. The therapist needs to be able to explore perceived obstacles to the therapeutic work in a non-defensive and non-controlling manner, also reflecting on their own contribution to the co-construction of hindrances. Actively inviting the patient’s perspective may foster the restoration of a collaborative therapeutic relationship, lead to corrective experiences and shifts toward security and reestablished mentalization processes (Allen, Fonagy, & Bateman, 2008).

Suggestions for future research The results of this thesis suggest that the development of a secure attachment to the therapist may be an important change mechanism in psychotherapy. Further, the formation and maintenance of an insecure attachment to the therapist may obstruct the therapeutic process, leading to non-improvement or even deterioration. However, in order to test these hypotheses more thoroughly, future studies need to be specifically designed to evaluate the impact of patient-therapist attachment quality on outcome. Due to a number of ethical and practical reasons, direct experimental manipulation of the patients’ attachment to their therapist is not

57

feasible. The most realistic step in this direction would be to investigate attachment to therapist as mediator of outcome (Kazdin, 2007, 2009).

In order to do this, it is imperative that future studies assess patient-therapist attachment, as well as the target outcomes, at several points in time in order to establish the temporal link between growth in attachment and change. Hypotheses regarding the temporal development of secure attachment and outcome could be tested within a multilevel growth model framework (e.g., Ljótsson et al., 2013) or with structural equation modeling (SEM). Similar to the findings regarding alliance and outcome (Falkenström et al., 2013, 2014; Tasca & Lampard, 2012; Zilcha-Mano et al., 2013), a reciprocal relationship between growth in secure attachment to therapist and therapeutic gains might be expected, which would also be theoretically consistent with the broaden-and-build cycle of attachment security development. In contrast, higher levels of insecure attachment to therapist may be expected to relate to less change or even predict deterioration.

One issue for future studies is how to measure attachment to therapist. While self-report may still be the most economic and practical approach, the results of this thesis suggests that an observer-based approach may be preferable, particularly if alliance is assessed from the patients’ perspective in the same study. In such studies, conducting attachment-focused interviews and assessing them with PAT-RS could be an appropriate approach. However, if the development of attachment to therapist were to be studied in detail, frequent interview assessments would likely be too cumbersome. An alternative would be to assess attachment to therapist at session level using PACS (Talia et al., 2014) or, perhaps, adopt PAT-RS for use on video recordings of sessions.

Additionally, patients’ pre-treatment attachment should be taken into account in order to investigate the possible role of attachment to therapist as mediator between global attachment styles and outcome. Recent research also indicates that the therapists’ attachment styles, as well as the match between therapists and patients in terms of attachment (Petrowski, Nowacki, Pokorny, & Buchheim et al., 2011; Petrowski, Pokorny, Nowacki, & Buchheim, 2013; Schauenburg et al., 2010; Wiseman & Tishby, 2014), may influence treatment outcome. While more research on the match between patients’ and therapists’ global attachment styles may yield important results, we suggest that future research might benefit even more from a relationship specific view of attachment development in the therapeutic process. Such a view could lead to more detailed process-oriented research focusing on how and when both secure and insecure attachments strategies are evoked and maintained in particular therapeutic relationships. This could potentially lead to the identification of specific in-session behaviors that foster secure attachment, as well as dissolve insecure strategies, an area in need of further research (Mallinckrodt, 2010; Obegi, 2008).

Lastly, since the attachment system is likely to influence any kind of helping relationship (Farber et al., 1995; Obegi, 2008), the quality of patient-therapist attachment may be regarded a common factor across psychotherapy orientations. Although emanating from the psychoanalytic tradition, attachment theory has also been assimilated into several models in the cognitive-behavioral tradition in the last decade (e.g., Gilbert, 2009; Liotti, 2007; Young, 2003). Future studies should investigate the development and impact of patient-therapist attachment in different treatment modalities. As suggested by Connors (2011), ttachment theory may provide a “secure base” for the future of psychotherapy integration.

59

ACKNOWLEDGEMENTS IN SWEDISH Först av allt vill jag tacka alla patienter och terapeuter som medverkade i YAPP-projektet. Utan er villighet att dela med er av era erfarenheter och upplevelser för forskning hade denna avhandling inte varit möjlig. Tack också till de sekreterare, PTP-psykologer och annan personal på forna Psykoterapiinstitutet som var med och samlade in materialet i YAPP. Era ansträngningar har inneburit ett enormt privilegium för mig! Jag vill uttrycka min djupaste tacksamhet till min huvudhandledare, Andrzej Werbart. Du har visat en (närmast obegripligt) stor tilltro till mig från första början av min PTP-tjänst på Psykoterapiinstitutet. Din tillförsikt och uppmuntran har varit ovärderlig i stunder när jag själv tvivlat. Du har dessutom varit enormt generös och gett mig förtroende att arbeta vidare med frågeställningar i YAPP som jag vet ligger dig själv varmt om hjärtat. Det har alltid varit väldigt inspirerande att diskutera alla teoretiska och metodologiska frågor som dykt upp under arbetets gång med dig. Även om vi kanske inte alltid sett saker på exakt samma sätt så har jag alltid upplevt att du stöttat mig att komma fram till mina egna slutsatser. Tack för allt du bidragit med – tid, engagemang, din breda och djupa kunskap på området samt lite välbehövlig utmaning ibland också – det har fått mig att växa! Jag vill även rikta ett stort tack till mina bihandledare Pia Risholm-Mothander och Marie-Louise Ögren. Ni har båda utgjort en viktig ”trygg bas” för mig (ja, även efter det att du slutade på Psykologiska Institutionen, Marie-Louise!). Tack för all uppmuntran, engagemang och stöd, både vad gäller avhandlingsarbetet och när det gäller den undervisning jag har genomfört på institutionen. Stort tack också till Rolf Sandell. Redan under psykologutbildningen var du en viktig inspirationskälla till att jag överhuvudtaget började intressera mig för psykoterapiforskning. Tack för den nyfikenhet och intresse du visat för mitt avhandlingsarbete. Särskilt tack för din hjälp med statistiken i studie III och IV – jag tror jag börjar greppa en del på egen hand nu! Tack också till Fredrik Falkenström för samarbetet med revisionen av studie IV. Avund kan ibland vara en värdefull drivkraft och kanske lyckas jag en dag lära mig göra multilevel-analyser lika självklart som du! Men det troliga är väl att jag mailar dig när kör fast nästa gång igen… Tack till August Ekström och Susanna Sjögren för allt jobb ni la ner på kodningsarbetet till studie III.

60

Jag vill också tacka Björn Philips för all värdefull feedback och uppmuntrande kommentarer till de olika delstudierna. Tack till Pehr Granquist för värdefull feedback på det första utkastet av PAT-RS. Thanks to Alessandro Talia and Madeleine Miller-Bottome for inspiring discussions and great company at the SEPI 2013 and SPR 2014 meetings. Looking forward to following your scientific work and to meeting you again soon! I also want to thank professor John Clarkin and professor Anna Buchheim for their valuable feedback on the development of PAT-RS at the Research Training Program (RTP) of the International Psychoanalytic Association in Berlin, 2012. Jag vill tacka Paul Bjerre stiftelsen för de stipendiemedel som har gett mig möjlighet att skapa goda kontakter med andra forskare på området och sprida mina arbeten vidare. Tack också till Groshinsky fonden för de stipendiemedel som gjorde studie II möjlig att genomföra. Tack till familj och vänner för att ni varit nyfikna på avhandlingen och för att ni haft tålamod med mina svårigheter att beskriva innehållet på ett begripligt sätt! Till sist, varmt tack till Maud och Arvid. För att ni är de fantastiska människor ni är helt enkelt och för ert tålamod med min mentala frånvaro under delar av skrivarbetet. Jag älskar er. Stockholm, oktober 2014 Peter Lilliengren

61

REFERENCES Abbass, A., Kisely, S., Town, J. M., Leichsenring, F., Driessen, E., de Maat, S. C. M., … Crowe, E. (2014). Short-term

psychodynamic psychotherapies for common mental disorders (Review). Cochrane Library, 7, 1–84. doi:10.1002/14651858.CD004687.pub4

Ablon, J., & Jones, E. (1998). How expert clinicians’ prototypes of an ideal treatment correlate with outcome in psychodynamic and cognitive-behavioral therapy. Psychotherapy Research, 8(1), 71–83. doi:10.1080/10503309812331332207

Adamson, L., & Lyxell, B. (1996). Self-concept and questions of life: Identity development during late adolescence. Journal of Adolescence, 19(6), 569–582. doi:10.1006/jado.1996.0055

Adatto, C.P. (1980). Late adolescence to early adulthood. In S.G. Greenspan & G.H. Pollock (Eds.), The course of life: Psychoanalytic contributions towards understanding personality development (pp. 463–476). Adelphi, MD: NIMH, Mental Health Study Center.

Alexander, F., & French, T. M. (1946). Psychoanalytic therapy: Principles and applications. New York, NY: Ronald Press. Allen, J. G., & Fonagy, P. (Eds.). (2006). Handbook of mentalization-based treatment. Chichester, UK: John Whiley &

Sons. Allen, J. G., Fonagy, P., & Bateman, A. W. (2008). Mentalizing in clinical practice. Arlington, VA: American Psychiatric

Publishing. Anderson, T., Lunnen, K. M., & Ogles, B. M. (2010). Putting models and techniques in context. In B. L. Duncan, S. D.

