Alliance Ratings Research Articles

Psychological versus pharmacological treatments of bulimia nervosa: Predictors and processes of change.

This article extends the acute outcome findings from a study comparing psychological and pharmacological interventions for bulimia nervosa (B.T. Walsh et al., 1997) by examining 3 additional domains: predictive factors, therapeutic alliance, and time course of change. One hundred twenty women were randomized to cognitive-behavioral therapy (CBT), supportive psychotherapy (SPT) plus antidepressant medication or a placebo, or a medication-alone condition. Results indicate that high baseline frequencies of binge eating and vomiting, as well as a positive history of substance abuse or dependence, are negative prognostic indicators. Although a greater overall therapeutic alliance may increase the likelihood of remission, symptom change over the course of treatment may have as much of an impact on patient ratings of alliance as the reverse. CBT was significantly more rapid than SPT in reducing binge eating and vomiting frequencies.

Outcome assessment after milieu-oriented rehabilitation: New considerations

Objective: This study incorporated level of functional impairment ratings at program admission into rehabilitation outcome (return to work/school) at the time of program discharge. Hypotheses: (1) Patients and families with better working alliance ratings will show better adjusted outcome; (2) patients seeking compensation will have poorer outcome than those not seeking compensation or receiving benefits; (3) speed of information processing and memory will relate to the level of adjusted outcome. Design: Consecutive neurorehabilitation admissions from March 1992 to May 1996. Setting: Outpatient milieu-based interdisciplinary day treatment program. Subjects: Sixty-four patients with heterogeneous brain injury etiologies. Main Outcome Measures: Adjusted outcome, defined as level of discharge productivity adjusted by staff ratings of functional severity of impairment at program admission; work readiness and work eagerness, based on average staff ratings. Results: At discharge, 89.5% of patients showed fair or good adjusted outcome; 62% were gainfully employed/full-time students; 15.6% resumed preinjury status. Better working alliance predicted better adjusted outcome. Patients seeking compensation showed significantly lower work eagerness ratings. Poorer outcome was associated with better neuropsychological status. Conclusions: Efficacy of neurorehabilitation was demonstrated for patients with better working alliance who were not seeking compensation. Adjusted outcome demonstrated greater sensitivity and utility by incorporating the variable of functional severity of impairment at program admission. Inclusion of “process” variables addressing working alliance, motivation, and capacity to work provide important contributions to understanding rehabilitation outcome.

Contribution of therapist experience and personal characteristics to the working alliance.

The authors hypothesized that level of experience and selected personal characteristics of therapists would predict clients' perceptions of the working alliance early in treatment. Therapists (N = 73) in university settings completed measures that assessed level of self-directed hostility, perceived social supports, and degree of comfort with attachment, and their clients completed a measure that estimated the working alliance between the 3rd and 5th sessions of therapy. As predicted, level of self-directed hostility, perceived social support, and degree of comfort with closeness in interpersonal relationships were significantly predictive of the bond component of the working alliance. Contrary to predictions, experience level was not uniquely predictive of clients' alliance ratings on the goal and task components of the alliance.

Therapeutic alliance and outcome in a VA intensive case management program.

A total of 143 clients and their case managers in a Veterans Affairs (VA) intensive case management program modeled on the Program for Assertive Community Treatment rated their therapeutic alliance after two years in the program. Strong case-manager-rated alliance was associated with reduced symptom severity and improved global functioning as rated by independent assessors; it was also associated with higher client ratings of community living skills and more positive outcome as perceived by both clients and case managers. Strong client-rated alliance was associated only with more positive client-perceived outcome. Alliance ratings were not associated with use of inpatient psychiatric hospitalization. The case manager-client alliance appears to be a significant component of therapeutic effectiveness.

