Investigating couple relationships and change during couple (PDF)

Investigating couple relationships and change during couple and family therapy   Rune Zahl-Olsen Investigating couple relationships and change during couple and family therapy: A specific focus on domestic violence and work functioning Dissertation for the degree of philosophiae doctor (ph.d.) University of Agder Faculty of Health and Sport Sciences Sørlandet Hospital Clinic for Mental Health Department of Child and Adolescent Mental Health 2020  Doctoral dissertation at University of Agder 266 ISSN: 1504-9272 ISBN: 978-82-7117-967-0  Rune Zahl-Olsen, 2020 Print: 07 Media, Kristiansand  V Summary The aims of this thesis were to investigate couple relationships and change in those relationships during couple and family therapy, with a specific focus on domestic violence and work functioning. The first subsidiary aim was to assess, theoretically and empirically, what level of simplification should be applied to the comprehensive measure used in this study. The second subsidiary aim was to examine the physical couple and family violence among clients receiving couple and family therapy. The final subsidiary aim was to analyze the clients’ change in work functioning from pre- to post- treatment. The clients who participated in this study (N=841) can be considered as representative of clients in couple and family therapy in Norway when considering income, the number of children, educational level, prior experience with therapy, and levels of distress on individual and relational measures. Domestic violence is commonly mentioned in the news because the consequences sometimes are fatal. However, research on domestic violence on clinical samples is sparse, and it has often failed to differentiate between emotional and physical violence. Furthermore, violence between others in the family besides the couple has, to the best of my knowledge, not been included. Outcome studies of couple and family therapy most often focus on relational functioning or individual distress. However, work functioning has, to the best of my knowledge, not been used as an outcome measure. Paper 1 investigated the level of simplification that should be applied to the comprehensive measure used in this study, the Systemic Therapy Inventory of Change (STIC). Since the use of standardized measures implies simplification of what can be captured of human experiences, the challenge of such use is to ensure that the information collected represents the human experience as close as possible to reality. However, the measure should be feasible when considering the time and energy spent in answering and examining the results. Therapists and researchers need simplified information to get an overview of the complex lives of the clients, but if the overview is too simplistic, it might mislead more than inform. The question is, therefore, how much simplification can be conducted before the measures no longer inform us as therapists and researchers in a  VI credible way about the clients’ levels of distress and the process of change in therapy? Paper 1 investigated what level of simplification could be conducted on one of the scales of the comprehensive self-assessment questionnaire, STIC. We found that the scale could not be simplified beyond the sub-scales model and still be clinically and scientifically useful to understand the clients and the possible change in their lives. The conclusion was that even though it is possible to simplify conceptually different items into one total scale, this approach should not be taken when evaluating problems or changes in therapy. The reason is that the simplification could increase the risk of misunderstanding clients and their therapeutic experience. In this thesis, I discuss the result from the analysis concerning measuring changes, in general, and specifically in couple and family therapy when using comprehensive measures. Paper 2 examined physical couple and family violence among clients seeking therapy with a specific focus on identifiers and predictors of the violence exerted. In total, 25 % of our sample, derived from a clinical setting, responded that they had experienced physical family violence, on a range from “sometimes” to “all of the time.” Family violence was detected in 50 % of the cases with violent couples, compared to 18 % in the cases with non-violent couples. A model for predicting physical violence was tested. The model explained 53 % of the physical violence, and the strongest predictor was found to be the expectation level towards the partner, specifically regarding household chores. This fact means that the more the clients experienced their partners expecting too much of them, the more physical violence they reported. The more they experienced being filled with anger towards their partner, the more physical violence they reported happening. Self-control was found to be a negative predictor of physical violence, meaning that more self-control was associated with less physical violence. Contrary to our expectations, the more sexually satisfied they reported being in the relationship, the more physical violence they experienced. Paper 3 analyzed the change in work functioning from pre- to post-treatment, concerning distress on three dimensions. In this study, the clients reported their individual level of distress as well as levels of distress on two relational dimensions. At the individual level, we measured the level of depression with the Becks Depression Index (BDI). The distress in their couple relationship was measured with the Revised Dyadic Adjustment Scale (RDAS), and family functioning was measured with the Family Assessment Device (FAD). The group  VII of clients improved from start to end of treatment on work functioning as well as on the individual-, couple- and family measures. The level of work functioning was significantly better predicted when the relational measures were included, compared to when only the individual measure was used. None of the levels of these measures (i.e., individual, couple, or family measures) at pre-treatment could predict the work functioning at post-treatment. However, the change in these measures from pre- to post-treatment could predict 54 % of the work functioning at post-treatment when controlling for the level of work functioning at pre-treatment. List of papers 1. Zahl-Olsen, R., Gausel, N., Håland, Å.T., Tilden, T. (In review). Monitoring therapeutic change through diversity or simplicity? A conservative, critical Confirmatory Factor Analysis test of a Routine Outcome Monitoring system. 2. Zahl-Olsen, R., Gausel, N., Zahl-Olsen A., Bjerregaard Bertelsen T., Håland Å.T., Tilden T. (2019). Physical couple and family violence among clients seeking therapy: identifiers and predictors. Frontiers in Psychology, section Psychology for Clinical Settings. 3. Zahl-Olsen, R., Håland, Å.T., Gausel, N., Wampold, B., Tilden, T. (2019). Change in work functioning from pre- to post-treatment in feedback- informed couple and family therapy in Norway. Journal of Family Therapy.  