Survival analysis, a methodology employed in studies of survival after diagnosis of a life-threatening physical illness, is used to examine psychiatric recidivism. The probability of survival in the community without rehospitalization was calculated for 400 consecutive admissions to Thresholds, a psychiatric rehabilitation center serving formerly hospitalized patients. It was found that the probability of a client's being rehospitalized at least once within 15 months of Thresholds intake was .55 and that the probability of rehospitalization varied with sex, race, age, and number of prior hospitalizations. The authors note that early intervention may be indicated for clients at higher risk of rehospitalization. They suggest that survival analysis be used in studies of psycbiatric recidivism because it permits direct statistical comparisons between studies.