Miller, B. E. Wampold, & M. A. Hubble (Eds.), The heart and soul of change. Delivering what works in therapy (2nd ed., pp. 143-166). Washington, DC: American Psychological Association.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

American Psychological Association. (2005). Policy statement on evidence-based practice in psychology. Retrieved from: https://www.apa.org/practice/guidelines/evidence-based-statement.aspx

Armelius, K. (2001). Reliabilitet och validitet för den svenska versionen av SASB - självbildstestet [Reliability and validity for the Swedish version of SASB introject questionnaire]. Department of Applied Psychology, Umeå University, Sweden.

Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55(5), 469–480. doi:10.1037//0003-066X.55.5.469

Arnett, J. J. (2007). Emerging Adulthood: What is it, and what is it good for? Child Development Perspectives, 1(2), 68–73. doi:10.1111/j.1750-8606.2007.00016.x

ATLAS.TI. (2000). The knowledge workbench 4.2. Computer software. Berlin, Germany: Scientific Software Development. Auerbach, J., & Blatt, S.J. (1996). Self-representation in severe pathology: The role of reflexive self-awareness.

Psychoanalytic Psychology, 13(3), 297–341. doi:10.1037/h0079659 Baardseth, T. P., Goldberg, S. B., Pace, B. T., Minami, T., Wislocki, A. P., Frost, N. D., … Wampold, B. E. (2013).

Cognitive-behavioral therapy versus other therapies: Redux. Clinical Psychology Review, 33(3), 395–405. doi:10.1016/j.cpr.2013.01.004

Bachelor, A., Meunier, G., Laverdiére, O., & Gamache, D. (2010). Client attachment to therapist: Relation to client personality and symptomatology, and their contributions to the therapeutic alliance. Psychotherapy, 47(4), 454–68. doi:10.1037/a0022079

Baldwin, S. A., & Imel, Z. E. (2013). Therapist effects: Findings and methods. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed., pp. 258–297). Hoboken, NJ: John Wiley & Sons.

Barber, J. P., Luborsky, L., Crits-Christoph, P., Thase, M. E., Wiess, R., Frank, A., ... Gallop, R. (1999). Therapeutic alliance as a predictor of outcome in treatment of cocaine dependence. Psychotherapy Research, 9(1), 54–73. doi:10.1080/10503309912331332591

62

Barber, J. P., Muran, C. J., McCarthy, K. S., & Keefe, J. R. (2013). Research on dynamic therapies. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed., pp. 443–494). Hoboken, NJ: John Wiley & Sons.

Barnett, J. (1971). Dependency conflicts in the young adult. Psychoanalytic Review, 58, 111–125. Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social and Personal Relationships,

7(2), 147– 178. doi:10.1177/0265407590072001 Bartholomew, K., & Horowitz, M. (1991). Attachment styles among young adults: A test of a four category model. Journal

of Personality and Social Psychology, 61(2), 226–244. doi:10.1037/ 0022-3514.61.2.226 Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York, NY: International Universities Press. Bender, R., & Lange, S. (2001). Adjusting for multiple testing - when and how? Journal of Clinical Epidemiology, 54, 343–

349. doi:10.1016/S0895-4356(00)00314-0 Benish, S., & Imel, Z. (2008). The relative efficacy of bona fide psychotherapies for treating post-traumatic stress disorder: a

meta-analysis of direct comparisons. Clinical Psychology Review, 28(7), 746–758. doi:10.1016/j.cpr.2007.10.005 Benjamin, L. S. (1984). Principles of prediction using Structural Analysis of Social Behavior. In R. A. Zucker, J. Aaronoff

& A. J. Rabin (Eds.), Personality and the prediction of behavior (pp. 121–174). New York: Academic Press. Benjamin, L. S. (2000). SASB Intrex user’s manual. Salt Lake City, UT: University of Utah. Benjamin, L. S., Rothweiler, J. C., & Critchfield, K. L. (2006). The use of Structural Analysis of Social Behavior (SASB) as

an assessment tool. Annual Review of Clinical Psychology, 2,83–109. doi:10.1146/annurev.clinpsy.2.022305.095337 Bernecker, S. L., Levy, K. N., & Ellison, W. D. (2014). A meta-analysis of the relation between patient adult attachment

style and the working alliance. Psychotherapy Research, 24(1), 12–24. doi:10.1080/10503307.2013.809561 Blatt, S. J., & Behrends, R. S. (1987). Internalization, separation-individuation, and the nature of therapeutic action.

International Journal of Psychoanalysis, 68, 279–297. Blatt, S. J., Stayner, D. A., Auerbach, J. S., & Behrends, R. S. (1996). Change in object and self-representations in long-

term, intensive, inpatient treatment of seriously disturbed adolescents and young adults. Psychiatry: Interpersonal and Biological Processes, 59, 82–107.

Blatt, S. J., & Auerbach, J. S. (2001). Mental representation, severe psychopathology, and the therapeutic process. Journal of the American Psychoanalytic Association, 49(1), 113–159. doi:10.1177/ 00030651010490010201

Blatt, S. J., Auerbach, J. S., & Behrends, R. S. (2008). Changes in the representation of self and significant others in the treatment process: Links between representation, internalization, and mentalization. In E. J. Jurist, A. Slade, & S. Bergner (Eds.), Mind to mind: Infant research, neuroscience, and psychoanalysis (pp. 225–263). New York, NY: Other Press.

Blatt, S. J. (2008). Polarities of experience: Relatedness and self-definition in personality development, psychopathology, and the therapeutic process. Washington, DC: American Psychological Association Press.

Blatt, S. J., & Luyten, P. (2009). A structural-developmental psychodynamic approach to psychopathology: two polarities of experience across the life span. Development and Psychopathology, 21(3), 793–814.

Blatt, S. J., Zuroff, D. C., Hawley, L. L., & Auerbach, J. S. (2010). Predictors of sustained therapeutic change. Psychotherapy Research, 20(1), 37–54. doi:10.1080/10503300903121080

Bohart, A. C., & Tallman, K. (1999). How clients make therapy work: The process of active self-healing. Washington, DC: American Psychological Association.

Bohart, A. C. (2000). The client is the most important common factor: Clients’ self-healing capacities and psychotherapy. Journal of Psychotherapy Integration, 10(2), 127–149. doi:10.1023/A:1009444132104

Bohart, A. C., & Greaves Wade, A. (2013). The client in psychotherapy. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed., pp. 219–257). Hoboken, NJ: John Wiley & Sons.

Bordin, E. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research & Practice, 16(3), 252–260. doi:10.1037/h0085885

Bowlby, J. (1969), Attachment and loss, Vol. 1: Attachment. New York, NY: Basic Books. Bowlby, J. (1973). Attachment and loss, Vol. 2: Separation. New York, NY: Basic Books. Bowlby, J. (1980). Attachment and loss, Vol. 3: Loss, sadness and depression. New York, NY: Basic Books. Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. New York, NY: Basic Books. Baruch, G., & Fearon, P. (2002). The evaluation of mental health outcome at a community-based psychodynamic

psychotherapy service for young people: a 12-month follow-up based on self-report data. Psychology and Psychotherapy, 75(3), 261–78. doi:10.1348/147608302320365181

Breuer, J., & Freud, S. (1955). Studies on hysteria. In J. Strachey (Ed. & Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 2). London, UK: Hogarth Press. (Original work published 1895)

63

Budge, S. L., Moore, J. T., Del Re, a C., Wampold, B. E., Baardseth, T. P., & Nienhuis, J. B. (2013). The effectiveness of evidence-based treatments for personality disorders when comparing treatment-as-usual and bona fide treatments. Clinical Psychology Review, 33(8), 1057–1066. doi:10.1016/j.cpr.2013.08.003

Castonguay, L. G., Goldfried, M. R., Wiser, S., Raue, P. J., & Hayes, A. H. (1996). Predicting outcome in cognitive therapy for depression: A comparison of unique and common factors. Journal of Consulting and Clinical Psychology, 64(3), 497– 504. doi:10.1037/0022-006X.64.3.497

Castonguay, L. G., Constantino, M. J., & Holtforth, M. G. (2006). The working alliance: Where are we and where should we go? Psychotherapy, 43(3), 271–279. doi:10.1037/0033-3204.43.3.271

Castonguay, L. G., & Hill, C. (Eds.). (2007). Insight in psychotherapy. Washington, DC: American Psychological Association.

Castonguay, L. G., Boswell, J. F., Zack, S. E., Baker, S., Boutselis, M. a., Chiswick, N. R., … Holtforth, M. G. (2010). Helpful and hindering events in psychotherapy: A practice research network study. Psychotherapy: Theory, Research, Practice, Training, 47(3), 327–344. doi:10.1037/a0021164

Christian, C., Safran, J. D., & Muran, J. C. (2012). The corrective emotional experience: A relational perspective and critique. In L. G. Castonguay & C. E. Hill (Eds.), Transformation in psychotherapy: Corrective experiences across cognitive behavioral, humanistic, and psychodynamic approaches (pp. 51–67). Washington, DC: American Psychological Association

Cicchetti, D. V. (1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment, 6(4), 284–290. doi:10.1037/1040-3590.6.4.284

Connolly, M. B., Crits-Christoph, P., Shelton, R. C., Hollon, S., Kurtz, J., Barber, J. P., & Thase, M. E. (1999). The reliability and validity of a measure of self-understanding of interpersonal patterns. Journal of Counseling Psychology, 46(4), 472–482. doi:10.1037/0022-0167.46.4.472

Connolly, M. B., Crits-Christoph, P., Shappel, S., Barber, J.P., Luborsky, L., & Shaffer, C. (1999). Relations of transference interpretations to outcome in the early sessions of brief supportive-expressive psychotherapy. Psychotherapy Research, 9(4), 485–495. doi:10.1080/10503309912331332881

Connolly Gibbons, M. B., Crits-Christoph. P., Barber, J. P., & Schamberger, M. (2007). Insight in psychotherapy: A review of the empirical literature. In L. G. Castonguay & C. Hill (Eds.), Insight in psychotherapy (pp. 143–166). Washington, DC: American Psychological Association.

