The overall age- and sex-adjusted incidence rate of 7.2 per 100,000 population for anorexia nervosa in Rochester, Minnesota, was recently obtained from a study covering the years 1935 through 1979. Using the incident cases identified for that study, as well as those for the years 1980 through 1984, a case-control study using medical record data will be carried out. The proposed study will use a controls two individuals matched by age and sex and seen at the same time and place as the case. This study will attempt to define risk factors for the condition; information at present is sadly lacking in the medical literature.