1 in 5 Trauma-Exposed Children Met PTSD Criteria in 2025 Meta-Analysis

TL;DR: A 2025 The British Journal of Psychiatry meta-analysis of 95 studies found about 1 in 5 trauma-exposed youth met older DSM-IV criteria for PTSD and 1 in 8 met newer DSM-5 criteria, with risk concentrated in girls and interpersonal trauma.

Source: The British Journal of Psychiatry (2025) | Visser et al.

The most important number in trauma care is not how many children experience something terrible. It is how many still meet full PTSD criteria after the immediate chaos has passed.

This paper updates that estimate with three-level meta-analysis — and, just as importantly, shows how much the answer shifts depending on the trauma type and the diagnostic instrument.

One in Five DSM-IV, One in Eight DSM-5

Traumatic exposure is common in childhood; PTSD is not automatic. That distinction can hide the clinical burden. Even when most exposed children recover, this meta-analysis suggests the affected minority is large enough to demand routine follow-up, not passive watching.

The authors updated the field’s last major prevalence estimate with a much broader evidence base and a three-level random-effects model that better accounts for dependence within studies. The result is sturdier than a simple pooled average.

Under DSM-IV criteria, pooled prevalence was 20.3%. Under DSM-5, the pooled estimate fell to 12.0% — but the evidence base there was much thinner and the confidence interval was wide enough to leave real uncertainty.

The temptation is to read the 20.3% vs. 12.0% gap as proof DSM-5 PTSD is simply less common.

The safer reading is narrower.

DSM-5 changed the symptom structure, added developmental nuance, and dropped some older criteria, so the frame is not identical.

The DSM-5 estimate came from only eight samples; each study carries much more weight in the pooled number. The paper is explicit that there were not enough DSM-5 samples to run the same moderator analyses — which means the questions clinicians actually care about (who is at highest risk, after which trauma, under which measurement conditions) were mostly answerable only in the DSM-IV set.

Interpersonal Trauma Was the Risk Multiplier

The clearest moderator was trauma type. In the DSM-IV analysis, children exposed to interpersonal trauma — assault, abuse, other human-caused violations — had a pooled PTSD prevalence of 32.1%. Children exposed to non-interpersonal trauma like accidents or disasters: 11.8%.

That gap is too large to shrug off as statistical noise. The meaning and context of the trauma, not just the presence of fear, changes the psychiatric aftermath. Interpersonal trauma often carries betrayal, ongoing threat, shame, secrecy, or disruption inside the home — ingredients that make recovery much harder than recovery from a frightening but bounded event.

The risk pattern was easier to read once moderators were separated:

• Interpersonal trauma: markedly higher PTSD prevalence than after accidents or disasters.
• Gender mix: samples with fewer boys showed higher PTSD prevalence — consistent with greater risk among girls.
• Time since trauma: prevalence did not significantly differ between children assessed within 3 months and those assessed later in DSM-IV. Time alone did not produce recovery in this finding.

The Interview Tool Swung Prevalence Sevenfold

One of the most underappreciated findings is the spread across diagnostic interviews. In the DSM-IV analysis, prevalence ranged from 8.8% with DICA to 62.3% with CPTSDI, with other instruments like CAPS-CA and K-SADS landing in between.

That does not necessarily mean one interview is wrong and another is right. It means prevalence is partly a measurement problem.

Different interviews emphasize different thresholds, symptom wording, informants, and developmental interpretations. If a school district, pediatric trauma clinic, or emergency follow-up program wants to compare its PTSD rates to the literature, the diagnostic tool matters much more than most reports acknowledge.

The implication for families is also important. A child does not become less symptomatic because one instrument is stricter.

From a service-planning view, instrument choice determines how many children qualify for specialized follow-up and how many are counted as “subthreshold” despite real impairment. A prevalence number without the interview method attached is less informative than it looks.

Post-Trauma Screening Should Include Children Without Classic PTSD Referral Signals

Clinical systems should stop assuming the base rate is trivial. If one in five trauma-exposed youth met DSM-IV PTSD criteria across the pooled literature, the burden is large enough that passive watch-and-wait approaches will miss a lot of children.

Screening has to be staged intelligently. Immediate distress after trauma is expected, but persistent intrusive memories, avoidance, hyperarousal, negative mood, or functional impairment need follow-up — not reassurance alone. The meta-analysis supports a tiered approach: universal early check-ins after serious trauma, closer monitoring after interpersonal trauma, and diagnostic interviews when symptoms persist or impair school, sleep, family life, or safety.

The paper especially strengthens the case for targeted follow-up after interpersonal trauma and a lower threshold for screening girls and adolescents who remain impaired even when the event is no longer immediate. It also argues for transparency in how PTSD is assessed.

The DSM-5 evidence base is still sparse, the studies vary by trauma context and country, and prevalence is not the same as chronicity or functional impact — but the update changes the center of gravity. The right starting assumption is no longer “PTSD in kids is relatively rare after trauma.” The better assumption is: many children recover, but the minority who do not is large enough to demand systematic screening and better measurement.