Selective accession waivers for medically disqualifying conditions like spinal curvature are one way the military meets its manpower needs. We evaluated retention patterns during the fi rst 2 years of service of a cohort of military recruits with waivers for pathological curvature of the spine (spinal curvature). Recruits waived for spinal curvature (n = 417), who accessed from 1998 to 2005 were identifi ed and matched with 3 qualifi ed recruits. Kaplan-Meier survival analysis and Cox proportional hazards model were used to compare survival patterns and adjusted attrition hazard estimates. Waived recruits experienced signifi cantly increased risk of "all cause" discharge (relative risk = 1.3; 95% confi dence interval: 1.1, 1.5) and "existing prior to service" discharge (relative risk = 2.4; confi dence interval: 1.6, 3.5). Despite the increased risk of discharge, current waiver criteria allowed a majority with spinal curvature to complete at least 2 years of service. Policy makers must consider risks and benefi ts before modifying the current accession standard for spinal curvature.