Meta-analysis on drugs in people with eating disorders.

OBJECTIVE The aim of this study was to examine whether drug use (DU) is higher in people with eating disorders (EDs) compared to a healthy control group and to perform a meta-analysis on the literature related to DU in people with EDs. METHOD We searched electronic databases (Medline, PsycINFO, Web of Science and CINAHL) and reviewed studies published from 1994 to August, 2007, in English, German or Spanish. A total of 16 papers fulfilled the inclusion criteria and were included. RESULTS The general meta-analysis revealed a negligible albeit significant effect size (0.119, p < .05). Risk was found to be higher in bulimia nervosa (BN, delta = 0.462, p = < .001), smaller in binge eating disorder (delta = 0.14, p < .05) and non-significant in anorexia nervosa (AN, delta = -.167, p = .070). CONCLUSIONS The differential risk observed in patients with BN might be related to differences in temperament or might be the result of reward sensitization.