Connolly Gibbons, M. B., Crits-Christoph, P., Barber, J. P., Wiltsey Stirman, S., Gallop, R., A. Goldstein, L. A., Temes, C. M., & Ring-Kurtz, S. (2009). Unique and common mechanisms of change across cognitive and dynamic psychotherapies. Journal of Consulting and Clinical Psychology, 77(5), 801–813. doi:10.1037/a0016596

Connors, M. E. (2011). Attachment theory: A “secure base” for psychotherapy integration. Journal of Psychotherapy Integration, 21(3), 348–362. doi:10.1037/a0025460

Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment Research & Evaluation, 10(7), 1–9.

Côté, J. (2000). Arrested adulthood: The changing nature of maturity and identity in the late modern world. New York, NY: New York University Press.

Coughlin Della Selva, P. (1996). Intensive short-term dynamic psychotherapy. London, UK: Karnac Books. Crits-Christoph, P., & Mintz, J. (1991). Implications of therapist effects for the design and analysis of comparative studies of

psychotherapies. Journal of Consulting and Clinical Psychology, 59(1), 20–26. doi:10.1037/0022-006X.59.1.20 Crits-Christoph, P., & Connolly Gibbons, M. B. (2001). Relational interpretations. Psychotherapy: Theory, Research,

Practice, Training, 38(4), 423–428. doi:10.1037/0033-3204.38.4.423 Crits-Christoph, P., Connolly Gibbons, M. B., & Mukherjee, D. (2013). Psychotherapy process–outcome research. In M. J.

Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change. (6th ed., pp. 298–340). Hoboken, NJ: John Wiley & Sons.

Cuijpers, P., Andersson, G., Donker, T., & van Straten, A. (2011). Psychological treatment of depression: Results of a series of meta-analyses. Nordic Journal of Psychiatry, 65(6), 354–64. doi:10.3109/08039488.2011.596570

Daniel, S. I. F. (2006). Adult attachment patterns and individual psychotherapy: A review. Clinical Psychology Review, 26(8), 968–84. doi:10.1016/j.cpr.2006.02.001

Daly, K. D., & Mallinckrodt, B. (2009). Experienced therapists’ approach to psychotherapy for adults with attachment avoidance or attachment anxiety. Journal of Counseling Psychology, 56(4), 549–563. doi:10.1037/a0016695

Davanloo, H. (1990). Unlocking the unconscious. Chichester, UK: John Wiley & Sons. de Haas, M. A, Bakermans-Kranenburg, M. J., & van Ijzendoorn, M. H. (1994). The Adult Attachment Interview and

questionnaires for attachment style, temperament, and memories of parental behavior. The Journal of Genetic Psychology, 155(4), 471–86. doi:10.1080/00221325.1994.9914795

64

DeFife, J. A., Hilsenroth, M. J., & Gold, J. R. (2008). Patient ratings of psychodynamic psychotherapy session activities and their relation to outcome. Journal of Nervous and Mental Disease, 196(7), 538–547. doi:10.1097/NMD.0b013e31817cf6d0

Derogatis, L. R. (1994). Symptom checklist-90-R: Administration, scoring and procedures manual (3rd ed.). Minneapolis, MN: National Computer Systems.

Derogatis, L. R., Lipman, R. S., & Covi, L. (1973). SCL-90: An outpatient psychiatric rating scale – preliminary report. Psychopharmacology Bulletin, 9,13–27.

DeRubeis, R. J., Brotman, M. A., & Gibbons, C. J. (2005). A conceptual and methodological analysis of the nonspecifics argument. Clinical Psychology: Science and Practice, 12, 174–183. doi:10.1093/clipsy/bpi022

Diamond, D., Blatt, S. J., Stayner, D., & Kaslow, N. (1995). Differentiation–relatedness of self and object representations (revised scale). Unpublished research manual, Yale University, New Haven, CT, USA.

Diamond, D., Stovall-Mclough, C., Clarkin, J. F., & Levy, K. N. (2003). Patient-therapist attachment in the treatment of borderline personality disorder. Bulletin of the Menninger Clinic, 67(3), 227–259. doi:10.1521/bumc.67.3.227.23433

Diener, M. J., Hilsenroth, M. J., & Weinberger, J. (2007). Therapist affect focus and patient outcomes in psychodynamic psychotherapy: A meta- analysis. American Journal of Psychiatry, 164(6), 936–941. doi:10.1176/appi.ajp.164.6.936

Diener, M. J., & Monroe, J. M. (2011). The relationship between adult attachment style and therapeutic alliance in individual psychotherapy: A meta-analytic review. Psychotherapy, 48(3), 237–48. doi:10.1037/a0022425

Eagle, M. N. (2013). Attachment and psychoanalysis: Theory, research, and clinical implications. New York, NY: Guilford Press.

Elliot, R., & James, E. (1989). Varieties of client experience in psychotherapy: An analysis of the literature. Clinical Psychology Review, 9(4), 443–467. doi:10.1016/0272-7358(89)90003-2

Elliot, R., Fischer, C. T., & Rennie, D. L. (1999). Evolving guidelines for publication of qualitative research studies in psychology and related fields. British Journal of Clinical Psychology, 38(3), 215–229. doi:10.1348/014466599162782

Ellis, A. (1962). Reason and emotion in psychotherapy. New York, NY: Lyle Stuart. Emde, R.N. (1985). From adolescence to midlife: Remodeling the structure of adult development. Journal of the American

Psychoanalytic Association, 33, 59–112. Erikson, E. H. (1959). Identity and the life cycle. New York, NY: International Universities Press. Erikson, E. H. (1968). Identity: Youth and crisis. New York, NY: Norton. Escoll, P.J. (1987). Psychoanalysis of young adults: An overview. Psychoanalytic Inquiry, 7(1), 5–30.

doi:10.1080/07351698709533657 Evans, M. E. (2009). Prevention of mental, emotional, and behavioural disorders in youth: The Institute of Medicine Report

and implications for nursing. Journal of Child and Adolescent Psychiatric Nursing, 22(3), 154–159. doi:10.1111/j.1744-6171.2009.00192.x

Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. doi:10.1037/1082-989X.4.3.272

Falkenström, F., Grant, J., Broberg, J., & Sandell, R. (2007). Self-analysis and post-termination improvement after psychoanalysis and long-term psychotherapy. Journal of the American Psychoanalytic Association, 55(2), 629–674. doi:10.1177/00030651070550020401

Falkenström, F. (2009). Does psychotherapy for young adults in routine practice show similar results as therapy in randomized clinical trials? Psychotherapy Research, 20(2), 181–192. doi:10.1080/10503300903170954

Falkenström, F., Granström, F., & Holmqvist, R. (2013). Therapeutic alliance predicts symptomatic improvement session by session. Journal of Counseling Psychology, 60(3), 317–328. doi:10.1037/a0032258

Falkenström, F., Granström, F., & Holmqvist, R. (2014). Working alliance predicts psychotherapy outcome even while controlling for prior symptom improvement. Psychotherapy Research, 24(2), 146–159. doi:10.1080/10503307.2013.847985

Farber, B., Lippert, R. A., & Nevas, D. B. (1995). The therapist as attachment figure. Psychotherapy: Theory, Research, Practice, Training, 32, 204–212. doi:10.1037/0033-3204.32.2.204

Farber, B. A., & Metzger, J. A. (2009). The therapist as a secure base. In J. H. Obegi & E. Berant (Eds.), Attachment theory and research in clinical work with adults (pp. 46–70). New York, NY: The Guilford Press.

Ferenczi, S., & Rank, O. (1925). The development of psychoanalysis. Washington, DC: Nervous and Mental Disease Publishing.

Fonagy, P., Gergely, G., Jurist, E., & Target, M. (2002). Affect regulation, mentalization, and the development of the self. New York, NY: Other Press.

65

Fonagy, P. (2003). Genetics, developmental psychopathology, and psychoanalytic theory: The case for ending our (not so) splendid isolation. Psychoanalytic Inquiry, 23(2), 218–247. doi:10.1080/07351692309349032

Fonagy, P., & Kächele, H. (2009). Psychoanalysis and other long-term dynamic psychotherapies. In M. G. Gelder, J. J. Lopez-Ibor, & N. Andreasen (Eds.), New Oxford textbook of psychiatry (2nd ed., pp. 1337–1349). Oxford, UK: Oxford University Press.