VIII  IX Acknowledgments It is hard to define an actual starting point of this thesis, but I recall when PhD Åshild Tellefsen Håland, one of the supervisors, came to my office just after I had finished a therapy session, asking if I wanted to participate in a research project. I was skeptical about the quantitative approach and the online feedback system that we were about to try out in clinical practice. Initially, I did not think it could gain any useful information for the therapy my colleagues and I performed. However, Åshild convinced me that it would be interesting, even fun, to be a part of the research project. As time passed by, I found the data we collected to be interesting and even helpful in therapy. At the same time, I slowly understood more and more of the research aspects of the project and started to ask questions about the data. Together with some people from the local team consisting of therapists and researchers, I twice visited our collaborators in the US and participated in discussions about the study design and how to understand the preliminary results. I wrote several drafts of research projects before I finally submitted and got a grant from Sørlandet Kompetansefond for the present project. I wish to express my most sincere gratitude to my main supervisor, Terje Tilden. He was the initiator of the STIC project in Norway, is still leading the project, and has tirelessly answered emails, read drafts and made comments, discussed on Skype and telephone, and invited me to the research department at Modum Bad several times to keep the momentum in this study. His patience, thorough comments on drafts, and supportive approach have been invaluable. I also want to express my gratitude to my co-supervisors, Åshild Tellefsen Håland and Nicolay Gausel. Tellefsen Håland has, for the last two years, been the leader of the research department at the department for Child and Adolescent Mental Health (ABUP), Sørlandet Hospital, where I am working. She has been supporting this project in several ways, as part of the steering group of STIC in Norway, local administrator for the first years of the project, and one of the co- supervisors for this project. She has provided, as my leader, the needed support but also been the one following the project on a day to day basis, always available for questions and suggestions on how to move on. Nicolay Gausel has brought statistical knowledge and feedback from an adjacent field that has lifted the discussions and improved the project and the papers in a  X way not possible without his participation. At critical time points, he invited me to his office to do analyses under his supervision. Please accept my deepest gratitude. I also want to credit all the coauthors for their important contributions to the papers. Thanks to Sørlandets Kompetansefond, Sparebanken Sør, and department of child and adolescent mental health (ABUP) that has financially supported this project. Thanks to my family, Agnes, Marius, Jonas and Julianne, for your support and understanding during these years. Finally, I am deeply grateful to God who gave me life, joy, energy and perseverance to fulfill this project.  XI Table of contents Summary................................................................................................................V List of papers .................................................................................................. VII Acknowledgments ............................................................................................IX Table of contents ..................................................................................................XI 1 Introduction .........................................................................................................1 1.1 Aims of this thesis.........................................................................................1 1.2 Definition and history of Couple and Family Therapy.................................1 1.2.1 Definition of CFT...................................................................................1 1.2.2 Historical background for CFT..............................................................2 1.2.3 The philosophy of science within CFT..................................................3 1.3 Developments and current state of CFT research.........................................6 1.4 Dilemmas when using questionnaires to measure change............................8 1.5 Should the focus be individual, relational, or societal?................................9 1.5.1 Relational focus – Domestic violence .....................................................10 1.5.2 Societal focus – Work functioning ..........................................................11 2 How to measure change in CFT? ......................................................................13 2.1 What to measure?........................................................................................13 2.1.1 Selecting focus: Individual, relational, or societal? .............................15 2.2 Who should measure levels of distress and change?..................................16 2.2.1 Single-informant strategy.....................................................................16 2.2.2 Benefits and challenges of the multi-informant strategy.....................16 2.3 When to measure change? ..........................................................................17 2.3.1 Routine outcome monitoring of clients’ subjective experiences .........18 2.3.2 Systemic Therapy Inventory of Change (STIC) ..................................19 2.3.3 Level of simplification in measures.....................................................22 2.4 How to measure change? ............................................................................23 2.4.1 Change defined by whom?...................................................................23 2.4.2 Different measuring levels of change ..................................................24 2.4.3 What is good evidence of treatment effect?.........................................25 2.4.4 How to assess if a change is clinically significant? .............................27 2.4.5 The possibility of arbitrary results when categorizing outcome..........29 2.4.6 Considerations regarding sub-samples when measuring change in CFT .......................................................................................................................30  XII 2.4.7 How to handle that a client’s understanding of a concept might change from pre to post-treatment?...........................................................................32 2.5 Decisions on what, who, when and how in this study................................33 3 Domestic violence .............................................................................................35 3.1 Definitions of violence................................................................................35 3.1.1 Types of domestic violence..................................................................36 3.2 Theories of domestic violence ....................................................................37 3.3 Consequences of