The Effect of Nonverbal Skills Training on Counselor Trainee Nonverbal Sensitivity and Responsiveness and on Session Impact and Working Alliance Ratings

Counselor trainees (N = 18) were randomly assigned to treatment (nonverbal sensitivity) or control (empathy training) conditions. Trainees saw a recruited client before (first 2 weeks) and after (last 2 weeks) a 15‐week counseling methods class. Trainees rated their sensitivity to nonverbal behaviors, counseling self‐efficacy, and the extent to which they focused on client nonverbal behavior. Clients filled out the Session Evaluation Questionnaire and Working Alliance Inventory at pre‐ and posttesting. Trained raters viewed videotapes of the counseling sessions and rated the extent that counselors focused on client nonverbal behavior. As hypothesized, trainees in the treatment condition, when compared with those in the control condition, increased their focus on client nonverbal behaviors. In addition, clients of treatment condition counselors showed significant differences in working alliance ratings. Trainees in both conditions increased their rating of nonverbal sensitivity and self‐efficacy from pre‐ to posttesting. Implications are discussed.

It Takes Two to Do Therapy: On Differential Aspects in the Formation of Therapeutic Alliance

For two therapies the dependence of the therapeutic alliance on the indices of the alliance of the partner is studied using multivariate dynamic analyses. The proposed model starts from the assumption that it is mainly the patient's alliance that is dependent on the therapist's alliance indicators, although the reverse relationship is considered as well. From videotapes, ratings of the achieved alliance had been made by two methods, the Penn-Helping Alliance Scales (Penn-HAS), and the Therapeutic Alliance Rating System (TARS). Whereas for the Penn-HAS the overall ratings for the patient and therapist were strongly mutually dependent, the dependence for the TARS was less strong, requiring more in-depth analyses based on the factor scores of the original ratings.

Effect of Rater Orientation on the Evaluation of Therapeutic Alliance

Alliance ratings from psychodynamic coders were compared with those of cognitive–behavioral coders on the same data set of significant psychodynamic–interpersonal (n = 13) and cognitive–behavior th...

Working alliance, attachment memories, and social competencies of women in brief therapy.

Current social competencies and memories of attachment bonds with each parent were examined as they related to influences on formation of the working alliance. Female clients (N = 76) at 4 university and community agencies completed surveys containing the Parental Bonding Instrument, the Self-Efficacy Scale, the Adult Attachment Scale, and the Working Alliance Inventory. Results indicated that parental bonds, especially with fathers, were significantly associated with social competencies. Multiple regression analyses indicated that social competencies (self-efficacy and adult attachment) accounted for 14% of the variance in client working alliance ratings, whereas recalled parental bonds accounted for 23% of the variance. Working alliance was negatively associated with father bonds, positively associated with mother bonds, and-among social competencies-significantly associated with capacity for adult attachment

Client attachment and perceptions of the working alliance with counselor trainees.

The relationships among 3 dimensions of client attachment (depend, anxiety, and close) and client ratings of the working alliance were examined in a sample of first-time clients (N = 60) seeking services through a university-based counseling clinic. Working alliance ratings were collected from all participants immediately after their 3rd counseling session with a counselor-in-training. A significant association was found between the depend dimension of attachment and client working alliance ratings

Alliance and Technique for Predicting Outcome in Short-and Long-Term Analytic Psychotherapy

A previous clinical trial provided the opportunity to examine the roles and interaction of alliance and technique in both short- and long-term analytic psychotherapy. Two separate teams of trained clinical judges rates either alliance on the CALPAS or therapist interventions on the ITS. Outcome included symptomatology and interpersonal problems. In short-term therapy, working and therapeutic alliance ratings contributed significantly to the reduction of symptomatology only. In long-term therapy, working alliance ratings contributed significantly to the reduction of interpersonal problems, and the interaction of working alliance ratings with supportive or exploratory intervention rating were found to account for significant amounts of the variance of both symptomatology and interpersonal problems at termination. These findings should be considered with caution, given the restricted sample sizes.

The Working Alliance: A Comparison of Two Therapies

Alliance ratings of single significant sessions of cognitive-behavioral and psychodynamic-interpersonal therapies were compared using the observer form of the Working Alliance Inventory (WAI-O). Eighteen cognitive-behavioral and 13 psychodynamic-interpersonal therapists, nominated by experts in the field, participated in the study. Results indicate significantly higher total alliance scores for cognitive-behavioral sessions, and greater variability in alliance for psychodynamic-interpersonal sessions. In addition, SCL-90 scores were negatively correlated with the alliance solely in psychodynamic therapy, indicating that more symptomatic patients may have greater difficulty with the work required in this kind of therapy.

