Trauma-focused therapy holds up in low- and middle-income settings

Design

Bayesian three-level random-effects meta-analysis of randomised controlled trials, PROSPERO CRD42024584333

Population

5,041 adults with PTSD symptoms across 55 RCTs in low- and middle-income countries, 1980-2025

Primary outcome

PTSD symptom severity at post-assessment and follow-up

Effect

Between-group g = -1.31 (95% CrI -1.66 to -0.95); within-group g = -1.88 (-2.25 to -1.51); both sustained at follow-up

Guidelines putting trauma-focused psychological therapy first for PTSD rest largely on trials done in high-income countries, and the question of whether that transfers has been answered mostly by assertion. This Bayesian meta-analysis pooled 55 randomised trials and 5,041 adults with PTSD symptoms in low- and middle-income countries, drawn from the PTSD Trials Standardized Data Repository with additional searching of Embase, Medline, PsycINFO and PTSDpubs, covering 1980 to June 2025. Effects were aggregated with three-level Bayesian random-effects meta-regression and quality assessed with RoB 2.

The effects were large: within-group g = -1.88 (95% credible interval -2.25 to -1.51) and between-group g = -1.31 (95% CrI -1.66 to -0.95) for PTSD symptoms at post-assessment, both sustained at follow-up. Depression and anxiety improved by medium to large margins, also sustained. Treatments did best when they were trauma-focused, delivered face to face, and given to older participants. Sensitivity analyses across different prior specifications held.

For Indian practice the useful part is the direction of the evidence rather than the magnitude. It supports prioritising trauma-focused work over generic supportive counselling, and it does not support the assumption that remote delivery is an equivalent substitute in these settings.

• Prioritise trauma-focused modalities over generic supportive counselling where you have a choice
• Face-to-face delivery outperformed remote in this pooled analysis — weigh that before defaulting to teletherapy
• Effects held at follow-up, so this is not a transient post-treatment bump
• The evidence base is thinnest in low-income and lower-middle-income countries specifically
• Depression and anxiety improved too; a single trauma-focused course may address more than one target

Why it matters

The first-line status of trauma-focused therapy in these settings has been an extrapolation until now.

Don't overread it

Comparisons between intervention types come from meta-regression across studies, not from randomised head-to-head trials.

The statistics, in plain English

A between-group Hedges' g of -1.31 is a very large effect, and effects this size in a psychotherapy meta-analysis usually reflect some combination of real benefit, small trials, and waitlist comparators that improve very little. The credible interval (-1.66 to -0.95) is narrow enough that the direction is not in doubt; the size probably is. 'Trauma-focused did best' is a meta-regression finding across trials, which is weaker evidence than a head-to-head comparison.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.