The aim of the current study was to investigate whether factors that predict acute postconcussion syndrome (PCS) are associated with persistent PCS, and how change in these predictors over time is related to change in PCS diagnosis. Prospective consecutive trauma admissions to a Level 1 trauma hospital were assessed a mean 4.88 (SD=2.72) days and again 106.23 (SD=14.85) days posttrauma. The final sample comprised 62 mild traumatic brain injury and 58 non-brain injured trauma controls. Multilevel logistic regression models were used to examine whether acute predictors were associated with persistent PCS, and to test associations between patient characteristics and the effect of time. We found that a diagnosis of PCS was not specific to mTBI either acutely or at 3-month assessment. Acute predictors of PCS (premorbid psychiatric disorder, and acute posttraumatic stress) predicted persistent PCS. A diagnosis of PCS for males and females differed over time. With increasing severity of PTSD symptoms the likelihood of PCS increased. Pain was associated with persistent PCS. If participants had both pain and PTSD they were 3.09 times more likely to have persistent PCS (95% CI: 1.14-8.40). We conclude that early intervention may prevent the development of subsequent PCS.