Initial findings on preventive intervention for families with parental affective disorders.

The purpose of this study was to develop a clinician-based cognitive, psychoeducational, preventive intervention for families with parental affective disorder that would be suitable to widespread use, test its feasibility and safety, and define the areas affected by the intervention. The intervention was designed to increase understanding of parental illness and resilience in the children. The authors studied the first seven families (14 parents) to receive the intervention. Enrollment criteria included affective disorder during the preceding year in at least one parent, presence of at least one child between the ages of 8 and 14 years who was not psychiatrically ill at the time of participation, and willingness to participate in the research study. The intervention consisted of parent, child, and family sessions. Assessment included semistructured interviews with parents about affective disorders, standard ratings of marital satisfaction and therapeutic alliance, and a recently developed semistructured interview to assess response to the intervention. Overall satisfaction with the intervention was rated moderate to high by parents. No harm was reported. Ten of 14 parent subjects reported five or more behavior and attitude changes that they attributed to the intervention. The most frequent behavior and attitudinal changes reported were increased discussion of the illness and related issues and increased understanding of information about affective illness. The authors conclude that the intervention is safe and feasible in families with parental affective disorder.

Therapeutic Alliance: What Can be Learned from Case Studies?

Two different video-taped short-term psychotherapies were rated on therapeutic alliance for 27 sessions. The ratings were based on two methods: the Penn Helping Alliance Scales (Penn-HAS) and the Therapeutic Alliance Rating Scale (TARS). Both scales provide separate scores for the therapist and the patient. After checking the reliabilities of the ratings, the strength and course of the alliance formation are compared within and across therapies. Furthermore it was hypothesized that the alliance is structured differently in the two therapies. This hypothesis is tested by making content-specific item comparisons across the therapies for the Penn-HAS and by a comparison of the factor structure over time (factor analytic P-technique) for the TARS.

Clients' representations of childhood emotional bonds with parents, social support, and formation of the working alliance.

Survey data were collected from 102 client-counselor dyads at 3 counseling centers with regard to (a) client and counselor 3rd-session working alliance ratings, (b) quality of clients' current social relationships, as evidenced by social-support satisfaction, and (c) clients' representations of care and overprotection in their memories of childhood emotional bonds with their mothers and fathers. Care pertains to memories of parents' emotional responsiveness, warmth, and attention to clients' needs. Overprotection pertains to memories of parents' intrusive control and resistance at attempts to gain independence. Results indicated that social support was a significant predictor of client-rated working alliance, whereas clients' bonds with fathers were a significant predictor of counselor-rated working alliance. A trend toward interactions between social support and parental bonds in predicting counselor-rated alliance was also suggested. The working alliance, with conceptual origins in psychodynamic theory (cf. Sterba, 1934; Zeitzel, 1956), has been proposed as a part of the counseling relationship that is crucial to successful outcome regardless of the theoretical underpinnings of the therapy (Bordin, 1979, 1980). Drawing from the work of Greenson (1967) and Bordin (1979), Gelso and Carter (1985) described the working alliance as an emotional alignment between the counselor and the client that engages the self-observing, rational aspects of the client and the working, therapizing aspects of the counselor. Bordin (1979) suggested that the working alliance is composed of emotional bonds of trust and attachment between counselor and client, as well as agreement about the goals of treatment and the therapeutic tasks needed to achieve those goals. Evidence documenting the importance for successful therapy outcome of the counseling relationship in general (Orlinsky & Howard, 1986), and the working alliance in particular (for reviews see Gaston, 1990; Horvath & Symonds, 1991), continues to accumulate. Some studies have reported quite strong associations, with the working alliance accounting for up to 30%-45% of the variance in therapeutic outcome (Horvath & Greenberg, 1989). A recent meta-analysis reported a more modest, but nevertheless significant, effect size averaged across 24 studies (Horvath & Symonds, 1991). Although the bulk of previous research has examined the working alliance as a predictor of therapy outcome, more attention has been given recently to predictors of the working alliance itself, with studies of factors that affect counselors' capacity to form working alliances (Mallinckrodt & Nelson, 1991), as

