Abstract
Objective
Identifying modifiable predictors of outcomes following treatment for eating disorders may help to tailor interventions to patients’ individual needs, improve treatment efficacy, and develop new interventions. The goal of this meta‐analysis was to quantify the association between pretreatment motivation and posttreatment changes in eating disorder symptomology.
Method
We reviewed 196 longitudinal studies reporting on change on indices of overall eating‐disorder symptomatology, weight gain, binge‐eating, vomiting, anxiety/depression, and treatment adherence. Meta‐analyses were performed using two complementary approaches: (a) combined probability analysis using the added Z‘s method; (b) effect size analyses. Using random‐effect models, effect sizes were pooled when there were at least three studies with the same type of statistical design and reporting statistics on the same outcome. Heterogeneity in study outcome was evaluated using Q and I
2 statistics. Studies were reviewed qualitatively when the number of studies or reported data were insufficient to perform a meta‐analysis.
Results
Forty‐two articles were included. Although samples and treatments differed substantially across studies, results across studies were remarkably consistent. Both combined‐probability and effect‐size analyses indicated positive effects of pretreatment motivation on improvement in general eating‐disorder symptoms (Cohen’s r = .17), and an absence of effects on anxiety/depression symptoms. Remaining outcome indices were subject to selective reporting and/or small sample size bias.
Discussion
Our findings underscore the importance of incorporating treatment engagement approaches in the treatment of eating disorders. Optimal reporting of study findings and improving study quality would improve future efforts to obtain an in‐depth understanding of the relationship between motivation and eating disorder symptoms.