Group member attachment style interacts with actor and partner helping alliance to predict decreasing binge eating episodes.

OBJECTIVE: An individual's attachment style may impact how they interact with their therapy group. This study examined the moderating role of a group member's attachment on the dynamic relationships between that group member's (actor) and other group members' (partner) therapeutic alliances and symptom outcomes. Method: This is a secondary analysis of data from a trial testing the outcome of emotionally-focused group therapy for binge-eating disorder. The sample consisted of 2,360 sessions nested within 118 group members who attended a 20-session treatment. Patients recorded binge eating episodes (BEE), their body weight and an alliance measure session-by-session. RESULTS: Dynamic structural equation modelling showed decreases in BEE and weight over the therapy. When attachment style was not included in the model, higher-than-average partner's alliance scores in the previous session were related to decreases in BEE in the current session. Attachment style moderated these actor and partner effects. For patients with preoccupied attachments, higher-than-average actor alliance in the previous session was related to subsequent decreases in BEE. For patients with dismissing or disorganized attachments, higher partner alliance in the previous session was related to subsequent decreases in BEE. CONCLUSION: Group members' attachment characteristics can play a role in the development of alliance-outcome patterns in group therapy.

The Promise and Challenges of Practice-Oriented Research: A Commentary on the Special Issue.

At the centre of POR is the concept of collaboration between patients, therapists, agencies, and third-party payers. For this commentary, I review the articles of the special issue with attention to both the opportunities and challenges offered by practice-oriented research (POR). I also reviewed some previous research on practice-research networks and how that research might inform POR. The use of routine outcome monitoring (ROM), artificial intelligence (AI), and program evaluation (PE) models show promise for advancing POR. However, each comes with its challenges. The use of ROM to improve patient outcomes has research support. However, researchers have identified problems with implementing ROM because of low uptake by clinicians and because clinicians may see ROM as a potential intrusion. AI shows promise to improve patient outcomes by potentially providing therapists with immediate and nuanced data to inform interventions and interpersonal stances specific to each patient. However, the scaling up of AI runs the risk of dehumanizing psychological interventions. PE may provide a context for allowing therapists to engage in POR to address real-world processes and outcomes of mental health interventions. However, like ROM PE faces the challenge of trust among clinicians and patients who may be reluctant to participate. Despite these challenges, and because of efforts to nurture and maintain respectful collaborations, articles in this special issue highlight how POR can play a pivotal role in bridging the gap between theory and practice.

Detecting defense mechanisms from Adult Attachment Interview (AAI) transcripts using machine learning.

OBJECTIVE: Defensive functioning (i.e., unconscious process used to manage real or perceived threats) may play a role in the development of various psychopathologies. It is typically assessed via observer rating measures, however, human coding of defensive functioning is resource-intensive and time-consuming. The purpose of this study was to develop a machine learning approach to automate coding of defense mechanisms from interview transcripts. METHOD: Participants included a clinical sample of women with binge-eating disorder (n = 92) and a community sample without binge-eating disorder (n = 66). We trained and evaluated five RoBERTa-based models to detect the presence of defenses in 16,785 interviewer-participant talk-turn pairs nested within 192 interviews. A model detected the presence of any defense, while four additional models detected the most common defenses in this sample (repression, intellectualization, reaction formation, undoing). RESULTS: The models were capable of distinguishing defenses (ROC-AUC .82-.90) but were not proficient enough to warrant replacing human coders (PR-AUC .28-.60). Follow-up analysis was performed to assess other practical uses of these models. DISCUSSION: Our machine learning models could be used to assist coders. Future research should conduct a deployment study to determine if human coding of defense mechanisms can be expedited using machine learning models.

Psychological changes in successful completers of an HIV-tailored smoking cessation program: mood, attachment and self-efficacy.

High rates of cigarette smoking is the leading contributor to the increasing risk of cardiovascular disease (CVD) among people living with HIV (PLH). Relapse rates among PLH who quit smoking are high among those receiving standard care, which may be due to several unique social and psychological challenges PLH face when they attempt to quit smoking. The purpose of the current study was to examine change in relevant psychological factors in a subgroup of participants (n = 14) who remained smoke-free at 6-months follow-up in an HIV-tailored smoking cessation counselling program (N = 50). We examined self-reported depressive symptoms, attachment style and self-efficacy across 5 time points (baseline, quite date, 4, 12 and 24 weeks). At study baseline, mean depression scores fell above the clinical cut off of 16 (M = 16.31; SD = 13.53) on the Centre for Epidemiological Studies - Depression (CES-D) scale and fell below the clinical cut off at 24 weeks post quit date (M = 13.36; SD = 10.62). Results of multi-level modeling indicated a significant linear reduction in depressive symptoms and a significant linear improvement in self-efficacy to refrain from smoking across study visits. These results suggest that positive change in mood and self-efficacy may be helpful for PLH who remain smoke-free during a quit attempt.

