Staffan Sohlberg’s research while affiliated with Uppsala University and other places

The study investigated initial self-image (structural analysis of social behavior) and its relation to 36-month outcome, among patients with anorexia nervosa and bulimia nervosa. Hypotheses were that degree of different aspects of self-image would predict outcome in the groups. Participants were 52 patients with anorexia and 91 with bulimia from a longitudinal naturalistic database, and outcome measures included eating disorder and psychiatric symptoms and a general outcome index. Stepwise regression was used to investigate which self-image variables were related to outcome, and multiple regression contrasted the groups directly on each obtained predictor. Consistent with hypotheses, in bulimia degree of self-hate/self-love moderately predicted outcome, whereas self-control-related variables powerfully predicted outcome in anorexia. It is important to focus on self-image in the treatment of both diagnostic groups, but especially in anorexia nervosa, where control-submission interactions between patient and therapist should be handled with care.

There is a lack of knowledge about the outcome of eating disorder patients who terminate treatment prematurely. The present study followed-up eating disorder patients who had previously dropped out of treatment and examined clinical status 36 months after intake. Dropouts (n = 30) were compared with treatment completers (n = 52) on diagnostic status, clinical symptoms, psychosocial adjustment and treatment satisfaction at follow-up. Patterns of change from intake to follow-up within groups, as well as between groups, were explored. No significant differences were found between groups at follow-up, except for more treatment dissatisfaction reported among dropouts. When patterns of change were examined between groups, patients who completed treatment were found to have made significantly greater changes (less eating disorder symptoms, less psychological problems and more positive self-image) compared to dropouts. Although no significant differences in outcome were found between dropouts and completers, greater clinical improvement was found among those who completed treatment. The dropouts examined in this study did well despite premature termination of treatment. Clinical and research implications are discussed.

Drop-out from treatment is a serious problem in eating disorders which remains poorly understood. The present study investigated whether self-image and interpersonal theory could help to explain why eating disorder patients drop out of treatment. Intake data on eating disorder patients who terminated treatment prematurely (N=54) were compared with patients who had completed treatment (N=54) and those who were still in treatment after 12 months (N=54). Self-image was assessed using the structural analysis of social behaviour (SASB), and comparisons were made on demographic and clinical variables. Patients who dropped out had initially presented with less negative self-image and fewer psychological problems compared with remainers. Low levels of SASB self-blame discriminated drop-outs from completers and remainers and significantly predicted treatment drop-out. Drop-out in eating disorders appears to be a complex phenomenon, not necessarily as pathological as often assumed. There may be important differences in the treatment goals of drop-outs and therapists; patients who drop out may be choosing to disengage at a time when symptom improvement creates space for closer examination of interpersonal issues.

Persistent and non-trivial effects of unconscious stimuli have been reported (Sohlberg & Birgegard, 2003). This raises the ethical question of whether informing participants about such stimuli effectively returns them to a normal state. Two experiments (sex-mixed, N = 70 and 118) tested two kinds of debriefing to participants following subliminal (tachistoscopic) attachment-related or control stimulation. Measures were the Beck Depression Inventory, Structural Analysis of Social Behavior, and Retrospective Attachment. Results showed persistent effects. Simple debriefing about the stimulus was effective in preventing these, while more elaborate debriefing also describing the effects and mechanisms for them was less effective. Persistent effects were also strongly related to participant sex, but debriefing effects were similar in men and women with regard to the purpose of debriefing. The findings have implications for ethical recommendations for subliminal research, and suggest that this unexplored area requires more attention.

Negative self-image has been hypothesised to be of aetiological significance in eating disorders; however, its relationship to outcome remains unclear. The present study examined the relationship between self-image and follow-up status in a heterogeneous sample of eating disorder patients (N=246). Patients were assessed at intake and after 36 months. Self-image was measured using SASB, and a general outcome measure was calculated comprising eating disorder symptoms, occupational status, interpersonal relationships and general psychopathology. SASB self-hate was the most important variable for predicting poor outcome, followed by occupational status, interpersonal relationships, eating disorder symptoms, SASB self-emancipation and general psychiatric symptoms. Together these variables predicted 23% of the variance in outcome. High levels of self-hate may increase the risk of poor outcome in eating disorders by adversely affecting interpersonal relationships and making it difficult for patients to engage in treatment.

