The objective of this longitudinal study was to identify risk factors for combat-related psychiatric disorders. The sample consisted of 6442 enlisted U.S. Marines who completed a questionnaire during basic training, deployed to a combat zone with no prior psychiatric diagnoses, and completed a postdeployment assessment form. Cox proportional hazards regression was used to determine associations between predeployment and postdeployment self-reports and subsequent mental health outcomes. During the observation period, 6.8% of the sample were diagnosed with a psychiatric disorder. The strongest predictors of postdeployment psychiatric disorders were, in order of importance, low paygrade, hospitalization during deployment, low education, preservice smoking, and post-traumatic stress disorder symptoms at deployment's end. The impact of war zone variables was smaller than expected. It was recommended that the combat experience section of the military's postdeployment assessment form be expanded to enhance the military's ability to identify and refer personnel who may be at risk for psychiatric disorders.