Self-esteem and binge eating: Do patients with binge eating disorder endorse more negatively worded items of the Rosenberg Self-Esteem Scale?

OBJECTIVE: Self-esteem is a core aspect of eating disorder symptomatology. This study aims to examine whether method effects associated with negatively worded items of the Rosenberg Self-Esteem Scale (RSES) may interact the negative self-evaluations experienced by patients with obesity and binge eating disorder (BED). We also examined whether negatively worded items were associated with psychological distress and eating symptoms. METHOD: Five hundred thirty three female outpatients (mean age: 42.59) with BED (n = 160) or obesity without BED (n = 373) completed the RSES and measures of interpersonal problems, psychological distress, and eating symptoms. RESULTS: Patients with BED responded more strongly to the negatively worded items of the RSES than those with obesity. The RSES negatively worded item factor was negatively associated with higher interpersonal problems, psychological distress, and binge eating. CONCLUSIONS: Patients with BED may be more responsive to negatively phrased items on the RSES consistent with their negative self-evaluations and self-perceptions.

Self-reported reflective functioning mediates the association between attachment insecurity and well-being among psychotherapists.

Objective: Subjective well-being is a crucial variable for mental health practitioners. This study examines the influence of therapists' attachment dimensions and self-reported reflective functioning on their perceived well-being. Further, it examines if reflective functioning mediates the association between attachment insecurity and well-being. Method: A total of 416 experienced psychotherapists were enrolled in this cross-sectional study, and completed self-report measures of attachment insecurity, reflective functioning, and well-being. We tested the hypothesized mediation model with path analysis that examined indirect effects. Results: Both attachment anxiety and avoidance dimensions had a significant negative association with perceived well-being with small to medium effects. "Certainty" in reflective functioning had a small positive effect on therapist well-being. Reflective functioning mediated the association between insecure attachment dimensions and well-being, suggesting that therapist's lower ability to mentalize may partially account for the effects of higher attachment insecurity on lower well-being. Conclusion: The well-being of psychotherapists with greater insecure attachment may deserve special attention, and therapists' mentalizing capacities may be targeted by researchers and trainers as a core ability to be cultivated in order to preserve therapists' professional and personal resources.

Mediators and moderators of change in mindfulness-based stress reduction for painful diabetic peripheral neuropathy.

Painful diabetic peripheral neuropathy (PDPN) is a chronic pain condition with modest response to pharmacotherapy. Participation in mindfulness-based stress reduction (MBSR) leads to improvements in pain-related outcomes but the mechanisms of change are unknown. The present study examined the mediators and moderators of change in 62 patients with PDPN who participated in a randomized controlled trial comparing MBSR to waitlist. Changes in mindfulness and pain catastrophizing were tested simultaneously as mediators. Increased mindfulness mediated the association between participation in MBSR and improved pain severity, pain interference, and the physical component of health-related quality of life (HRQoL) 3 months later. The mediation effect of pain catastrophizing was not significant. Linear moderated trends were also found. Post-hoc moderated mediation analyses suggested that MBSR patients with longer histories of diabetes might increase their mindfulness levels more, which in turn leads to improved pain severity and physical HRQoL. These results allow for a deeper understanding of pathways by which MBSR benefits patients with PDPN.

Testing a stepped care model for binge-eating disorder: a two-step randomized controlled trial.

BACKGROUND: A stepped care approach involves patients first receiving low-intensity treatment followed by higher intensity treatment. This two-step randomized controlled trial investigated the efficacy of a sequential stepped care approach for the psychological treatment of binge-eating disorder (BED). METHODS: In the first step, all participants with BED (n = 135) received unguided self-help (USH) based on a cognitive-behavioral therapy model. In the second step, participants who remained in the trial were randomized either to 16 weeks of group psychodynamic-interpersonal psychotherapy (GPIP) (n = 39) or to a no-treatment control condition (n = 46). Outcomes were assessed for USH in step 1, and then for step 2 up to 6-months post-treatment using multilevel regression slope discontinuity models. RESULTS: In the first step, USH resulted in large and statistically significant reductions in the frequency of binge eating. Statistically significant moderate to large reductions in eating disorder cognitions were also noted. In the second step, there was no difference in change in frequency of binge eating between GPIP and the control condition. Compared with controls, GPIP resulted in significant and large improvement in attachment avoidance and interpersonal problems. CONCLUSIONS: The findings indicated that a second step of a stepped care approach did not significantly reduce binge-eating symptoms beyond the effects of USH alone. The study provided some evidence for the second step potentially to reduce factors known to maintain binge eating in the long run, such as attachment avoidance and interpersonal problems.

