Impulsive, undercontrolled personalities and major family stresses are known predictors of impaired adjustment, but long-term health effects are unclear. In an archival prospective cohort design, we followed up on L. M. Terman's (Terman & Oden, 1947) sample of gifted children by collecting and coding death certificates for the half of the sample that is now dead. Statistical survival analyses were used to predict longevity and cause of death as a function of parental divorce during childhood, unstable marriage patterns in adulthood, childhood personality, adult adjustment, and possible mediating health behaviors. Psychosocial factors emerged as important risks for premature mortality.