Sheffield Psychotherapy Project Research Articles

Rewiring efficacy studies to increase their relevance to routine practice

Current efficacy literature relies heavily on the Beck Depression Inventory (BDI) as the gold standard patient self-report measure. In contrast, the evaluation of psychological therapies in routine practice relies heavily on the CORE-OM. Although the two measures are conceptually distinct, they have been shown to be highly correlated. This suggests the possibility of replacing one measure with the other - a procedure we refer to as rewiring - in service of making the results of efficacy studies using the BDI have greater relevance of practitioners who routinely use the CORE-OM. We tested this proposition using transformation tables (Leach et al., in press) to convert BDI-I scores into CORE-OM scores and reran the analysis of a major efficacy study of depression - the Second Sheffield Psychotherapy Project (Shapiro et al., 1994). Results showed a near perfect replication of the original results and examples of benchmarks concerning the overall effects of treatment as well as differences between treatments are provided against which outcomes in routine practice can be contrasted. The implications for bridging efficacy and effectiveness research are discussed.

Relations of the alliance with psychotherapy outcome: Findings in the second Sheffield Psychotherapy Project.

Relations of the alliance with psychotherapy outcome: Findings in the second Sheffield Psychotherapy Project.

Alliance structure assessed by the Agnew Relationship Measure (ARM).

We used a new instrument, the Agnew Relationship Measure (ARM), to examine the alliance's dimensionality and to construct scales for use in future studies. We studied the alliance as reported on the ARM by both clients and therapists in the Second Sheffield Psychotherapy Project, a randomized comparison of two contrasting time-limited psychotherapies for depression. Clients (N = 95) and therapists (N = 5) completed parallel forms of the (ARM) after every session (N = 1120). Five scales were constructed, based on results of simultaneous components analyses and considerations of conceptual coherence and comparability across client and therapist perspectives. Bond, partnership, and confidence overlapped statistically (consistent with previous studies), but there are conceptual and empirical reasons for retaining the distinctions. Openness represented a relatively independent dimension. The client initiative scale had low internal consistency but the items may have value for future investigations.

When and why does cognitive-behavioural treatment appear more effective than psychodynamic-interpersonal treatment? Discussion of the findings from the Second Sheffield Psychotherapy Project

The main outcome and process findings from the Second Sheffield Psychotherapy Project (SPP2) support the equivalence paradox: the overall outcomes of cognitive-behavioural (CB) and psychodynamic-interpersonal (PI) therapy were roughly the same, but the content of the two treatments were different. However, when we compared specific groups of clients or specific problems CB therapy showed an outcome advantage to PI therapy. Subsequent analyses of the data suggested that two factors accounted for these differences in treatment outcome and processes: the extent to which clients had assimilated their problems at the beginning of therapy and therapists' responsiveness to clients' interpersonal style. The implications of the SPP2 findings for clinical practice are discussed.

Relations of the alliance with psychotherapy outcome: findings in the Second Sheffield Psychotherapy Project.

Clients (n = 79) and therapists (n = 5) rated their alliance using parallel forms of the Agnew Relationship Measure (ARM) after every session of their time-limited psychodynamic-interpersonal or cognitive-behavioral treatments for depression. The ARM assesses 5 dimensions of the alliance: Bond, Partnership, Confidence, Openness, and Client Initiative. Treatment outcome was assessed as residual gain from pretreatment assessment to end of treatment, 3-month follow-up, and 1-year follow-up on 6 standard measures. Some aspects of the alliance as measured by the ARM were correlated with clients' gains in treatment. The strength of the association varied across assessment measures, occasions of outcome assessment, ARM scales, and the session number when the alliance was measured.

Outcomes of time-limited psychotherapy in applied settings: Replicating the Second Sheffield Psychotherapy Project.

Outcomes of time-limited psychotherapy in applied settings: Replicating the Second Sheffield Psychotherapy Project.

Therapist session intentions in cognitive–behavioral and psychodynamic–interpersonal psychotherapy.