Higher levels of Depressive Symptoms are Associated with Increased Resting-State Heart Rate Variability and Blunted Reactivity to a Laboratory Stress Task among Healthy Adults.

Laboratory stress tasks induce strong changes in linear and non-linear indices of heart rate variability (HRV) among healthy adults, due to a task-induced parasympathetic withdrawal. Previous findings suggested that negative affectivity and its correlates (i.e., depressive symptoms, anxiety, hostility, type D personality, and situational stress) could profoundly affect autonomic activity. However, to date no studies considered these psychological dimensions simultaneously while trying to disentangle their acute effects on HRV during a laboratory stress task. A total of 65 healthy participants completed a battery of questionnaires and later underwent a psychosocial stress protocol, which involves a stressful and a non-stressful mental arithmetic task, with the latter serving as a control condition for the former. During the entire procedure, autonomic activity was recorded through a portable ECG device. We analysed longitudinal changes in HRV indices using Mixed Models, taking into account respiration rates and the associations between psychophysiological variables through bivariate Pearson's r (partial) correlation indices. We found significant changes in linear (e.g., HF power, RMSSD) and non-linear (e.g., Poincaré Plot and Correlation Dimension D2) HRV indices during the procedure, with the lowest point reached during the stressful mental arithmetic task. Interestingly, only depressive symptomatology was significantly and positively related to a higher resting-state HRV and to a blunted reactivity to the stress task, even after controlling for baseline values. Results suggest that healthy individuals with higher levels of depressive symptoms could experience atypical cardiovascular responses to stressful events: several speculative interpretations, considering autonomic, behavioral, and motivational dysregulations, are discussed.

Attachment comes of age: adolescents' narrative coherence and reflective functioning predict well-being in emerging adulthood.

This study investigated the effects of adolescents' attachment security and reflective functioning (RF) (assessed by the adult attachment interview [AAI]) in the prediction of well-being in adulthood. Adolescents (N = 79; M = 14.6 years old; SD = 3.5 years) completed the AAI at Time 1 (T1), which was subsequently coded for inferred attachment experiences, narrative coherence, and RF by three nonoverlapping teams of raters. Participants completed the Psychological General Well-being Index at T1 and 8 years later (Time 2, T2). Analyses showed that (a) both adolescent narrative coherence and RF were significant predictors of almost all indices of well-being at T2 in adulthood; (b) both narrative coherence and RF indirectly linked inferred loving parental care and T2 well-being; (c) when included in the same model, RF was a significant indirect effect linking inferred loving parental care and T2 well-being. These findings contribute to theory in suggesting that both RF and narrative coherence are predictive of subsequent psychological well-being and operate as links between inferred parental care and subsequent adjustment. Possible mechanisms underlying these findings are discussed.

Persistence of friendly and submissive interpersonal styles among those with binge-eating disorder: Comparisons with matched controls and outcomes after group therapy.

Interpersonal problems play a prominent role in the development of binge-eating disorder (BED), so reducing their intensity may be a key focus of many psychological interventions. In recent years, several interpersonal treatments for BED were developed, which posit that binge eating arises to manage relational problems. However, few studies have evaluated the prototypical interpersonal problems, and no studies evaluated the longitudinal changes in interpersonal functioning after treatment within this population. We investigated the severity and prototypicality of interpersonal problems of 101 overweight women with BED from pre-group psychodynamic-interpersonal psychotherapy (GPIP) to 12 months post-GPIP. At baseline, we compared patients' interpersonal problems with two groups of matched controls (46 overweight and 49 normal weight women without a diagnosis of BED) and examined circular correlations between relational problems, depressive symptoms, and binge-eating frequency. Results showed that participants with BED had significantly higher levels of interpersonal problems compared with the matched control samples, with predominantly nonassertive and exploitable styles. Depressive symptoms were related to the presence of friendly-submissive interpersonal problems only among those with BED. Although the intensity of nonassertive interpersonal problems of patients with BED decreased post-group treatment, their profiles remained prototypically nonassertive and exploitable across all time points. Women with BED experience higher levels of interpersonal difficulties exemplified by an exploitable/nonassertive style that significantly improve but continue to prevail even after treatment. Clinicians might modify interventions to focus on increasing interpersonal skills and decreasing interpersonal dysfunctions among those with BED.

