Purpose of reviewTo review the recent literature about the epidemiology, course, and outcome of eating disorders in accordance with the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).Recent findingsThe residual category eating disorder not otherwise specified' (EDNOS) was the most common DSM-IV eating disorder diagnosis in both clinical and community samples. Several studies have confirmed that the DSM-5 criteria for eating disorders effectively reduce the proportion of EDNOS diagnoses. The lifetime prevalence of DSM-5 anorexia nervosa among women might be up to 4%, and of bulimia nervosa 2%. In a cross-national survey, the average lifetime prevalence of binge eating disorder (BED) was 2%. Both anorexia nervosa and bulimia nervosa are associated with increased mortality. Data on long-term outcome, including mortality, are limited for BED. Follow-up studies of BED are scarce; remission rates in randomized controlled trials ranged from 19 to 65% across studies. On a community level, 5-year recovery rates for DSM-5 anorexia nervosa and bulimia nervosa are 69 and 55%, respectively; little is known about the course and outcome of BED in the community.SummaryApplying the DSM-5 criteria effectively reduces the frequency of the residual diagnosis EDNOS, by lowering the threshold for anorexia nervosa and bulimia nervosa, and adding BED as a specified eating disorder. Course and outcome studies of both anorexia nervosa and bulimia nervosa show that no significant differences exist between DSM-5 and DSM-IV definitions.