Elsevier

Behaviour Research and Therapy

Volume 74, November 2015, Pages 72-79
Behaviour Research and Therapy

Predictors of response to CBT and IPT for depression; the contribution of therapy process

https://doi.org/10.1016/j.brat.2015.09.003Get rights and content

Highlights

  • We examined predictors of response to CBT and IPT in a RCT with depressed outpatients.

  • Predictors were therapy logic, recurrent depression, childhood reason for depression.

  • Comorbid personality disorders symptoms predicted a poorer response to IPT only.

  • Patient psychic distress, patient participation in therapy and patient alliance were associated with outcome.

  • Only end of therapy therapeutic processes and alliance contributed to outcome beyond the specific predictors.

Abstract

Little is known about the factors that contribute to a positive psychotherapy outcome. There is still considerable debate as to whether specific factors (e.g. severity of symptoms, comorbidity) or nonspecific factors (e.g. alliance, therapy process) are most important in influencing outcome. This study examined the additional contribution that therapeutic process and alliance made to previously identified specific predictors of response to CBT and IPT for depression over the course of therapy. The previously identified specific factors were belief that childhood reasons caused the depression, recurrent depression, perceptions about how logical therapy was and comorbid personality disorder symptoms. One hundred and sixty five adult outpatients with major depression were treated for depression in a randomised clinical trial examining predictors of response to Cognitive Behavior Therapy and Interpersonal Psychotherapy. All therapy sessions were audiorecorded to enable objective ratings of therapeutic process and alliance. Process factors – patient psychic distress, patient participation and patient alliance had the strongest associations with outcome. The early and middle stage of therapy process did not account for any additional variance other than that previously identified by the patient predictors, however, at the end stage of therapy process contributed a further 14%.

Introduction

The efficacy of Cognitive Behavior Therapy (CBT) and Interpersonal Psychotherapy (IPT) for depression has been demonstrated over many decades of research. Despite this, relatively little is known about the factors that contribute to patients making a positive therapeutic response. Given that only 50–60% of those with depression will benefit from a structured short term therapy, understanding factors that contribute to patient improvement in therapy must remain a research imperative. Clinical characteristics such as high initial depression severity, chronic or recurrent depression, earlier age of onset of depression and comorbid Axis I disorders have all been consistently associated with a poorer response (Blom et al., 2007, Constantino et al., 2008, Frank et al., 2011, Jarrett et al., 1991, Thase et al., 1994). Other factors that have shown some association with a poorer response to therapy include higher levels of dysfunctional attitudes (Jarrett et al., 1991), perfectionism (Enns, Cox, & Pidlubny, 2002), beliefs about the causes of depression (Addis & Jacobson, 1996), and patient expectancy of treatment outcome (Gaston et al., 1989, Ravitz et al., 2011). To date there has been little support for specific therapeutic mechanisms proposed for each therapy, such as cognitive changes for CBT and resolution of interpersonal problems for IPT (Lipstiz and Markowitz, 2013, Longmore and Worrell, 2007). On the contrary, there are some indications that therapy that targets a person's relative strengths rather than deficits is more likely to be beneficial (Cheavens et al., 2010, Imber et al., 1990, Sotsky et al., 1991).

In previous studies, we examined patient predictors of response following weekly sessions of CBT and IPT for depressed outpatients in the Christchurch Psychotherapy for Depression study (CPDS) (Carter et al., 2011, Joyce et al., 2007). Previous analyses of outcome in the CPDS indicated there was no difference in outcome between CBT and IPT (Joyce et al., 2007, Luty et al., 2007). Patient predictors, accounting for about 20% of the variance in response, were the perceived logic of therapy (measured immediately after the first session of therapy), recurrent depression, and childhood reasons for depression (Carter et al., 2011). More specifically, those who believed that the therapy they were about to receive seemed logical, who did not have a recurrent depression and who held moderate (not mild or strong) beliefs that childhood reasons were the cause of their depression had a better response to therapy. Only one differential predictor of treatment response was identified, increasing comorbid personality disorder symptoms were associated with a decrease in response to IPT but not CBT. The average percentage change from pretreatment to post treatment, when more than ten comorbid personality disorder symptoms were present, was 30% for IPT and 58% for CBT, with five to ten comorbid personality disorder symptoms the average percentage change for IPT was 45% and 62% for CBT and with four or fewer comorbid personality disorder symptoms the average percentage change for IPT was 50% and for CBT was 58% (Carter et al., 2011).

Many argue that nonspecific factors or those common to all therapies make the largest contribution to treatment outcome (Goldfried, 2013, Lambert and Barley, 2001, Messer and Wampold, 2002, Safran and Muran, 1995, Wompold, 2001) and that this is where research attention should be directed. Therapist attitudes and behaviours, such as therapist credibility, skills, empathic understanding and affirmation of the patient, have been found to have a positive impact on outcome (Norcross, 2011). In particular, some have proposed that the therapeutic alliance is the most important contributor to therapeutic outcome and is a causal factor in patient improvement (Barber, Connolly, Crits-Cristoph, Gladis, & Siqeuland, 2000). Meta-analyses of the therapeutic alliance–outcome relationship suggest the therapeutic alliance makes a small to moderate contribution (accounting for 5–7% of the variance) to a positive psychotherapy outcome, irrespective of the type of psychotherapy (Castonguay et al., 2006, Horvath et al., 2011, Martin et al., 2000). There is consistent evidence from different lines of research, that patient processes in therapy may be a particularly important determinant of outcome. For example, the patient's contribution to the alliance, as opposed to the therapist's contribution, relates most strongly to outcome (Ablon and Jones, 1999, Stiles et al., 1986). Further, positive patient expectancy (Lambert & Barley, 2001), and patient belief in the treatment rationale (Carter et al., 2006) have been found to be associated with a positive therapy outcome, and poor outcome is consistently predicted by patient unwillingness or inability to become actively involved in therapy (Butler & Strupp, 1986).