Psychotherapy for eating disorders: A meta-analysis of direct comparisons.

Objective: We conducted a meta-analysis of randomized controlled trials (RCTs) of bona fide psychotherapy for adults with eating disorders (EDs). Method: Thirty-five RCTs with 54 direct comparisons were included. The majority of RCTs included participants with bulimia nervosa and/or binge-ED, while only two RCTs included participants with anorexia nervosa, and three RCTs included participants with an ED not otherwise specified. Results: There was a clear advantage of bona fide psychotherapy over wait-list controls. Bona fide psychotherapy was superior to non-bona fide treatment; however, the majority of results were not stable. There were no significant differences between bona fide cognitive-behavioral therapy (CBT) and bona fide non-CBT, with the exception of bona fide CBT resulting in greater reductions in ED psychopathology assessed by the ED Examination, which primarily assesses maintenance factors according to the CBT model. Conclusions: Generally, the results indicate that any bona fide psychotherapy will be equally effective. While the number of trials remains modest, we hope that as more research becomes available, treatment guidelines can be updated, and more evidence-based treatment options will be available for treating EDs.

The Role of Emotional Competence in Takotsubo Cardiomyopathy.

OBJECTIVE: The current study examined the role of emotional competences and depression in takotsubo cardiomyopathy (TTC). METHODS: We compared 37 TTC patients who experienced emotion triggers (TTC-t: M (SD) age = 66.4 (12.8) years, 33 women) with 37 TTC patients who did not experience emotion triggers (TTC-nt: M (SD) age = 65.8 (11.1) years, 33 women) and 37 patients with acute myocardial infarction who experienced an emotion trigger (AMI-t: M (SD) age = 66.1 (10.1) years, 33 women). Three aspects of emotional competence (emotional intelligence, metacognitive beliefs, and emotional processing deficits) were assessed using the Trait Meta-Mood Scale (TMMS), the Meta-Cognitions Questionnaire 30, and the Emotional Processing Scale. Differences between-group means were evaluated using multivariate analysis of covariance, adjusting for depressive symptom (Hamilton Rating Scale for Depression). RESULTS: Compared with the TTC-nt and AMI-t comparison groups, TTC-t patients had low scores on emotional intelligence (TMMS Attention: F(2, 184) = 23.10, p < .001; TMMS Repair: F(2, 184) = 11.98, p < .001) and high scores in metacognitive beliefs and emotional processing deficits (e.g., Meta-Cognitions Questionnaire 30 Negative Beliefs about Thoughts: F(2, 184) = 56.93, p < .001), independent of the levels of depressive symptom. TTC-nt patients also had significantly lower scores on the Hamilton Rating Scale for Depression scale compared with AMI-t (p = .021) and TTC-t (p = .004) patients. CONCLUSIONS: TTC-t patients showed a specific dysfunctional profile of emotional competence, even after adjusting for depressive symptom. These results provided a better understanding of the psychological factors that contribute to TTC.

Effectiveness of Intraindividual Variability in Detecting Subtle Cognitive Performance Deficits in Breast Cancer Patients.

OBJECTIVES: The purpose of this study was to determine if intraindividual variability would be more sensitive than speed or accuracy in detecting subtle cancer-related cognitive disturbance. METHODS: Data were from a previous study in which 60 breast cancer (BC) patients underwent neuropsychological assessment before commencement of chemotherapy and again following each chemotherapy cycle. Sixty healthy controls were tested at equivalent intervals. Hierarchical linear modeling was used to compare the BC and control groups in terms of accuracy, mean reaction time, and intraindividual variability in reaction time on a computerized continuous performance test with three conditions: a simple reaction time task, a "1-back" task, and a "2-back" task. RESULTS: An increase in accuracy and response speed over sessions was noted on some tasks in the sample as a whole but there were no differences in these parameters between the BC patients and the controls on any condition. There was a significant group difference in change in intraindividual variability across sessions (i.e., a "group × session interaction"), albeit only on the most complex "2-back" task. Intraindividual variability declined in the control group (i.e., consistency improved with practice) but this practice effect was significantly attenuated in the BC patients. There was no main effect of group on the "2-back" task. CONCLUSIONS: Results support our hypothesis that intraindividual variability is a more sensitive indicator of subtle cognitive disturbance than conventional speed or accuracy measures and may have potential in the assessment of mild cognitive impairment in patients with non-central nervous system cancers. (JINS, 2018, 24, 724-734).