William B. Stiles Miami University Mike Startup University of Wales, Bangor Gillian E. Hardy, Michael Barkharn, Shirley Reynolds Anne Rees, and David A. Shapiro University of East Anglia University of Leeds Theories of psychotherapy are implemented by therapists' intentional actions within sessions. This study examined the structure and construct validity of the Therapist Session Intentions (TSI) form. Ten therapists rated their therapeutic intentions following each of 2,305 therapy sessions of cognitive-behavioral or psychodynamic-interpersonal therapy in the Second Sheffield Psychotherapy Project. Seven conceptually coherent clusters of intentions, or therapeutic foci, were identified: treatment context, session structure, affect, obstacles, encouraging change, behavior, and cognition-insight. Contrasting patterns of therapeutic focus across treatments and changes in focus across sessions within treatments appeared conceptually coherent. Correlations across a subsample of sessions confirmed correspon- dences of the TSI foci with dimensions found previously in observers' ratings. The therapist is responsible for translating psychotherapy theory into practice by a series of intentional actions. This article reports a study of therapists' intentions in sessions of two theoretically different psychotherapies--one cognitive- behavioral (CB) and one psychodynamic-interpersonal (PI)--as implemented in the Second Sheffield Psychother- apy Project (Shapiro, Barkham, Hardy, & Morrison, 1990; Shapiro et al., 1994). Our goal was to assess the structure and construct validity of a postsession self-report instrument and to identify subscales for use in future studies. We examined how therapists' intentions clustered together within sessions of each treatment and how they changed systematically across sessions. We also compared thera- pists' reported intentions in sessions with ratings of their behaviors by raters who had listened to a subsample of the session tape recordings to assess the degree of convergence of descriptions of sessions' contents from these different perspectives. William B. Stiles, Department of Psychology, Miami Univer- sity, Oxford, Ohio; Mike Startup, School of Psychology, Univer- sity of Wales, Bangor, Wales; Gillian E. Hardy, Michael Barkham, Anne Rees, and David A. Shapiro, Psychological Therapies Re- search Centre, University of Leeds, England; Shirley Reynolds, Health Policy and Practice Unit, University of East Anglia, Nor- wich, England. This article was written in part while William B. Stiles was at the MRC/ESRC Social and Applied Psychology Unit, University of Sheffield, Sheffield, England, supported in part by Senior Interna- tional Fellowship 1 F06 TW01808-01 from the Fogarty Interna- tional Center of the National Institutes of Health. We thank Jane Hall and Mary Lou Hughes for administration of the clinic. Correspondence concerning this article should be addressed to William B. Stiles, Department of Psychology, Miami University, Oxford, Ohio 45056. Electronic mail may be sent via Internet to wbstiles @miamiu.muohio.edu. 402 As reported elsewhere (Shapiro et al., 1994, 1995), most clients in all cells of the Sheffield project design improved substantially across treatment, suggesting that both CB and PI treatments were very effective. There were no substantial mean differences in outcomes between clients who received CB versus PI treatment at termination or 3-month follow- up. Thus, the intentions we studied took place, in general, in the context of equivalently successful treatments. The Concept of Therapist Intention According to Hill and O'Grady (1985), Intentions can be defined as a therapist's rationale for select- ing a specific behavior, response mode, technique, or inter- vention to use with a client at any given moment within the session. Intentions represent what the therapist wants to ac- complish through his or her behavior within the session. An intention is the cognitive component that mediates the choice of intervention. (p. 3) Defined in this way, therapist intentions are epistemologi- cally private, and our most direct access to them is by asking therapists to reveal their thoughts. This sense of intention should be distinguished from sense of intent used in re- sponse mode coding (e.g., Could you come closer? has an interrogative form but a directive intent), which refers to the on-record meaning of utterances (Stiles, 1987). We use the former, private sense of intention in this article. As Hill and O'Grady (1985) pointed out, a systematic classification of intentions, thus conceived, may be useful for clarifying therapists' internal processes in teaching, clinical supervi- sion and practice, and research. Based on an initial list developed by Elliott and Feinstein (1978), a literature search, and a series of formal and infor- mal clinical studies, Hill and O'Grady (1985) constructed a

