Background Understanding the natural course of child and adolescent posttraumatic stress disorder (PTSD) has significant implications for the identification of, and intervention for, at‐risk youth. We used a meta‐analytic approach to examine longitudinal changes in youth PTSD prevalence and symptoms over the first 12 months posttrauma. Methods We conducted a systematic review to identify longitudinal studies of PTSD in young people (5–18 years old), excluding treatment trials. The search yielded 27 peer‐reviewed studies and one unpublished dataset for analysis of pooled prevalence estimates, relative prevalence reduction and standardised mean symptom change. Key moderators were also explored, including age, proportion of boys in the sample, initial prevalence of PTSD and PTSD measurement type. Results Analyses demonstrated moderate declines in PTSD prevalence and symptom severity over the first 3–6 months posttrauma. From 1 to 6 months posttrauma, the prevalence of PTSD reduced by approximately 50%. Symptoms also showed moderate decline, particularly across the first 3 months posttrauma. There was little evidence of further change in prevalence or symptom severity after 6 months, suggesting that it is unlikely a child would lose a PTSD diagnosis without intervention beyond this point. Conclusions The current findings provide key information about the likelihood of posttrauma recovery in the absence of intervention and have important implications for our understanding of child and adolescent PTSD. Results are discussed with reference to the timing of PTSD screening and the potential role of early interventions. Findings particularly highlight the importance of future research to develop our understanding of what factors prevent the action of normal recovery from the ‘acute’ posttrauma period.