Fosha, D. (2000). The transforming power of affect: A model for accelerated change. New York, NY: Basic Books. Frank, J. D., & Frank, J. B. (1991). Persuasion and healing: A comparative study of psychotherapy. Baltimore, MD: Johns

Hopkins University Press. Frank, K. N. (1999). Psychoanalytic participation. Hillsdale, NJ: The Analytic Press. Frederickson, J. (2013). Co-creating change: Effective dynamic therapy techniques. Kansas, MO: Seven Leaves Press. Freud, A. (1937). The ego and the mechanisms of defense. London, UK: Hogarth Press. Freud, S. (1958). Remembering, repeating and working-through (Further recommendations on the technique of psycho-

analysis II). In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 12). London, UK: Hogarth Press. (Original work published in 1914)

Freud, S. (1959). Inhibitions, symptoms and anxiety. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 20). London, UK: Hogarth Press. (Original work published in 1926)

Freud, S. (1964). New introductory lectures on psychoanalysis. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 22). London, UK: Hogarth Press. (Original work published 1933)

Freud, S. (1966). The dynamics of transference. In J. Strachey (Ed. and Trans.), The standard edition of the Alliance and Technique complete psychological works of Sigmund Freud (Vol. 12). London, UK: Hogarth Press. (Original work published 1912)

Friedlander, M. L., Sutherland, O., Sandler, S., Kortz, L., Bernardi, S., Lee, H.-H., & Drozd, A. (2011). Exploring corrective experiences in a successful case of short-term dynamic psychotherapy. Psychotherapy, 49(3), 349–363. doi:10.1037/a0023447

Fuertes, J. N., Mislowack, A., Brown, S., Gur-Arie, S., Wilkinson, S., & Gelso, C. J. (2007). Correlates of the real relationship in psychotherapy: A study of dyads. Psychotherapy Research, 17(4), 423–430. doi:10.1080/10503300600789189

Gabbard, G. O., & Westen, D. (2003). Rethinking therapeutic action. International Journal of Psychoanalysis, 84(4), 823–841 doi:10.1516/N4T0-4D5G-NNPL-H7NL

Geller, J., & Farber, B. (1993). Factors influencing the process of internalization in psychotherapy. Psychotherapy Research, 3(3), 166–180. doi:10.1080/10503309312331333769

Gershefski, J. J., Arnkoff, D. B., Glass, C. R., & Elkin, I. (1996). Clients’ perceptions of treatment for depression: I. Helpful aspects. Psychotherapy Research, 6(4), 233–247. doi:10.1080/10503309612331331768

Gilbert, P. (2009). Introducing compassion-focused therapy. Advances in Psychiatric Treatment, 15, 199–208. doi:10.1192/apt.bp.107.005264

Glaser, B.G., & Strauss, A.M. (1967). The discovery of grounded theory: Strategies for qualitative research. New York, NY: Aldine.

Gold, J., & Stricker, G. (2001). A relational psychodynamic perspective on assimilative integration. Journal of Psychotherapy Integration, 11(1), 43–58.

Gold, J., & Stricker, G. (2011). Failures in psychodynamic psychotherapy. Journal of Clinical Psychology, 67(11), 1096–105. doi:10.1002/jclp.20847

Grande, T., Rudolf, G., Oberbracht, C., & Pauli-Magnus, C. (2003). Progressive changes in patients’ lives after psychotherapy: Which treatment effects support them? Psychotherapy Research, 13(1), 43–58. doi:10.1093/ptr/kpg006

Grant, J. E., & Potenza, M. N. (2009). Young adult mental health. New York, NY: Oxford University Press. Gray, P. (1986). On helping analysands observe intrapsychic activity. In A. D. Richards & M. S. Willick (Eds.)

Psychoanalysis: The science of mental conflict (pp. 245–262). Hillsdale, NJ: Analytic Press. Greenberg, L. S., & Pascual-Leone, A. (2006). Emotion in psychotherapy: A practice-friendly research review. Journal of

Clinical Psychology, 62(5), 611–630. doi:10.1002/jclp.20252 Greenson, R. R. (1967). The technique and practice of psychoanalysis. Madison, CT: International Universities Press. Grenyer, B. F. S., & Luborsky, L. (1996). Dynamic change in psychotherapy: Mastery of interpersonal conflicts. Journal of

Consulting and Clinical Psychology, 64(2), 411–416. doi:10.1037/0022-006X.64.2.411 Grünbaum, A. (1984). The foundations of psychoanalysis: A philosophical critique. Berkeley, CA: University of California

Press.

66

Harpaz-Rotem, I., Blatt, S. J. (2005). Changes in representations of a self–designated significant other in long–term intensive inpatient treatment of seriously disturbed adolescents and young adults. Psychiatry: Interpersonal and Biological Processes, 68(3), 266-282. doi:10.1521/psyc.2005.68.3.266

Hartman, J. (2001). Grundad teori: Teorigenerering på empirisk grund. [Grounded theory: Generating theory empirically]. Lund: Studentlitteratur.

Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. doi:10.1037/0022-3514.52.3.511

Heck, R. H., Thomas, S. L., & Tabata, L. N. (2014). Multilevel and longitudinal modeling with IBM SPSS (2nd ed.). New York, NY: Routledge.

Henkelman, J., & Paulson, B. (2006). The client as expert: Researching hindering experiences in counseling. Counseling Psychology Quarterly, 19(2), 139–150. doi:10.1080/09515070600788303

Hesse, E., & Main, M. (2000). Disorganized infant, child, and adult attachment: Collapse in behavioral and attentional strategies. Journal of the American Psychoanalytic Association, 48, 1097–1127. doi:10.1177/00030651000480041101

Hill, C. E., Nutt-Williams, E., Heaton, K. J., Thompson, B. J., & Rhodes, R. H. (1996). Counselor retrospective recall of impasses in long-term psychotherapy: A qualitative analysis. Journal of Counseling Psychology, 43(2), 207–217. doi:10.1037/0022-0167.43.2.207

Hill, C. E., & Knox, S. (2009). Processing the therapeutic relationship. Psychotherapy Research, 19(1), 13–29. doi:10.1080/10503300802621206

Hilsenroth, M. J., Blagys, M. D., Ackerman, S. J., Bonge, D. R., & Blais, M. A. (2005). Measuring psychodynamic-interpersonal and cognitive-behavioral techniques: Development of the Comparative Psychotherapy Process Scale. Psychotherapy: Theory, Research, Practice, Training, 42(3), 340–356. doi:10.1037/0033-3204.42.3.340

Hjälmdahl, Y., Claesson, M., Werbart, A., & Levander, S. (2001). Bedömning av differentiering-relaterande i själv- och objektrepresentationer I: En validitetsstudie [Differentiation- relatedness of self and object relations I: Validity of the Swedish version]. Psykoterapi: Forskning och utveckling, 20. Stockholm: Institute of Psychotherapy and Psychotherapy Section at Karolinska Institutet.

Holmes, J. (2001). The search for the secure base: Attachment theory and psychotherapy. New York, NY: Routhledge. Holmes, J. (2010). Exploring in security: Towards an attachment-informed psychoanalytic psychotherapy. New York, NY:

Routhledge. Horowitz, L. M., Rosenberg, S.E., Baer, B.A., Ureno, G., & Villaseñor, V. S. (1988). Inventory of interpersonal problems:

Psychometric properties and clinical applications. Journal of Consulting and Clinical Psychology, 56(6), 885–892. doi:10.1037/ 0022-006X.56.6.885

Horowitz, L. M., Alden, L. E., Wiggins, J. S., & Pincus, A. L. (2002). Inventory of Interpersonal Problems. Swedish manual. Stockholm: Psykologiförlaget.

Horvath, A. O. (2006). The alliance in context: Accomplishments, challenges, and future directions. Psychotherapy, 43(3), 258–63. doi:10.1037/0033-3204.43.3.258

Horvath, A. O., Del Re, AC., Flückinger, C., & Symonds, D. (2011). Alliance in individual psychotherapy. In John C. Norcross (Ed.), Psychotherapy relationships that work: Evidence-based responsiveness (2nd ed.). New York, NY: Oxford University Press.

Howe, D. (1993). On being a client: Understanding the processes of counseling and psychotherapy. London, UK: Sage. Hunsley, J., Aubry, T. D., Verstervelt, C. M., & Vito, D. (1999). Comparing therapist and client perspectives on reasons for

psychotherapy termination. Psychotherapy: Theory, Research, Practice, Training, 36(4), 380–388. doi:10.1037/h0087802

Høglend, P. (1993). Transference interpretations and long-term outcome after dynamic psychotherapy of brief to moderate length. American Journal of Psychotherapy, 47(4), 494–507.

Høglend, P., Engelstad, V., Sørbye, Ø., Heyerdahl, O., & Amlo, S. (1994). The role of insight in exploratory psychodynamic psychotherapy. British Journal of Medical Psychology, 67(4), 305–317.

Høglend, P. (2003). Long-term effects of brief dynamic psychotherapy. Psychotherapy Research, 13(3), 271–292. doi:10.1093/ptr/kpg031

Høglend, P., Amlo, S., Marble, A., Bøgwald, K.-P., Sørbye, O., Sjaastad, M. C., & Heyerdahl, O. (2006). Analysis of the patient-therapist relationship in dynamic psychotherapy: an experimental study of transference interpretations. American Journal of Psychiatry, 163(10), 1739–46. doi:10.1176/appi.ajp.163.10.1739

Høglend, P., Bøgwald, K.-P., Amlo, S., Marble, A., Ulberg, R., Sjaastad, M. C., … Johansson, P. (2008). Transference interpretations in dynamic psychotherapy: Do they really yield sustained effects? American Journal of Psychiatry, 165(6), 763–71. doi:10.1176/appi.ajp.2008.07061028

67

Høglend, P., Hersoug, A. G., Bøgwald, K.-P., Amlo, S., Marble, A., Sørbye, O., … Crits-Christoph, P. (2011). Effects of transference work in the context of therapeutic alliance and quality of object relations. Journal of Consulting and Clinical Psychology, 79(5), 697–706. doi:10.1037/a0024863

Høglend, P. (2014). Exploration of the patient-therapist relationship in psychotherapy. American Journal of Psychiatry. Advance online publication. doi:10.1176/appi.ajp.2014.14010121

Jacobs, T. J. (1988). Notes on the therapeutic process: Working with the young adult. In A. Rothstein (Ed.), How does treatment help? On the modes of therapeutic action of psychoanalytic psychotherapy. (pp. 61–80). Madison, CT: International University Press.