Reliability and validity of the Experiences in Close Relationships Scale-12: Attachment dimensions in a clinical sample with eating disorders.

OBJECTIVE: Attachment insecurity is a potential risk factor for the development and maintenance of eating disorders (EDs). To date, there are multiple psychometrically sound questionnaires for the evaluation of attachment in both clinical and healthy populations, such as the Experience in Close Relationships (ECR) scale. Composed by two subscales (i.e., attachment anxiety and avoidance), the ECR scale was recently adapted to a shorter, 12-item version (ECR-12). However, a validation of the ECR-12 among patients with EDs is still lacking. The present study sought to investigate the psychometric properties of the ECR-12, when used in a treatment-seeking sample with EDs. METHOD: A total of 1,262 treatment-seeking patients with various ED diagnoses completed the ECR-12, together with a commonly used measure of ED psychopathology (Eating Disorder Inventory-2). Subsamples also completed the Attachment Style Questionnaire (n = 66) or underwent a Day Hospital Program (n = 128). RESULTS: A multigroup confirmatory factor analysis indicated that the ECR-12 maintained its two-factor structure across the ED diagnostic groups. In addition, the scale demonstrated good convergent validity, internal consistency, concurrent, and incremental validity. Finally, both ECR-12 subscales demonstrated adequate test-retest reliability. DISCUSSION: Our results provide preliminary evidence of the psychometric properties of ECR-12 in assessing attachment anxiety and avoidance among patients with EDs. This short scale could help clinicians tailor interventions for EDs that take into account attachment dimensions.

Patients with Binge Eating Disorder and Obesity have qualitatively different interpersonal characteristics: Results from an Interpersonal Circumplex study.

BACKGROUND: Patients with Binge Eating Disorder (BED) and obesity experience distressing relationships, which could trigger negative affect and over-eating. To date no studies compared the interpersonal profiles and prototypicality of both groups using the Interpersonal Circumplex. METHOD: A sample of 177 patients with BED (mean age: 41.0 ±â€¯12.5 years; 11.3% males), 321 obese non-BED adults (mean age: 44.5 ±â€¯13.4 years; 28% males), and 108 normal weight adults (mean age: 37.3 ±â€¯9.6 years; 52.77% males) completed the Inventory of Interpersonal Problems (IIP-32), and scales of binge eating and psychological distress at one time-point. RESULTS: Compared to normal weight and obese participants, those with BED reported higher levels of interpersonal problems on all circumplex dimensions, except for Vindictive, with small to medium effects. All groups had highly prototypical profiles. Both obese and normal weight adults had very similar interpersonal profiles characterized by predominant friendly-dominant themes. On the contrary, those with BED had predominant friendly-submissive themes. Patients with BED reported significantly higher levels of psychological distress and binge eating severity, compared to the other two groups. Greater Domineering, Cold, Socially Inhibited and Non-Assertive scale scores, and lower Vindictive scale scores significantly predicted higher binge eating and psychological distress, regardless of group membership. DISCUSSION: Findings lend support to an interpersonal model of binge eating and to the presence of qualitative differences between patients with obesity and BED. Finally, results suggest some strategies for the clinical management of BED focused on non-assertion and problems with experiences and expression of anger.

Change in Attachment Dimensions in Women with Binge-Eating Disorder Following Group Psychodynamic Interpersonal Psychotherapy.

OBJECTIVE: To examine the role of attachment dimensions, including coherence of mind and reflective functioning, in developing and maintaining binge-eating disorder (BED) and in determining group psychotherapy outcomes for women with BED. We hypothesize that higher pre-treatment attachment dimension scores will predict better treatment outcomes for women with BED and will increase at follow-up. METHOD: Women with BED attended 16 sessions of group therapy and completed the Adult Attachment Interview (AAI) at pre-treatment. Participants completed outcome measures (i.e., binge-eating frequency and symptoms of depression) at pre-, post-, six months, and 12 months post-treatment. Treatment completers completed the AAI at six months post-treatment. RESULTS: Treatment outcomes improved significantly from pre- to 12 months post-treatment. Greater Reflective Functioning scores at pre-treatment were related to greater decreases in binge eating across the four time points, whereas Coherence of Mind scores were not. For treatment completers, there were significant increases in Reflective Functioning at six months post-treatment, and about a third of treatment completers experienced clinically significant increases in both attachment dimensions at six months post-treatment. CONCLUSIONS: Greater reflective functioning at the outset is important for improvements in binge eating in the longer term and group psychotherapy can facilitate change in reflective functioning.

