Moffitt's developmental taxonomy of adolescence-limited and life-course-persistent offenders has received much empirical attention, with researchers focusing on the etiology and trajectory of offending between the two groups. Recently, Moffitt articulated a new hypothesis that has yet to be empirically assessed--that life-course-persistent offenders will be at high risk in midlife for poor physical and mental health, cardiovascular disease, and early disease morbidity. Using data from the Baltimore portion of the National Collaborative Perinatal Project, a longitudinal study of several thousand individuals followed from birth to ages 27 to 33, the authors test this hypothesis. We find that, compared to adolescence-limited offenders, life-course-persistent offenders are more likely to experience adverse physical and mental health outcomes. We also find that life-course-persistent offenders are more likely than their counterparts to be involved in antisocial lifestyles, which in turn increase the chances of adverse health outcomes. Future theoretical and empirical research directions are identified. (Contains 5 tables and 8 notes.)