Objectives: The study conducted a longitudinal assessment of insomnia as an antecedent versus consequence of posttraumatic stress disorder (PTSD) and depression symptoms among combat veterans. Design: Two postdeployment time points were used in combination with structural equation modeling to examine the relative strength of two possible directions of prediction: insomnia as a predictor of psychological symptoms, and psychological symptoms as a predictor of insomnia. Participants were active duty soldiers (N = 659) in a brigade combat team who were assessed 4 months after their return from a 12-month deployment to Iraq, and then again eight months later. Results: Although both insomnia and psychological symptoms were associated at both time periods and across time periods, insomnia at 4 months postdeployment was a significant predictor of change in depression and PTSD symptoms at 12 months postdeployment, whereas depression and PTSD symptoms at 4 months postdeployment were not significant predictors of change in insomnia at 12 months postdeployment. Conclusions: Results support the role of insomnia in the development of additional psychological problems and highlight the clinical implications for combat veterans, to include the importance of longitudinal assessment and monitoring of sleep disturbances, and the need for early intervention. © 2011 Wiley Periodicals, Inc. J Clin Psychol 67:1–19, 2011.