Jacobsson, G. (2005). On the threshold of adulthood: Recurrent phenomena and developmental tasks during the period of young adulthood. Doctoral dissertation, Department of Education, Stockholm University, Sweden.

Jacobvitz, D., Curran, M., & Moller, N. (2002). Measurement of adult attachment: The place of self-report and interview methodologies. Attachment & Human Development, 4(2), 207–216. doi:10.1080/1461673021015422

Janzen, J., Fitzpatrick, M., & Drapeau, M. (2008). Processes involved in client-nominated relationship building incidents: Client attachment, attachment to therapist, and session impact. Psychotherapy, 45(3), 377–90. doi:10.1037/a0013310

Jeanneau, M., & Winzer, R. (2007). Psykodynamisk psykoterapi för unga. Utvärdering av projektet ungdomar och unga vuxna vid Ericastiftelsen [Psychodynamic psychotherapy for young people. Evaluation of the adolescent and young adult project at the Erica Foundation]. Psykisk hälsa - barn och unga. Rapport 1. Stockholm: Centrum för folkhälsa.

Johansson, P., & Høglend, P. (2007). Identifying mechanisms of change in psychotherapy: Mediators of treatment outcome. Clinical Psychology and Psychotherapy, 14(1), 1–9. doi:10.1002/cpp.514

Johansson, P., Høglend, P., Ulberg, R., Bøgwald, K-P., Amlo, S., Marble, A., … Heyerdahl, O. (2010). The mediating role of insight for long-term improvements in psychodynamic therapy. Journal of Consulting and Clinical Psychology, 78(3), 438–448. doi:10.1037/a0019245

Jones, E. E., & Pulos, S. M. (1993). Comparing the process in psychodynamic and cognitive-behavioral therapies. Journal of Consulting and Clinical Psychology, 61(2), 306–16. doi:10.1037/0022-006X.61.2.306

Kallestad, H., Valen, J., McCullough, L., Svartberg, M., Høglend, P., & Stiles, T. C. (2010). The relationship between insight gained during therapy and long-term outcome in short-term dynamic psychotherapy and cognitive therapy for cluster C personality disorders. Psychotherapy Research, 20(5), 526–34. doi:10.1080/10503307.2010.492807

Kantrowitz, J., Katz, A., & Paolitto, F. (1990). Follow-up of psychoanalysis five to ten years after termination. II. Development of the self-analytic function. Journal of the American Psychoanalytic Association, 38(3), 637–654. doi:10.1177/000306519003800305

Karlsson, R., & Kermott, A. (2006). Reflective-functioning during the process in brief psychotherapies. Psychotherapy: Theory, Research, Practice, Training, 43(1), 65–84. doi:10.1037/0033-3204.43.1.65

Katzman, J., & Coughlin, P. (2013). The role of therapist activity in psychodynamic psychotherapy. Psychodynamic Psychiatry, 41(1), 75–89. doi:10.1521/pdps.2013.41.1.75

Kazdin, A. E. (2003). Research design in clinical psychology (4th ed.). Boston, MA: Allyn & Bacon. Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical

Psychology, 3, 1–27. doi:10.1146/annurev.clinpsy.3.022806.091432 Kazdin, A. E. (2008). Evidence-based treatment and practice: New opportunities to bridge clinical research and practice,

enhance the knowledge base, and improve patient care. American Psychologist, 63(3), 146–159. doi:10.1037/0003-066X.63.3.146

Kazdin, A. E. (2009). Understanding how and why psychotherapy leads to change. Psychotherapy Research, 19(4-5), 418–28. doi:10.1080/10503300802448899

Kernberg, O. F. (1997). The nature of interpretation: intersubjectivity and the third position. American Journal of Psychoanalysis, 57(4), 297–312. doi:10.1023/A:1022463306053

Kernberg, O. F. (2007). The therapeutic action of psychoanalysis: Controversies and challenges. Psychoanalytic Quarterly, 93(2), 427–42. doi:10.1111/j.1745-8315.2011.00512.x

Kivlighan, D. M., Multon, K. D., & Patton, M. J. (2000). Insight and symptom reduction in time-limited psychoanalytic counseling. Journal of Counseling Psychology, 47(1), 50–58. doi:10.1037/0022-0167.47.1.50

Knox, S., Hess, S. A., Hill, C. E., Burkard, A. W., & Crook-Lyon, R. E. (2012). Corrective relational experiences: Client perspectives. In L. G. Castonguay & C. E. Hill (Eds.), Transformation in psychotherapy: Corrective experiences across cognitive behavioral, humanistic, and psychodynamic approaches (pp. 51–67). Washington, DC: American Psychological Association.

Kobak, R. R., Cole, H. E., Ferenz-Gillies, R., Fleming, W. S., & Gamble, W. (1993). Attachment and emotion regulation during mother-teen problem solving: a control theory analysis. Child Development, 64(1), 231–45.

68

Koelen, J. A., Houtveen, J. H., Abbass, A., Luyten, P., Eurelings-Bontekoe, E. H. M., Van Broeckhuysen-Kloth, S. A. M., … Geenen, R. (2014). Effectiveness of psychotherapy for severe somatoform disorder: meta-analysis. British Journal of Psychiatry, 204(1), 12–19. doi:10.1192/bjp.bp.112.121830

Kohut, H. (1984). How Does Analysis Cure? Chicago, IL: University of Chicago Press. Kottler, J. (1986). On being a therapist. San Francisco, CA: Jossey-Bass. Kraemer, H. C., Wilson, G. T., Fairburn, C. G., & Agras, W. S. (2002). Mediators and moderators of treatment effects in

randomized clinical trials. Archives of General Psychiatry, 59(10), 877–883. doi:10.1001/archpsyc.59.10.877 Lambert, M. J. (2011). What have we learned about treatment failure in empirically supported treatments? Some suggestions

for practice. Cognitive and Behavioral Practice, 18(3), 413–420. doi:10.1016/j.cbpra.2011.02.002 Lambert, M. J. (2013a). The efficacy and effectiveness of psychotherapy. In M. J. Lambert (Ed.), Bergin and Garfield’s

handbook of psychotherapy and behavior change (6th ed., pp. 169–218). Hoboken, NJ: John Wiley & Sons. Lambert, M. J. (Ed.). (2013b). Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed.). Hoboken,

NJ: John Wiley & Sons. Lampropoulos, G. (2001). Bridging technical eclecticism and theoretical integration: Assimilative integration. Journal of

Psychotherapy Integration, 11(1), 5–19. doi:10.1023/A:1026672807119 Leichsenring, F. (2001). Comparative effects of short-term psychodynamic psychotherapy and cognitive-behavioral therapy

in depression: A meta-analytic approach. Clinical Psychology Review, 21(3), 401–19. doi:10.1016/S0272-7358(99)00057-4

Leichsenring, F., & Leibing, E. (2003). The effectiveness of psychodynamic therapy and cognitive behavior therapy in the treatment of personality disorders: A meta-analysis. American Journal of Psychiatry, 160(7), 1223–1232. doi:10.1176/appi.ajp.160.7.1223

Leichsenring, F. (2004). Randomized controlled versus naturalistic studies: A new research agenda. Bulletin of the Menninger Clinic, 68(2), 137–51. doi:10.1521/bumc.68.2.137.35952

Leichsenring, F., & Klein, S. (2014). Evidence for psychodynamic psychotherapy in specific mental disorders: A systematic review. Psychoanalytic Psychotherapy, 28(1), 4–32. doi:10.1080/02668734.2013.865428

Leuzinger-Bohleber, M. (2002). A follow-up study – critical inspiration for our clinical practice? In M. Leuzinger-Bohleber & M. Target (Eds.), Outcomes of psychoanalytic treatments: Perspectives for therapists and researchers (pp. 143–173). London, UK: Whurr.

Levy, J. A., Glass, C. R., Arnkoff, D. B., & Gershefski, J. J. (1996). Clients’ perceptions of treatment for depression: II. Problematic or hindering aspects. Psychotherapy Research, 6(4), 249-262. doi:10.1080/10503309612331331778

Levy, K. N., Meehan, K. B., Kelly, K. M., Reynoso, J. S., Weber, M., Clarkin, J. F., & Kernberg, O. F. (2006). Change in attachment patterns and reflective function in a randomized control trial of transference-focused psychotherapy for borderline personality disorder. Journal of Consulting and Clinical Psychology, 74(6), 1027–1040. doi:10.1037/0022-006X.74.6.1027

Levy, K. N., Ellison, W. D., Scott, L. N., & Bernecker, S. L. (2011). Attachment style. Journal of Clinical Psychology, 67(2), 193–203. doi:10.1002/jclp.20756

Lilliengren, P. (2011). Patient Attachment to Therapist Rating Scale. Unpublished manual, Department of Psychology, Stockholm University, Sweden.