The aim of the current study was to examine the added contribution of non-specific therapy process factors measured at three stages (early, middle and penultimate sessions) over the course of therapy to the specific patient predictors of response to therapy previously identified (Carter et al., 2011).

Section snippets

Method

Data for this study came from 165 (males n = 46; females n = 119) of 177 outpatients in the CPDS and for whom Vanderbilt Psychotherapy Process and Vanderbilt Therapeutic Alliance Scale data were available. Participants in the CPDS had a principal current diagnosis of major depressive episode (DSM-IV) and were over the age of 18 years. Participants were required to be free for a minimum of two weeks of any centrally acting drug, except for the occasional hypnotic and the oral contraceptive pill.

Vanderbilt Therapeutic Alliance Scale – revised

The VTAS-R is 37 item modification of the original Vanderbilt Therapeutic Alliance Scale (Krupnick, Sotsky, Watkins, Eklin, & Pilkonis, 1996). The VTAS-R is an observer rated scale with three subscales, therapist contribution, patient contribution and therapist and patient interaction, which can be summed to give a total alliance score. High scores indicate positive therapeutic alliance. Theoretically the VTAS represents a blend of dynamic and eclectic frameworks. The VTAS-R has demonstrated

Results

Demographic and clinical characteristics of the sample can be seen in Table 1. The mean age was 35 years, 44% of the sample was married, and the majority were female and New Zealand European. Approximately 60% had a current comorbid Axis I disorder and the majority of the sample had recurrent depression (74%). Consistent with previous results there was no significant difference between therapies in depression severity prior to or at the end of weekly therapy sessions. In the IPT group mean

Discussion

The current study examined the association of psychotherapeutic factors with outcome during CBT and IPT for depression, and examined the contribution significant psychotherapy process factors made to the previously identified specific patient predictors of outcome. Scores for the Vanderbilt Process Scale and the Vanderbilt Alliance Scale found in the current study were consistent with previous studies in this area (Cecero et al., 2001, Krupnick et al., 1994, Smith et al., 2003). Similarly, the

Funding

This research was funded by grants from the Health Research Council of New Zealand (98/134).

Acknowledgements

Particular thanks to researchers, therapists and clinicians who worked on this study and special thanks to those who participated in the study.

References (62)

  • J.P. Barber et al.

    Alliance predicts patients' outcome beyond in-treatment change in symptoms

    Journal of Consulting and Clinical Psychology

    (2000)
  • J. Beck

    Cognitive therapy: Basics and beyond

    (1995)
  • A. Beck et al.

    Cognitive therapy of depression

    (1979)
  • A. Brown et al.

    Is the therapeutic alliance overvalued in the treatment of eating disorders

    International Journal of Eating Disorders

    (2013)
  • E. Bulmash et al.

    Personality, stressful life events and treatment response in major depression

    Journal of Consulting and Clinical Psychology

    (2009)
  • S.F. Butler et al.

    Specific and nonspecific factors in psychotherapy: a problematic paradigm for psychotherapy research

    Psychotherapy: Theory, Research, Practice, Training

    (1986)
  • J. Carter et al.

    Predictors of response to interpersonal psychotherapy and cognitive behaviour therapy

    Journal of Affective Disorders

    (2006)
  • L.G. Castonguay et al.

    The working alliance: where are we and where should we go?

    Psychotherapy: Theory, Research, Practice, Training

    (2006)
  • J.J. Cecero et al.

    Focus on therapeutic alliance: the psychometric properties of six measures across three treatments

    Psychotherapy: Theory, Research, Practice, Training

    (2001)
  • J. Cheavens et al.

    The compensation and capitalization models: a test of two approaches to individualising the treatment of depression

    behavior Rsearch and Therapy

    (2010)
  • M.J. Constantino et al.

    Interpersonal styles of chronically depressed outpatients: profiles and therapeutic change

    Psychotherapy: Theory, Research & Practice

    (2008)
  • R.J. DeRubeis et al.

    Determinants of change in cognitive therapy for depression

    Cognitive Therapy & Research

    (1990)
  • K.S. Dobson et al.

    Reliability of a measure of the quality of cognitive therapy

    British Journal of Clinical Psychology

    (1985)
  • I. Elkin et al.

    Therapist effects in the national institute of mental health treatment of depression collaborative research program

    Psychotherapy Research

    (2006)
  • M.W. Enns et al.

    Group cognitive behaviour therapy for residual depression: effectiveness and predictors of response

    Cognitive Behaviour Therapy

    (2002)
  • M. Feeley et al.

    The temporal relation of adherence and alliance to symptom change in cognitive therapy for depression

    Journal of Consulting and Clinical Psychology

    (1999)
  • E. Frank et al.

    Predictors and moderators of time to remission of major depression with interpersonal psychotherapy and SSRI pharmacotherapy

    Psychological Medicine

    (2011)
  • L. Gaston et al.

    Impact of confirming patient expectations of change process in behavioural, cognitive, and brief dynamic psychotherapy

    Psychotherapy

    (1989)
  • M.R. Goldfried

    What should we expect from psychotherapy?

    Clinical Psychology Review

    (2013)
  • D. Hartley et al.

    The therapeutic alliance: its relationship to outcome in brief psychotherapy

  • C.E. Hill et al.

    Applying the Collaborative Study Psychotherapy Rating Scale to rate therapist adherence in cognitive-behavior therapy, interpersonal therapy, and clinical management

    Journal of Consulting and Clinical Psychology

    (1992)
  • Cited by (11)

    View all citing articles on Scopus
    View full text