The Second Sheffield Psychotherapy Project and Collaborative Project: some initial findings and their clinical implications

The Second Sheffield Psychotherapy Project and Collaborative Project: some initial findings and their clinical implications

The structure, validity and clinical relevance of the Inventory of Interpersonal Problems

This paper presents preliminary data on the psychometric and clinical properties of the Inventory of Interpersonal Problems (IIP: Horowitz, Rosenberg, Baer, Ureno & Villasenor, 1988) obtained from a UK sample of clients presenting with depression/anxiety for individual psychotherapy as part of two concurrent outcome studies: the Second Sheffield Psychotherapy Project (SPP2) and the Medical Research Council/NHS Collaborative Psychotherapy Project (CPP). The factor structure of the IIP is reported on a sample of 250 clients and eight factor-analytically derived scales are presented. Data on the reliability of the IIP together with its relationship to the Symptom Checklist-90R (SCL-90R: Derogatis, 1983) are investigated in a subsample of 143 clients. In addition, data are presented on the clinical relevance of the IIP using case material. The structure of the IIP was further analysed using ipsatized scores. Four bipolar factors were extracted and these findings are discussed with reference to four primary sets of social competencies.

Therapist treatment fidelity in prescriptive vs. exploratory psychotherapy

Therapist fidelity to the manuals for prescriptive and exploratory psychotherapies was assessed via the Sheffield Psychotherapy Rating Scale. Ratings on 220 sessions drawn from the Second Sheffield Psychotherapy Project showed adequate inter-rater reliabilities within and between treatments. Discriminant analysis showed that the treatments could be differentiated almost perfectly even though the same five therapists delivered both of them. Relationship enhancing skills, measured by a facilitative conditions scale, were held constant across treatments and contributed nothing to the discriminant function. There was no evidence that adherence varied with the severity of the clients' symptoms and only very limited evidence that it varied with the duration of treatment, despite there being adequate statistical power to detect small effects. Small variations in adherence with the stage of treatment were found but only for sessions of prescriptive therapy.

Two-Year Follow-up of the Sheffield Psychotherapy Project

Of 40 depressed and anxious patients who received treatment during the Sheffield Psychotherapy Project, 31 completed the BDI and SCL-90 at two-year follow-up. At both group and individual levels, outcomes were very similar to those obtained at the completion of treatment. Substantial minorities of patients reported major life events and seeking further help for psychological problems during the two years, but these reports were not reliably associated with symptoms at follow-up. Patients recalled very different elements in prescriptive and exploratory therapies as helpful, with a notable emphasis upon prescriptive therapy's relaxation and anxiety management as a source of problem solution.

The Second Sheffield Psychotherapy Project: Rationale, design and preliminary outcome data

The cost-efficiency of psychotherapy research can be enhanced by (a) dissemination of the rationale and design of large-scale investigations whilst still under way, (b) incorporation of process measures within comparative outcome studies, (c) cumulative development of a series of related studies, and (d) sharing of recordings and other data with other investigators. We present the rationale and design, including full details of process and outcome measures, of the second Sheffield psychotherapy project, in which 120 white-collar, professional and managerial employees presenting with major depressive disorder receive either eight or 16 sessions of either prescriptive (cognitive/behavioural) or exploratory (relationship-oriented) therapy in a 2 x 2 design. Ethical considerations prompted the analysis of preliminary outcome data on 48 clients. Results suggested substantial clinical improvement, of similar magnitude for both methods and both durations of treatment, except that exploratory therapy appeared more effective than prescriptive in relation to interpersonal difficulties. Other workers are invited to build collaboratively on this research by making their own process analyses of recordings available from this study, relating these analyses to the existing impact and outcome data, and developing comparable data sets based on the design and initial results of this project.