Lilliengren, P., Werbart, A., Mothander, P. R., Ekström, A., Sjögren, S., & Ögren, M.-L. (2014). Patient Attachment to Therapist Rating Scale: Development and psychometric properties. Psychotherapy Research, 24(2), 184–201. doi:10.1080/10503307.2013.867462

Lindgren, A., Werbart, A., & Philips, B. (2010). Long-term outcome and post-treatment effects of psychoanalytic psychotherapy with young adults. Psychology and Psychotherapy, 83(1), 27–43. doi:10.1348/147608309X464422

Linehan, M. M. (1993). Cognitive behavioral therapy of borderline personality disorder. New York, NY: Guilford Press. Liotti, G. (2007). Internal working models of attachment in the therapeutic relationship. In P. Gilbert & R. Leahy (Eds.), The

therapeutic relationship in cognitive-behavioural psychotherapies (pp. 143–162). New York, NY: Routhledge Lipsey, M. W., & Wilson, D. B. (1993). The efficacy of psychological, educational and behavioural treatment. American

Psychologist, 48(12), 1181–1209. Ljótsson, B., Hesser, H., Andersson, E., Lindfors, P., Hursti, T., Rück, C., … Hedman, E. (2013). Mechanisms of change in

an exposure-based treatment for irritable bowel syndrome. Journal of Consulting and Clinical Psychology, 81(6), 1113–11126. doi:10.1037/a0033439

Loewald, H. W. (1960). On the therapeutic action of psychoanalysis. International Journal of Psychoanalysis, 41, 16–33. Loewald, H. (1962). Internalization, separation, mourning, and the superego. Psychoanalytic Quarterly, 31(4), 483–504. Luborsky, L. (1976). Helping alliances in psychotherapy: The groundwork for a study of their relationship to its outcome. In

J. L. Claghom (Ed.), Successful Psychotherapy (pp. 92–116). New York, NY: Brunner/Mazel.

69

Luborsky, L., Barber, J. P., Siqueland, L., Johnson, S., Najavits, L. M., Frank, A. (1996). The revised Helping Alliance Questionnaire (HAq-II). Journal of Psychotherapy Practice and Research, 5(3), 260–270.

Lunsford, T. P. (2010). Attachment to therapist, the working alliance, and emotional processing of traumatic material in session among veterans diagnosed with Post-Traumatic Stress Disorder. Doctoral dissertation, University of Missouri-Colombia, Columbia, MO, USA.

Lutz, W., & Knox, S. (Eds.). (2014). Quantitative and qualitative methods in psychotherapy research. New York, NY: Routledge

Luyten, P., Blatt, S. J., & Corveleyn, J. (2006). Minding the gap between positivism and hermeneutics in psychoanalytic research. Journal of the American Psychoanalytic Association, 54(2), 571–610. doi:10.1177/00030651060540021301

Luyten, P., & Blatt, S. J. (2013). Interpersonal relatedness and self-definition in normal and disrupted personality development: Retrospect and prospect. American Psychologist, 68(3), 172–183. doi:10.1037/a0032243

Maier, M., Bernier, A., Pekrun, R., Grossmann, K., & Zimmermann, P. (2004). Attachment working models as unconscious structures: An experimental test. International Journal of Behavioral Development, 28(2), 180–189. doi:10.1080/01650250344000398

Malan, D. H. (1979). Individual psychotherapy and the science of psychodynamics. London, UK: Butterworths. Mallinckrodt, B., Gantt, D. L., & Coble, H. M. (1995). Attachment patterns in the psychotherapy relationship: Development

of the Client Attachment to Therapist Scale. Journal of Counseling Psychology, 42(3), 307–317. doi:10.1037//0022-0167.42.3.307

Mallinckrodt, B., Coble, H. M., & Gantt, D. L. (1995). Toward differentiating client attachment from working alliance and transference: Reply to Robbins (1995). Journal of Counseling Psychology, 42(3), 320–322. doi:10.1037//0022-0167.42.3.320

Mallinckrodt, B., Porter, M. J., & Kivlighan, D. M. (2005). Client attachment to therapist, depth of in-session exploration, and object relations in brief psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 42(1), 85–100. doi:10.1037/0033-3204.42.1.85

Mallinckrodt, B. (2010). The psychotherapy relationship as attachment: Evidence and implications. Journal of Social and Personal Relationships, 27(2). 262-270. doi:10.1177/0265407509360905

Malterud, K. (2001). Qualitative research: Standards, challenges, and guidelines. The Lancet, 358(9280), 483–488. doi:10.1016/S0140-6736(01)05627-6

Martin, P., & Stayer, M. A. (1990). The experience of micro- and macroevents: A life span analysis. Research on Aging, 12(3), 294–310. doi:10.1177/0164027590123002

Maunder, R. G., & Hunter, J. J. (2012). A prototype-based model of adult attachment for clinicians. Psychodynamic Psychiatry, 40(4), 549–573. doi:10.1521/pdps.2012.40.4.549

Mayes, L. C., & Spencer, D. P. (1994). Understanding therapeutic action in the analytic situation: A second look at the developmental metaphor. Journal of the American Psychoanalytic Association, 42(3), 789-817. doi:10.1177/000306519404200306

McGorry, P. D., Purcell, R., Goldstone, S. G., & Amminger, P. (2011). Age of onset and timing of treatment for mental and substance use disorders: Implications for preventive intervention strategies and models of care. Current Opinion in Psychiatry, 24, 301–306. doi:10.1097/YCO.0b013e3283477a09

McLeod, J. (1990). The practitioner’s experience of counseling and psychotherapy: A review of the research literature. In D. Mearns & W. Dryden, (Eds.), Experiences of counseling in action (pp. 66–79). London, UK: Sage.

McLeod, J. (2013). Qualitative research: Methods and contributions. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed., pp. 49–84). Hoboken, NJ: John Wiley & Sons.

McMurran, M., Huband, N., & Overton, E. (2010). Non-completion of personality disorder treatments: A systematic review of correlates, consequences, and interventions. Clinical Psychology Review, 30(3), 277–287. doi:10.1016/j.cpr.2009.12.002

Messer, S. B., & Winokur, M. (1984). Ways of knowing and visions of reality in psychoanalytic therapy and behavior therapy. In S. B. Messer & H. Arkowitz (Eds.), Psychoanalytic therapy and behavior therapy: Is integration possible? (pp. 63–100). New York, NY: Plenum.

Messer, S. B. (2001). Introduction to the special issue on assimilative integration. Journal of Psychotherapy Integration, 11(1), 1–4. doi:10.1023/A:1026619423048

Messer, S. B., & Wampold, B. E. (2002). Let’s face facts: Common factors are more potent than specific therapy ingredients. Clinical Psychology: Science and Practice, 9(1), 21–25. doi:10.1093/clipsy.9.1.21

Messer, S. B., & McWilliams, N. (2007). Insight in psychodynamic therapy: Theory and assessment. In L. G. Castonguay & C. Hill (Eds.), Insight in psychotherapy (pp. 9–29). Washington, DC: American Psychological Association.

70

Mikulincer, M., Shaver, P. R., & Pereg, D. (2003). Attachment theory and affect regulation  : The dynamics, development, and cognitive consequences of attachment-related strategies. Motivation and Emotion, 27(2), 77–102. doi:10.1023/A:1024515519160

Mikulincer, M., & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamics, and change. New York, NY: Guilford Press.

Mintz, J., Auerbach, A. H., Luborsky, L., & Johnson, M. (1973). Patient’s, therapist’s and observers’ view s of psychotherapy: A ‘Rashomon’ experience or a reasonable consensus? British Journal of Medical Psychology, 46, 83–89.

Mohr, D.C., (1995). Negative outcome in psychotherapy: A critical review. Clinical Psychology: Science and Practice 2(1), 1–27. doi:10.1111/j.1468-2850.1995.tb00022.x

Moore, S. R., & Gelso, C. J. (2011). Recollections of a secure base in psychotherapy: Considerations of the real relationship. Psychotherapy, 48(4), 368–73. doi:10.1037/a0022421

Muller, R. T. (2010). Trauma and the avoidant client: Attachment-based strategies for healing. New York, NY: Norton. Muran, C. J., & Barber, J. P. (Eds.). (2010). The therapeutic alliance: An evidence-based guide to practice. New York, NY:

The Guilford Press. Najavits, L. M. (1997). Therapists’ implicit theories of psychotherapy. Journal of Psychotherapy Integration, 7(1), 1–16. Nilsson, T., Svensson, M., Sandell, R., & Clinton, D. (2007). Patients’ experiences of change in cognitive-behavioral therapy

and psychodynamic therapy: A qualitative comparative study. Psychotherapy Research, 17(5), 553–566. doi:10.1080/10503300601139988

Norcross, J. C. (Ed.). (2011). Psychotherapy relationships that work. New York, NY: Oxford University Press Obegi, J. H. (2008). The development of the client-therapist bond through the lens of attachment theory. Psychotherapy:

Theory, Research, Training, Practice, 45(4), 431–446. doi:10.1037/a0014330 Obegi J. H., & Berant, E. (Eds.). (2009). Attachment theory and research in clinical work with adults. New York, NY: The

Guilford Press. Ogrodniczuk, J. S., Piper, W. E., Joyce, A. S., & McCallum, M. (1999). Transference interpretations in short-term dynamic

psychotherapy. Journal of Nervous and Mental Disease, 187(9), 571–578. Orlinsky, D. E., Rønnerstad, M. H., & Wilutzki, U. (2004). Fifty years of psychotherapy process-outcome research:

Continuity and change. In M. Lambert (Ed.), Bergin and Garfield´s handbook of psychotherapy and behavior change (5th ed., pp. 307–389). New York, NY: Wiley.

