Objective: To investigate if baseline shape and weight concern (SWC) moderated outcomes in Prevention Across the Spectrum, a randomized-controlled trial (RCT) of 3 school-based programs aimed at reducing eating disorder and obesity risk factors. Method: N = 1,316 Grade 7 and 8 girls and boys (M age = 13.21 years) across three Australian states were randomly allocated to: Media Smart; Life Smart; Helping, Encouraging, Listening and Protecting Peers Initiative (HELPP) or control (usual school class). Moderation was explored by testing interaction effects for group (Media Smart; Life Smart; HELPP; Control) × moderator (SWC: higher-SWC; lower-SWC) × time (post-program; 6-month follow-up; 12-month follow-up), with baseline risk factor scores entered as covariates. Results: Moderation effects were found for shape concern, weight concern, eating concern, regular eating (i.e., meal skipping), physical activity, body dissatisfaction, dieting, and perfectionism. Post-hoc testing found eating concern at post-program was the only variable where higher-SWC Media Smart participants experienced a reduction in risk relative to controls. Both higher-SWC Life Smart and HELPP participants reported an increase in eating concern relative to controls and both groups were skipping more meals than controls at 12-month follow-up. Amongst lower-SWC participants, Media Smart was the only group to experience a benefit relative to controls (physical activity). Conclusions: This study highlights the need for moderator analyses to become more routinely conducted in universal trials, to ensure that participants across baseline risk levels are benefiting and not harmed from program participation.