Portal de Periódicos da CAPES

Predictors of treatment outcome in individuals with eating disorders: A systematic review and meta-analysis

2016; Wiley; Volume: 49; Issue: 4 Linguagem: Inglês

10.1002/eat.22518

1098-108X

Eva Vall, Tracey Wade,

Behavioral Health and Interventions

This erratum incorporates data from two papers that were inadvertently omitted from analyses in the initial publication of the article, “PREDICTORS OF TREATMENT OUTCOME IN INDIVIDUALS WITH EATING DISORDERS: A SYSTEMATIC REVIEW & META-ANALYSIS” (Vall & Wade, 2015). Specifically, both studies investigated motivation as a predictor of outcome. The first (Geller, Cockell & Drab, 2001) examined drop-out from treatment, while the second (Geller, Drab-Hudson, Whisenhunt & Srikameswaran, 2004) considered outcomes at end of treatment and at 6-month follow-up. Both studies used the Readiness and Motivation Interview (RMI: Geller et al, 2001), which measures readiness and motivation to change eating disorder symptoms and behaviours from a ‘stages of change’ perspective, including pre-contemplation, contemplation, and action/maintenance. Each stage of change is measured across three separate symptom domains: dietary restriction, compensation and cognitive. The RMI also measures internality: the reasons for change as internal versus externally driven. Both studies are described in Table 1. In line with the method of the original manuscript, multiple effect sizes have been combined for each outcome. The updated meta-analytic outputs are given in Table 2. Overall, motivation as measured by the RMI was a small but significant predictor of drop-out. This effect was driven largely by the pre-contemplation subscale, as the other subscales (contemplation and action) were not significant predictors. The inclusion of these effects into the meta-analysis had no impact on the original outputs, with the mean effect size staying constant at r = .23. End of treatment outcomes included drive for thinness and body dissatisfaction. When grouped together, motivation as measured by two of the RMI subscales (pre-contemplation and action) was a small but significant predictor of outcome. This effect was driven by motivation as it relates to dietary restriction: motivation related to the other symptom domains (compensatory and cognitive) did not predict outcomes. The inclusion of this effect into the meta-analysis had little impact on the original outputs, with the mean effect size dropping from r = (0).26 to r = .25. Follow-up outcomes at 6-months post-treatment included drive for thinness, body dissatisfaction, and for underweight patients a drop in body mass index (BMI) to below 18.5. There were no predictors of body dissatisfaction, but motivation as measured by two of the RMI subscales (pre-contemplation and internality related to dietary restriction) predicted drive for thinness. Two measures predicted change in BMI, namely, pre-contemplation for compensatory symptoms and internality for dietary restriction. Following the inclusion of these effects into the meta-analysis, the mean effect size increased from r = .15 to r = .20. Following the inclusion of these two studies to the original meta-analysis, motivation continues to be classed as a modest predictor of outcome in the eating disorders. The results of these two studies highlight the variability in predictive value depending on which facet of motivation (and which outcome measure) is examined. Ongoing investigation into motivation as it relates to treatment outcomes in the eating disorders is needed.