Ottosson, H., Bodlund, O., Ekselius, L., von Knorring, L., Kullgren, G., … Lindström, E. (1995). The DSM-IV and ICD-10 personality questionnaire (DIP-Q): Construction and preliminary validation. Nordic Journal of Psychiatry, 49(4), 285–291. doi:10.3109/08039489509011918

Ottosson, H., Bodlund, O., Ekselius, L., Grann, M., von Knorring, L., … Kullgren, G. (1998). DSM-IV and ICD-10 personality disorders: A comparison of a self-report questionnaire (DIP-Q) with a structured interview. European Psychiatry, 13(5), 246–253. doi:10.1016/S0924-9338(98)80013-8

Pandey, S., & Elliott, W. (2010). Suppressor variables in social work research: Ways to identify in multiple regression models. Journal of the Society for Social Work and Research, 1(1), 28–40. doi:10.5243/jsswr.2010.2

Parish, M., & Eagle, M. N. (2003a). A new measure of components of attachment. Unpublished manuscript, Derner Institute of Advanced Psychological Studies, Adelphi University, Garden City, NY, USA.

Parish, M., & Eagle, M. N. (2003b). Attachment to the therapist. Psychoanalytic Psychology, 20(2), 271–286. doi:10.1037/0736-9735.20.2.271

Pearl, E. (2008). Psychotherapy with adolescent girls and young women: Fostering autonomy through attachment. New York, NY: Guilford Press.

Perelberg, R. J. (1993). The psychoanalytic treatment of young adults as a rite of passage: Discussion of the conference. Bulletin of the Anna Freud Centre, 16, 95–103.

Petrowski, K., Nowacki, K., Pokorny, D., & Buchheim, A. (2011). Matching the patient to the therapist: the roles of the attachment status and the helping alliance. Journal of Nervous and Mental Disease, 199(11), 839–44. doi:10.1097/NMD.0b013e3182349cce

Petrowski, K., Pokorny, D., Nowacki, K., & Buchheim, A. (2013). The therapist’s attachment representation and the patient’s attachment to the therapist. Psychotherapy Research, 23(1), 25–34. doi:10.1080/10503307.2012.717307

Philips, B., Wennberg, P., Werbart, A., & Schubert, J. (2006). Young adults in psychoanalytic psychotherapy: Patient characteristics and therapy outcome. Psychology and Psychotherapy, 79(1), 89–106. doi:10.1348/147608305X52649

Philips, B., Werbart, A., & Schubert, J. (2005). Private theories and psychotherapeutic technique. Psychoanalytic Psychotherapy, 19(1), 48–70. doi:10.1080/02668730512331341573

71

Philips, B., Wennberg, P., & Werbart, A. (2007). Ideas of cure as a predictor of premature termination, early alliance and outcome in psychoanalytic psychotherapy. Psychology and Psychotherapy, 80(2), 229–245. doi:10.1348/147608306X128266

Pilkonis, P.A. (1988). Personality prototypes among depressives: Themes of dependency and autonomy. Journal of Personality Disorders, 2(2), 144−152. doi:10.1521/pedi.1988.2.2.144

Piper, W. E., Azim, H. F., Joyce, A. S., & McCallum, M. (1991). Transference interpretations, therapeutic alliance, and outcome in short-term individual psychotherapy. Archives of General Psychiatry, 48(10), 946–953. doi:10.1001/archpsyc.1991.01810340078010.

Popper, K. R. (1959). The logic of scientific discovery. London, UK: Hutchinson. Quintana, S. M., & Meara, N. M. (1990). Internalization of therapeutic relationships in short-term psychotherapy. Journal of

Counseling Psychology, 37(2), 123–130. doi:10.1037//0022-0167.37.2.123 Ravitz, P., Maunder, R., Hunter, J., Sthankiya, B., & Lancee, W. (2010). Adult attachment measures: A 25-year review.

Journal of Psychosomatic Research, 69(4), 419–432. doi:10.1016/j.jpsychores.2009.08.006 Regan, A., & Hill, C. (1992). Investigation of what clients and counselors do not say in brief psychotherapy. Journal of

Counseling Psychology, 39(2), 168–174. doi:10.1037/0022-0167.39.2.168 Rennie, D. L. (1994). Clients’ deference in psychotherapy. Journal of Counseling Psychology, 41(4), 427–437.

doi:10.1037/0022-0167.41.4.427 Rennie, D. L. (1998). Grounded theory methodology: The pressing need for a coherent logic of justification. Theory &

Psychology, 8(1), 101–119. doi:10.1177/0959354398081006 Rennie, D. L. (2002). Experiencing psychotherapy: Grounded theory studies. In R. C. Page, J. F. Weiss, & G. Lietaer (Eds.),

Humanistic psychotherapies: Handbook of research and practice (pp. 117–144). Washington, DC: American Psychological Association.

Rennie, D. L. (2006). Embodied categorizing in the Grounded Theory method: Methodical hermeneutics in action. Theory & Psychology, 16(4), 483–503. doi:10.1177/0959354306066202

Robbins, A., & Wilner, A. (2001). Quarterlife crisis: The unique challenges of life in your twenties. New York, NY: Tarcher/Putnam

Rogers, C. R. (1951). Client-centered therapy: Its current practice, implications, and theory. Boston, MA: Houghton Mifflin Romano, V., Fitzpatrick, M., & Janzen, J. (2008). The secure-base hypothesis: Global attachment, attachment to counselor,

and session exploration in psychotherapy. Journal of Counseling Psychology, 55(4), 495–504. doi:10.1037/a0013721

Roos, J., & Werbart, A. (2013). Therapist and relationship factors influencing dropout from individual psychotherapy: A literature review. Psychotherapy Research, 23(4), 394–418. doi:10.1080/10503307.2013.775528

Rudden, M., Milrod, B., Target, M., Ackerman, S., & Graf, E. (2006). Reflective functioning in panic disorder patients: A pilot study. Journal of the American Psychoanalytic Association, 54(4), 1339–1343. doi:10.1177/00030651060540040109

Ryum, T., Stiles, T. C., Svartberg, M., & McCullough, L. (2010). The role of transference work, the therapeutic alliance, and their interaction in reducing interpersonal problems among psychotherapy patients with Cluster C personality disorders. Psychotherapy, 47(4), 442–453. doi:10.1037/a0021183

Safran, J.D., & Messer, S. B. (1997). Psychotherapy integration: A postmodern critique. Clinical Psychology: Science and Practice, 4, 140–152. doi:10.1111/j.1468-2850.1997.tb00106.x

Safran, J. D., & Muran, J. C. (2000). Negotiating the therapeutic alliance: A relational treatment guide. New York, NY: Guilford Press.

Safran, J. D., & Muran, J. C. (2006). Has the concept of the therapeutic alliance outlived its usefulness? Psychotherapy: Theory, Research, Practice, Training, 43(3), 286–291. doi:10.1037/0033-3204.43.3.286

Safran, J. D., Muran, J. C., & Eubanks-Carter, C. (2011). Repairing alliance ruptures. Psychotherapy, 48(1), 80–87. doi:10.1037/a0022140

Sandler, J. (1983). Reflections on some relations between psychoanalytic concepts and psychoanalytic practice. International Journal of Psychoanalysis, 64(1), 35–45.

Satran, G. (1995). The patient’s sense of therapeutic action: An introduction. Contemporary Psychoanalysis, 31(1), 124–132. doi:10.1080/00107530.1995.10746900

Sauer, E. M., Anderson, M. Z., Gormley, B., Richmond, C. J., & Preacco, L. (2010). Client attachment orientations, working alliances, and responses to therapy: A psychology training clinic study. Psychotherapy Research, 20(6), 702–711. doi:10.1080/10503307.2010.518635

Saypol, E., & Farber, B. A. (2010). Attachment style and patient disclosure in psychotherapy. Psychotherapy Research, 20(4), 462–471. doi:10.1080/10503301003796821

72

Schauenburg, H., Buchheim, A., Beckh, K., Nolte, T., Brenk-Franz, K., Leichsenring, F., … Dinger, U. (2010). The influence of psychodynamically oriented therapists’ attachment representations on outcome and alliance in inpatient psychotherapy. Psychotherapy Research, 20(2), 193–202. doi:10.1080/10503300903204043

Schut, A. J., Castonguay, L. G., Flanagan, K. M., Yamasaki, A. S., Barber, J. P., Bedics, J.D., & Smith, T. L. (2005). Therapist interpretation, patient-therapist interpersonal process, and outcome in psychodynamic psychotherapy for avoidant personality disorder. Psychotherapy: Theory, Research, Practice, Training, 42(4), 494–511. doi:10.1037/0033-3204.42.4.494

Schön, D. (1983). The reflective practitioner: How professionals think in action. New York, NY: Basic Books. Seligman, M. E. P. (1995). The effectiveness of psychotherapy: The Consumer Reports study. American Psychologist,

50(12), 965-974. doi:10.1037/0003-066X.50.12.965 Shaver, P. R., & Mikulincer, M. (2009) An overview of adult attachment theory. In J. H. Obegi & E. Berant (Eds.),

Attachment theory and research in clinical work with adults (pp. 46-70). New York, NY: The Guilford Press. Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2),

420–428. doi:10.1037/0033-2909.86.2.420 Slade, A. (2008). The implications of attachment theory and research for adult psychotherapy: Research and clinical

perspectives. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 762–782). New York, NY: Guilford Press.