Change in prescriptive versus exploratory therapy: Older clients' responses to therapy

Abstract This paper evaluates the relationship between expected rates of client improvement, based upon the dose-effect curve, and a specified client population: older employed clients. A subset of data from a comparative outcome study of psychotherapy (the first Sheffield Psychotherapy Project) was used in which a total of 40 clients received either eight sessions of exploratory or eight sessions of prescriptive therapy. Analyses of younger (N = 28) and older (N = 12) clients' responses to therapy using the Beck Depression Inventory (BDI) as a change measure, are presented. Further analyses of the 12 older clients' (aged 47-65) responses to therapy, using depression personal questionnaire data, are presented. Results pertain to (1) the differential rate of responses to treatment by the older as compared with the younger client group, and (2) differential responses of older clients to the two therapy modes. Results are discussed in relation to research findings and clinical practice.

Personal questionnaire changes in prescriptive vs. exploratory psychotherapy

Personal Questionnaire (PQ) data were collected within a larger study (the Sheffield Psychotherapy Project) in which 40 professional and managerial workers with depression and/or anxiety received eight sessions of prescriptive (cognitive‐behavioural) and eight sessions of exploratory (relationship‐orientated) therapy in a cross‐over design with each client seeing the same therapist throughout. Concurrent validity of PQ ratings with two established outcome measures are presented and suggest differing patterns of association with different outcome measures. Results based on a group comparison design as well as differential trend analyses, in which each case is analysed as a single case within a group design, are presented and suggest a greater differential response to treatments when the treatment order is prescriptive followed by exploratory therapy. Findings are presented relating clients' responses to treatment with the individual characteristic of clients' age. Implications for clinical research and practice are discussed.

Brief Structured Recall: a more efficient method for studying significant therapy events.

Previously, significant therapy events have been studied using a time-consuming form of tape-assisted recall. Brief Structured Recall (BSR) was developed to overcome difficulties with this earlier method. We carried out two case studies of time-limited therapy following the protocol of the Sheffield Psychotherapy Project (eight sessions each of Prescriptive and Exploratory therapies). In the first case study, we developed a more efficient method of identifying significant therapy events. In the second case study we developed structured interview schedules for obtaining client and therapist ratings and descriptions of a broad range of factors involved in significant events. The events identified in the two cases were typically 5-10 min in length, occurred in the second halves of sessions, contained 2-4 'peak' (most helpful) speaking turns, and were most often attributed to the therapist. The most common predominant impact of these events was Problem Clarification, followed by Personal Insight and Problem Solution. Treatment differences were more marked in the therapist's ratings of event impacts than in the clients' ratings.

Do sessions of different treatments have different impacts?

Each of 40 clients received eight sessions of an interpersonal/psychodynamic treatment, called exploratory, and eight sessions of a cognitive/behavioral treatment, called prescriptive, in a crossover design. The treatments were compared with respect to clients', counselors', and external raters' evaluations of the session and of participants' postsession moods. Consistent with theoretical expectations, exploratory sessions were rated as deeper and more powerful by counselors and external raters, whereas prescriptive sessions were rated as smoother and easier from all three perspectives. After prescriptive sessions clients' postsession mood was rated more positively by both the external raters and the clients themselves. Measuring a session's impact--its immediate effects, including participants' evaluations of the session and their postsession affective state--permits a closer look at what intervenes between process (events within sessions) and longterm outcome (effects of a series of sessions or of a whole treatment). Impact research offers a two-stage strategy for studying process-outcome relations, first linking process to impact and then linking impact to outcome(Stiles, 1980). This article addresses questions at the first of these two stages. We compared the session impact of two forms of counseling whose session processes differ and whose theoretical rationales suggest that they have different session impacts. The data were drawn from the Sheffield Psychotherapy Project (Firth, 1985; Firth, Shapiro, & Parry, 1986; Hardy &