Quality of Object Relations versus Interpersonal Functioning as Predictors of Therapeutic Alliance and Psychotherapy Outcome

The purpose of the study was to compare an interview measure of quality of object relations to questionnaire and interview measures of recent interpersonal functioning with respect to the prediction of therapeutic alliance and psychotherapy outcome. The sample consisted of 64 patients who had received approximately 20 sessions of short-term individual psychotherapy within a controlled, clinical-trial investigation. Ratings of therapeutic alliance were provided independently by the patient and the therapist after each session. Outcome measures, which were provided by three sources (patient, therapist, and independent assessor), covered the areas of interpersonal functioning, psychiatric symptomatology, self-esteem, and life satisfaction. Quality of object relations, which characterizes the patient's lifelong pattern of relationships, was the best predictor. It was significantly related to patient-rated and therapist-rated therapeutic alliance and to patient improvement of both general symptomatology and specific target problems. The study also replicated previous studies that have reported significant relationships between therapeutic alliance and therapy outcome. The advantages of pretherapy predictors of therapy outcome, such as quality of object relations, are discussed.

The helping alliance and premature termination

Abstract The study was conducted to determine the relationship between strength of the helping alliance and type of client termination (premature or with mutual knowledge of client and counselor). Participants were 102 client-counselors dyads at a university counseling center. After an average of 8 sessions, clients and couneselors completed Alexander & Luborsky's (1986) Helping Alliance Questionnaires. Clients terminated after an average of 19 sessions. Clients who later terminated with mutual knowledge of their counselors gave significantly higher strength of helping alliance ratings than did clients who later terminated unilaterally and prematurely. Counselors' ratings of strength of helping alliance were only modestly related to clients' ratings and unrelated to type of client termination.

Outcome at Discharge for Patients in an Ongoing Follow-up Study of Hospital Treatment

Methodological and practical difficulties have limited the growth of knowledge about outcome of psychiatric hospital treatment. The authors report on outcome at hospital discharge for 103 long-term and 93 short-term patients treated at the C. F. Menninger Memorial Hospital, part of an ongoing follow-up study of hospital treatment. Discharge outcome is based mainly on ratings of symptoms, global functioning, and therapeutic alliance as well as on patients' reports of satisfaction. At discharge both long- and short-term patients were found to have low levels of symptoms (based on the Brief Psychiatric Rating Scale) and a relatively adequate level of functioning (in the 51-to-60 range on the Global Assessment Scale) and to have been highly satisfied with treatment.

Toward the validation of the California Therapeutic Alliance Rating System.

The California Therapeutic Alliance Rating System, was used by judges to rate four therapy sessions for each of 52 subjects treated in brief dynamic psychotherapy for pathological grief

Therapeutic alliance and hospital treatment outcome.

The effect of therapeutic alliance on treatment outcome in a stratified sample of 96 hospital inpatients was assessed by means of a 6-point alliance scale as well as by demographic, treatment history, and psychopathological data. Alliance was correlated with improvement at discharge but not with demographic or prior treatment characteristics. Symptom patterns and personality dysfunction were shown to be differentially related to the quality of the alliance. Patients with axis I substance abuse, adjustment, and somatoform disorders had the poorest alliance and the poorest outcomes. The authors generated a model to predict outcome of hospital treatment, using patient alliance ratings, axis I and II diagnoses and their combinations, and admission Global Assessment Scale scores.

Predicting the outcomes of psychotherapy by the Penn Helping Alliance Rating Method.

The Penn Psychotherapy Project reports further progress in its search for the factors predicting the outcomes of psychotherapy through the construction of the Penn Helping Alliance Rating Method. This method is based on the following two types of patients' statements during psychotherapy sessions: type 1, that the therapist is helping the patient, and type 2, that the patient and therapist are working together in a team effort to help the patient. The scales were applied reliably to the transcripts from ten patients who had improved the most and ten who had improved the least among the 73 in the project. Significant predictive correlations were found for the early helping alliance measures (eg, with status at the termination of treatment and with the composite of "success, satisfaction, and improvement"). In contrast, other theoretically important treatment variables were not significant predictors.