This paper provides econometric evidence on the prevalence and childhood antecedents of Posttraumatic Stress Disorder (PTSD) using data from a large-scale cross-sectional survey of Vietnam-Era veterans. The empirical strategy accounts for potential errors in the survey PTSD diagnosis using a method that is particularly suitable for secondary analysis of public-use surveys of the health of the general population. The intensity of a veteran’s reactions to war-time stress is higher for middle-born veterans, for veterans whose parents have a history of mental illness and for veterans reared in, what was then considered, nontraditional households. The latter two factors also substantially raised the probability of a positive diagnosis. These findings are consistent with much of the psychological literature on the relationship between family psychopathology and vulnerability to the disease. They are also consistent with much of household economics literature on the relationship between family structure and a person’s achievements. Adjusted for PTSD classification errors, the sample PTSD prevalence rate is high, 18%, but roughly 4 percentage points lower than the unadjusted rate. This result is consistent with recent studies of the biases in PTSD prevalence estimates.