LAUSR

Background Anorexia nervosa (AN) is a psychiatric disorder with high mortality. Method A retrospective register study of 609 males who received hospitalized care for AN in Sweden between 1973 and 2010 was performed. The standardized mortality ratios (SMRs) and Cox regression-derived hazard ratios (HRs) were calculated as measures of mortality. The incidence rate ratios (IRRs) were calculated to compare the mortality rates in patients with AN and controls both with and without psychiatric diagnoses. Results The SMR for all causes of death was 4.1 [95% confidence interval (CI) 3.1–5.3]. For those patients with psychiatric co-morbidities, the SMR for all causes of death was 9.1 (95% CI 6.6–12.2), and for those without psychiatric co-morbidity, the SMR was 1.6 (95% CI 0.9–2.7). For the group of patients with alcohol use disorder, the SMR for natural causes of death was 11.5 (95% CI 5.0–22.7), and that for unnatural causes was 35.5 (95% CI 17.7–63.5). The HRs confirmed the increased mortality for AN patients with psychiatric co-morbidities, even after adjusting for confounders. The IRRs revealed no significant difference in mortality patterns between the AN patients with psychiatric co-morbidity and the controls with psychiatric diagnoses, with the exceptions of alcohol use disorder and neurotic, stress-related and somatoform disorders, which seemed to confer a negative synergistic effect on mortality. Conclusion Mortality in male AN patients was significantly elevated compared with the general population among only the patients with psychiatric co-morbidities. Specifically, the presence of alcohol and other substance use disorders was associated with more profound excess mortality.