domestic violence ...........................................................38 3.4 Prevalence of couple violence in the population in general and in clinical samples..............................................................................................................39 3.4.1 Prevalence of domestic violence in Norway........................................40 3.4.2 Change in the understanding of domestic violence .............................41 3.5 Recommended treatments for domestic violence.......................................43 3.6 Uncovering violence...................................................................................43 3.6.1 Considerations for standard assessment of violence at pre-treatment .45 3.7 Predictors of couple and family violence ...................................................46 4. Work functioning..............................................................................................49 5 Method...............................................................................................................53 5.1 Me as researcher, therapist and local administrator....................................53 5.2 Participants..................................................................................................55 5.3 Recruitment and procedures .......................................................................57 5.4 Measurements .............................................................................................58 5.4.1 Individual distress – Level of depression.............................................58 5.4.2 Relational – first level – Couple distress .............................................58 5.4.3 Relational – second level – Family distress.........................................59 5.4.4 Societal functioning – Work functioning.............................................59 5.4.5 Systemic Therapy Inventory of Change – STIC..................................60 5.4.6 Domestic violence................................................................................60 5.5 Summary of the method, data, and sample.................................................61 6. Summary of results...........................................................................................63 6.1 Paper 1 ........................................................................................................63 6.2 Paper 2 ........................................................................................................63 6.3 Paper 3 ........................................................................................................64 7 Discussion..........................................................................................................65 7.1 Discussion of main findings .......................................................................65 7.1.1 Paper 1 – Simplification of measures ..................................................65  XIII 7.1.2 Paper 2 – Domestic violence................................................................66 7.1.3 Paper 3 – Work functioning.................................................................69 7.2 General discussion of the method...............................................................73 7.2.1 General discussion of study design......................................................73 7.2.2 Discussion of reliability and validity ...................................................74 7.3 Discussion of strengths and limitations ......................................................77 7.3.1 Paper 1..................................................................................................77 7.3.2 Paper 2..................................................................................................79 7.3.3 Paper 3..................................................................................................81 7.3.4 Implications and handling of missing data ..........................................82 8 Conclusions .......................................................................................................87 9 References .........................................................................................................89 10 Appendix, paper 1-3 ......................................................................................107  XIV List of abbreviations in this thesis APA American Psychological Association BDI Beck Depression Inventory CFT Couple and Family Therapy DV Domestic Violence EBP Evidence-Based Practice EBT Evidence-Based Treatment FAD Family Assessment Device IPS Individual Problems and Strength scale IPV Intimate Partner Violence RCT Randomized clinical trial RDAS Revised Dyadic Adjustment Scale ROM Routine Outcome Monitoring systems STIC Systemic Therapy Inventory of Change List of tables and figures in this thesis Table 1: Overview of the domain, type of data and sample for the three papers Figure 1: Different levels of simplification  1 Introduction 1.1 Aims of this thesis The aims of this thesis were to investigate couple relationships and change in those relationships during couple and family therapy (CFT), with a specific focus on domestic violence and work functioning. The first subsidiary aim was to assess, theoretically and empirically, what level of simplification should be applied to the comprehensive measure used in this study. The second subsidiary aim was to examine physical couple and family violence among clients receiving CFT. The final subsidiary aim was to analyze the change in the clients' work functioning from pre- to post-treatment. In the following introduction, I will start by placing this thesis within its context. I start by defining CFT and describe its historical background and philosophical basis. I will then present some main developments and findings within CFT research. Further, I will present a more specific background for each of the three papers. I start by presenting some of the complexity of measuring change in psychotherapy and, thereby, laying out the rationale for Paper 1. I will then present the different focuses that have been applied within psychotherapy research and present arguments for why I have chosen to focus on domestic violence (Paper 2) and work functioning (Paper 3) in this thesis. 1.2 Definition and history of Couple and Family Therapy 1.2.1 Definition of CFT The term CFT captures a vast area of different perspectives and treatment programs (Lorås & Ness, 2019; Ness, 2017; Sexton, Datchi, Evans, LaFollette, & Wright, 2013), even if all are related to general systems theory (Von Bertalanffy, 1972) and communications analysis (Bateson, Jackson, Haley, & Weakland, 1956). CFT is the branch of psychotherapy with a focus on working with families and couples in intimate relationships where the aim is to contribute to change and development (Johnsen & Torsteinsson, 2012). The preferred focus of change is on the relational level rather than on the individual, i.e., the couple or the family. Change is, from this perspective, most often seen in patterns of cooperation or how people relate to each other (Johnsen & Torsteinsson, 2012). CFT is 1