Introduction: Social support’s association with posttraumatic stress disorder (PTSD) in Veterans is well established. One explanation for this link is social causation – support inhibits PTSD. Inversely, within the social erosion model, PTSD erodes support. The aim of the present study was to examine if the social causation or social erosion model better explained the association between support and PTSD within a psychosocial intervention context. Methods: Veterans ( N = 218) participating in a multimodal transition program were assessed pre-program, post-program, and at 3-month follow-up on their perceived social support and PTSD symptoms. We used path analysis to conduct a three-wave cross-lagged panel model to compare the social erosion and social causation models. Results: PTSD symptoms were associated with attenuated improvements in social support, while social support was not associated with increased reductions in PTSD symptoms. This association was observed from pre- to post-program and from post-program to follow-up. Discussion: These findings support the social erosion model over the social causation model. Clinical implications of PTSD inhibiting interpersonal gains are discussed.