The efficacy of psychological interventions for adult post-traumatic stress disorder following exposure to single versus multiple traumatic events: a meta-analysis of randomised controlled trials

Thole H. Hoppen, Richard Meiser-Stedman, Ahlke Kip, Marianne Skogbrott Birkeland, Nexhmedin Morina

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Background: Previous meta-analyses of psychological interventions for adult post-traumatic stress disorder (PTSD) did not investigate whether efficacy is diminished in individuals with PTSD related to multiple (vs single) traumatic events. We aimed to assess whether treatment efficacy would be lower in randomised controlled trials involving multiple-event-related PTSD versus single-event-related PTSD.  

Methods: For this meta-analysis, we searched PsycINFO, MEDLINE, Web of Science, and PTSDpubs from database inception to April 18, 2023. Randomised controlled trials involving adult clinical samples (≥70% meeting full PTSD criteria) with adequate size (≥10 participants per arm) were included. We extracted data on trial characteristics, demographics, and outcome data. Random-effects meta-analyses were run to summarise standardised mean differences (Hedges' g). Trials involving 100% of participants with single-event-related PTSD versus at least 50% of participants with multiple-event-related PTSD (ie, associated with ≥two traumatic events) were categorised. Quality of evidence was assessed using the Cochrane criteria. The review protocol was registered in PROSPERO (CRD42023407754).  

Findings: Overall, 137 (85%) of 161 randomised controlled trials were included in the quantitative synthesis, comprising 10 684 participants with baseline data and 9477 with post-treatment data. Of those randomly assigned, 5772 (54%) of 10 692 participants identified as female, 4917 (46%) as male, and three (<1%) as transgender or other. 34 (25%) of 137 trials exclusively involved women, 15 (11%) trials exclusively involved men, and the remainder were mixed samples. Mean age across the trials was 40·2 years (SD 9·0) ranging from 18·0 years to 65·4 years. 23 (17%) of 137 trials involved participants from low-income and middle-income countries (23 [17%] of 137). Data on ethnicity were not extracted. At treatment endpoint, psychological interventions were highly effective for PTSD when compared with passive control conditions in both samples with single-event-related PTSD (Hedges' g 1·04 [95% CI 0·77–1·31]; n=11; I2=43%) and multiple-event-related PTSD (Hedges’ g 1·13 [0·90–1·35]; n=55, I2=87%), with no efficacy difference between these categories (p=0·48). Heterogeneity between studies was substantial but outlier-corrected analysis yielded similar results. Moderate-sized effects were found compared with active control conditions with no significant difference between single-trauma and multiple-trauma trials. Results were robust in various sensitivity analyses (eg, 90% cutoff for multiple-trauma trials) and analyses of follow-up data. The quality of evidence was moderate to high.  

Interpretation: Contrary to our hypothesis, we found strong evidence that psychological interventions are highly effective treatments for PTSD in patients with a history of multiple traumatic events. Results are encouraging for clinical practice and could counteract common misconceptions regarding treatment and treatment barriers.
Original languageEnglish
Pages (from-to)112-122
Number of pages11
JournalThe Lancet Psychiatry
Volume11
Issue number2
Early online date11 Jan 2024
DOIs
Publication statusPublished - Feb 2024

Cite this