An HIV-tailored quit-smoking counselling pilot intervention targeting depressive symptoms plus Nicotine Replacement Therapy.

Cardiovascular disease (CVD) rates among people living with HIV/AIDS (PHAs) are high. Rates of cigarette smoking, a leading contributor to CVD among PHAs, are 40-70% (2-3 times higher than the general population). Furthermore, PHAs have high rates of depression (40-60%), a risk factor for smoking cessation relapse. The current pilot study examined the effectiveness of a specifically tailored 5-session smoking cessation counselling programme for PHAs, which addressed depression, in combination with Nicotine Replacement Therapy (NRT) in a cohort of PHA smokers (n = 50). At 6-month follow-up, 28% of participants demonstrated biochemically verified abstinence from smoking. This result compares favourably to other quit-smoking intervention studies, particularly given the high percentage of HIV+ smokers with depression. At study baseline, 52% of HIV+ smokers scored above the clinical cut-off for depression on the Centre for Epidemiological Studies - Depression (CES-D) scale. HIV+ smokers with depression at study baseline demonstrated quantitatively lower depression at 6-month follow-up with a large effect size (d = 1), though it did not reach statistical significance (p = .058). Furthermore, those with depression were no more likely to relapse than those without depression (p = .33), suggesting that our counselling programme adequately addressed this significant barrier to smoking cessation among PHAs. Our pilot study indicates the importance of tailored programmes to help PHAs quit smoking, the significance of addressing depressive symptoms, and the need for tailored counselling programmes to enhance quit rates among PHAs.

Group psychotherapy for eating disorders: A meta-analysis.

OBJECTIVE: In the current meta-analysis, we review the effect of group psychotherapy compared to both wait-list controls and other active treatments for adults with eating disorders (EDs). METHOD: Twenty-seven randomized controlled trials (RCTs) that provide direct comparisons with a total of 1,853 participants were included. RESULTS: Group psychotherapy is significantly more effective than wait-list controls at achieving abstinence rates of binge eating and/or purging (RR = 5.51, 95% CI: 3.73, 8.12), decreasing the frequency of binge eating and/or purging (g = 0.70, 95% CI: 0.51, 0.90), and reducing related ED psychopathology (g = 0.49, 95% CI: 0.32, 0.66) after treatment. The effects of group psychotherapy and other active treatments (e.g., behavioral weight loss, self-help, individual psychotherapy) did not differ on any outcome at post-treatment or at follow-ups. Group cognitive behavioral therapy (CBT) and other forms of group psychotherapy did not differ significantly on outcomes at any time point. DISCUSSION: Additional research is needed to evaluate other group psychotherapy approaches, along with CBT, to provide more evidence-based treatment options for individuals with an ED. Group psychotherapy appears as effective as other common treatments and is perhaps more cost-effective than the most popular treatment, individual psychotherapy. Only 8.33% of comparisons in the current meta-analysis had at least 80% power to detect a moderate effect (d = .50) and we recommend that future RCTs be adequately powered.

A meta-analysis of the relation between therapeutic alliance and treatment outcome in eating disorders.

The therapeutic alliance has demonstrated an association with favorable psychotherapeutic outcomes in the treatment of eating disorders (EDs). However, questions remain about the inter-relationships between early alliance, early symptom improvement, and treatment outcome. We conducted a meta-analysis on the relations among these constructs, and possible moderators of these relations, in psychosocial treatments for EDs. Twenty studies met inclusion criteria and supplied sufficient supplementary data. Results revealed small-to-moderate effect sizes, ßs = 0.13 to 0.22 (p < .05), indicating that early symptom improvement was related to subsequent alliance quality and that alliance ratings also were related to subsequent symptom reduction. The relationship between early alliance and treatment outcome was partially accounted for by early symptom improvement. With regard to moderators, early alliance showed weaker associations with outcome in therapies with a strong behavioral component relative to nonbehavioral therapies. However, alliance showed stronger relations to outcome for younger (vs. older) patients, over and above the variance shared with early symptom improvement. In sum, early symptom reduction enhances therapeutic alliance and treatment outcome in EDs, but early alliance may require specific attention for younger patients and for those receiving nonbehaviorally oriented treatments.