In two experiments we explored Tomkins's (1963) concept of shame, comparing conscious versus unconscious shame activation. In line with Tomkins' theory, an impeded positive feedback sequence, consciously or unconsciously perceived, elicited more shame than continuously negative feedback. This was, however, true only for participants with an initially low degree of internalized shame. Participants with a high degree of internalized shame unexpectedly displayed shame following the positive feedback intended to elicit positive emotion. Whether this has implications for Tomkins's theory or rather for methodological issues is discussed. Exploring consequences of shame for social perception and self-image, we found reversed results depending on level of consciousness. Effects were consistently greater for women, although at marginal significance levels. Results partly support Tomkins's notion of shame, but imply that his theory might need modification in terms of the role played by consciousness and, possibly, by individual differences such as sex and shame proneness.

Important, but not unequivocal, support has been found for Sociotropy and Autonomy as vulnerability factors for depression. Using a novel approach we gave subliminal sociotropy-related and autonomy-related visuoverbal stimulation to nondepressed students (Study 1, N=36), and patients treated for clinical depression (Study 2, N=34) in completely balanced within-groups designs. Autonomy, as measured with the Personal Style Inventory, was unrelated to response. Sociotropy predicted negative outcome after a sociotropy-related abandonment stimulus in both Study 1 (r=−.34 to a decrease in positively valenced autobiographical memories) and Study 2 (r=.38 to an increase in selection of depression-related words on a word-selection task). This finding may interest developers of depression therapies, and support further attempts to study Sociotropy as a putative vulnerability factor for depression.

The practice of statistical inference in psychological research is critically reviewed. Particular emphasis is put on the fast pace of change from the sole reliance on null hypothesis significance testing (NHST) to the inclusion of effect size estimates, confidence intervals, and an interest in the Bayesian approach. We conclude that these developments are helpful for psychologists seeking to extract a maximum of useful information from statistical research data, and that seven decades of criticism against NHST is finally having an effect.

The present study explored treatment satisfaction among eating disorder patients participating in a Swedish multicentre study (N ¼ 469) who had been followed up 36 months after initial assessment. Satisfaction was measured using a questionnaire focusing on initial reception at the treatment unit, suitability of treatment, ability of staff to listen and understand, confidence in the unit and agreement on treatment goals. After 36 months, 38% of patients were 'highly satisfied' with treatment, while 39% were 'satisfied' and 23% were 'unsatisfied'. Satisfaction was predicted by treatment interventions focusing on support and increased control of eating problems. Although unsatisfied patients were characterized by significantly higher levels of eating disorder psychopathology and psychiatric symptoms as well as more negative interpersonal profiles at follow-up, these patients had been virtually indistinguishable from the other groups at initial assessment, with a few notable exceptions. Unsatisfied patients had previously been less prepared to change their eating habits, had higher levels of conflict with their fathers, had a lesser degree of present weight acceptance and tended to expect less from treatment interventions focusing on control of eating problems. It is concluded that, although we may be successful in engaging and satisfying patients who are predisposed to therapeutic strategies focusing on support and control of eating problems, we may be less successful with those who have a greater number of interpersonal problems and who are not as focused on their symptoms of disordered eating. Contract/grant sponsors: The Swedish Foundation for Health Care Sciences and Allergy Research; The Foundation Sö derströ m-Kö nigska Nursing Home; The PKF Foundation.

Although evidence suggests that interpersonal psychotherapy may be an efficacious treatment for eating disorders, there is surprisingly little systematic knowledge about the interpersonal world of these patients. SASB self-image ratings were used to explore interpersonal profiles in a large heterogeneous sample of eating disorders (N = 830), matched normal controls (N = 105) and a small group of controls with subclinical depression (N = 26). Eating disorder patients clearly presented with significantly more negative interpersonal profiles compared to controls. Within the eating disorder group, anorexics were characterized by high self-control, self-blame and self-attack. Patients with binge eating disorder expressed the least negative self-image, and were significantly more self-affirming than bulimics and less self-controlling than patients with atypical eating disorders. Eating disorder patients may have distinct interpersonal profiles that increase the risk of negative therapeutic reaction. Better knowledge of interpersonal processes in eating disorders may help to improve both diagnostic assessment and treatment.