Smith, M. L., Glass, G. V., & Miller, T. I. (1980). The benefits of psychotherapy. Baltimore, MD: Johns Hopkins University Press.

Smith, A. E. M., Msetfi, R. M., & Golding, L. (2010). Client self rated adult attachment patterns and the therapeutic alliance: a systematic review. Clinical Psychology Review, 30(3), 326–337. doi:10.1016/j.cpr.2009.12.007

Socialstyrelsen. (2013a). Folkhälsan i Sverige 2013 [Public health in Sweden 2013]. Stockholm: Statens Folkhälsoinstitut Socialstyrelsen. (2013b). Psykisk ohälsa bland unga [Mental illness among young people]. Stockholm: National Board of

Health and Wellfare Sterba, R. F. (1934). The fate of the ego in analytic therapy. International Journal of Psychoanalysis, 15, 117–126. Stern, D. N., Sander, L. W., Nahum, J. P., Harrison, A. M., Lyons-Ruth, K., … Tronick, E. Z. (1998). Non-interpretive

mechanisms in psychoanalytic therapy: The ‘something more’ than interpretation. International Journal of Psychoanalysis, 79(5), 903–921.

Stoffers, J. M., Völlm, B. A., Rücker, G., Timmer, A., Huband, N., & Lieb, K. (2013). Psychological therapies for people with borderline personality disorder. Cochrane Library, 2, 1–255. doi:10.1002/14651858.CD005652.pub2

Strachey, J. (1934), The nature of the therapeutic action of psychoanalysis. International Journal of psychoanalysis, 15, 127–159.

Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). London, UK: Sage.

Strauss, B., Lobo-Drost, A., & Pilkonis, P. A. (1999). Einschätzung von Bindungsstilen bei Erwachsenen [Assessment of adult attachment styles]. Zeitschrift für Klinische Psychologie, Psychiatrie und Psychotherapie, 47, 347–364.

Strupp, H., Fox, R., & Lessler K. (1969). Patients view their psychotherapy. Baltimore, MD: John Hopkins University Press. Strunk, D. R., Brotman, M. A., & DeRubeis, R. J. (2010). The process of change in cognitive therapy for depression:

Predictors of early inter-session symptom gains. Behavior Research and Therapy, 48(7), 599–606. doi:10.1016/j.brat.2010.03.011

Strunk, D. R., Cooper, A. A., Ryan, E. T., DeRubeis, R. J., & Hollon, S. D. (2012). The process of change in cognitive therapy for depression when combined with antidepressant medication: Predictors of early intersession symptom gains. Journal of Consulting and Clinical Psychology, 80(5), 730–738. doi:10.1037/a0029281

Swift, J. K., & Greenberg, R. P. (2012). Premature discontinuation in adult psychotherapy: A meta-analysis. Journal of Consulting and Clinical Psychology, 80(4), 547–559. doi:10.1037/a0028226

Szajnberg, N.M., & Massie, H. (2003). Transition to young adulthood: A prospective study. International Journal of Psychoanalysis, 84(6), 1569–1586. doi: 10.1516/R70A-KVKF-9R16-0UA0

Tabachnick, B., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). New York, NY: Pearson. Talia, A., Daniel, S. I. F., Miller-Bottome, M., Brambilla, D., Miccoli, D., Safran, J. D., & Lingiardi, V. (2014). AAI

predicts patients’ in-session interpersonal behavior and discourse: a “move to the level of the relation” for attachment-informed psychotherapy research. Attachment & Human Development, 16(2), 192–209. doi:10.1080/14616734.2013.859161

Tanner, L., & Arnett, J. J. (2009). The emergance of ”emerging adulthood”. In A. Furlong (Ed.), Handbook of youth and young adulthood (pp. 39-45). New York, NY: Routhledge.

73

Tasca, G. A., & Lampard, A. M. (2012). Reciprocal influence of alliance to the group and outcome in day treatment for eating disorders. Journal of Counseling Psychology, 59(4), 507–517. doi:10.1037/a0029947

Timulak, L. (2007). Identifying core categories of client-identified impact of helpful events in psychotherapy: A qualitative meta-analysis. Psychotherapy Research, 17(3), 305–314. doi:10.1080/10503300600608116

Timulak, L. (2010). Significant events in psychotherapy: An update of research findings. Psychology and Psychotherapy: Theory, Research and Practice, 83(4), 421–447. doi:10.1348/147608310X499404

Ulvenes, P. G., Berggraf, L., Hoffart, A., Stiles, T. C., Svartberg, M., McCullough, L., & Wampold, B. E. (2012). Different processes for different therapies: Therapist actions, therapeutic bond, and outcome. Psychotherapy, 49(3), 291–302. doi:10.1037/a0027895

Vermote, R., Lowyck, B., Luyten, P., Vertommen, H., Corveleyn, J., Verhaest, Y., … Peuskens, J. (2010). Process and outcome in psychodynamic hospitalization-based treatment for patients with a personality disorder. Journal of Nervous and Mental Disease, 198(2), 110–115. doi:10.1097/NMD.0b013e3181cc0d59

von Below, C., & Werbart, A. (2012). Dissatisfied psychotherapy patients: A tentative conceptual model grounded in the participants’ view. Psychoanalytic Psychotherapy, 26(3), 211–229. doi:10.1080/02668734.2012.709536

Wachtel, P. L., & Messer, S. B. (Eds.). (1997). Theories of psychotherapy: Origins and evolution. Washington, DC: American Psychological Association.

Wachtel, P. L. (2014). Cyclical psychodynamics and the contextual self: The inner world, the intimate world, and the world of culture and society. New York, NY: Routhledge.

Wallin, D. J. (2007). Attachment in psychotherapy. New York, NY: Guilford Press. Wampold, B. E., & Serlin, R. C. (2000). The consequence of ignoring a nested factor on measures of effect size in analysis of

variance. Psychological Methods, 5(4), 425–433. doi:10.1037/1082-989X.5.4.425 Wampold, B. E. (2001). The great psychotherapy debate: Models, methods and findings. Mahwah, NJ: Lawrence Erlbaum. Wampold, B. E., Minami, T., Baskin, T. W., & Callen Tierney, S. (2002). A meta-(re)analysis of the effects of cognitive

therapy versus “other therapies” for depression. Journal of Affective Disorders, 68(2-3), 159–65. doi:10.1016/S0165-0327(00)00287-1

Wampold, B. E., & Budge, S. L. (2012). The 2011 Leona Tyler award address: The relationship--and its relationship to the common and specific factors of psychotherapy. The Counseling Psychologist, 40(4), 601–623. doi:10.1177/0011000011432709

Weinryb, R. M., Gustavsson, J. P., Hellström, C., Andersson, E., Broberg, A., & Rylander, G. (1996). Interpersonal problems and personality characteristics: Psychometric studies of the Swedish version of the IIP. Personality and Individual Differences, 20(1), 13–23. doi:10.1016/0191-8869(95)00137-U

Werbart, A., & Levander, S. (2006). Two sets of private theories in analysands and their analysts: Utopian versus attainable cures. Psychoanalytic Psychology, 23(1), 108–127. doi:10.1037/0736-9735.23.1.108

Werbart, A., & Levander, S. (2011). Vicissitudes of ideas of cure in analysands and their analysts: A longitudinal interview study. International Journal of Psychoanalysis, 92(6), 1455–1481. doi:10.1111/j.1745-8315.2011.00485.x

Werbart, A., Forsström, D., & Jeanneau, M. (2012). Long-term outcomes of psychodynamic residential treatment for severely disturbed young adults: A naturalistic study at a Swedish therapeutic community. Nordic Journal of Psychiatry, 66(6), 367–75. doi:10.3109/08039488.2012.654508

Whelton, W. J. (2004). Emotional processes in psychotherapy: evidence across therapeutic modalities. Clinical Psychology & Psychotherapy, 11(1), 58–71. doi:10.1002/cpp.392

Wiman, M., & Werbart, A. (2002). Unga vuxna i psykoterapi II: Hur uppfattar de själva sina problem? [Young adults in psychotherapy II: How do they perceive their problems?]. Psykoterapi: Forskning och utveckling, 23. Stockholm: Institute of Psychotherapy and Psychotherapy Section at Karolinska Institutet.

Wiseman, H., & Tishby, O. (2014). Client attachment, attachment to the therapist and client-therapist attachment match: How do they relate to change in psychodynamic psychotherapy? Psychotherapy Research, 24(3), 392–406. doi:10.1080/10503307.2014.892646

Woodhouse, S. S., Schlosser, L. Z., Crook, R. E., Ligiéro, D. P., & Gelso, C. J. (2003). Client attachment to therapist: Relations to transference and client recollections of parental caregiving. Journal of Counseling Psychology, 50(4), 395–408. doi:10.1037/0022-0167.50.4.395

Young, J. E., Klosko, J., Weishaar, M. E. (2003). Schema therapy: A practioner’s guide. New York, NY: Guilford Press. Zetzel, E. (1956). Current concepts of transference. International Journal of Psychoanalysis, 37, 369–375. Zilcha-Mano, S., Dinger, U., McCarthy, K. S., & Barber, J. P. (2013). Does alliance predict symptoms throughout treatment,

or is it the other way around? Journal of Consulting and Clinical Psychology. Advance online publication. doi:10.1037/a0035141

Exploring therapeutic action in psychoanalytic psychotherapy753745/FULLTEXT02.pdf · explore...

Documents

Transcript of Exploring therapeutic action in psychoanalytic psychotherapy753745/FULLTEXT02.pdf · explore...