checkedPsychopharmacology and Psychiatry Updates2026-11-02last verified 2026-11-03

were seven of ten placebo-controlled prazosin trials in ptsd positive?

verdict: overstated
directionally right; magnitude or certainty inflated
“Seven of the ten were positive.”

a guest, Dr. David Osser, psychiatrist treating PTSD in the VA system said on Psychopharmacology and Psychiatry Updates, 2026-11-02, at 03:56. hear it in context →

the episode: Prazosin for PTSD Nightmares (at 03:56). we link the show's own page; we do not re-host audio.

first, the part that holds

the count of ten is about right: a 2025 meta-analysis found ten placebo-controlled prazosin trials in PTSD, and more of them favored prazosin than not. pooled across trials, prazosin still beats placebo on nightmares.

what we found when we went and looked

we found a 2025 systematic review and meta-regression that set out the same ten-trial universe. its own table counts 6 of 10 trials as favoring prazosin. the authors add a footnote saying one of those six (an 8-person trial in women with bulimia and PTSD) was not significant in their reanalysis, which makes it 5 of 10. you can get to "seven" by adding a 2012 three-arm trial that the review left out. but that trial beat placebo on insomnia and daytime symptoms, not on nightmares, and 42% of its participants had subthreshold PTSD. that also turns the count into 7 of 11, not 7 of 10. the four or five null trials include the largest one by far: a 304-person VA trial published in 2018 that found no difference on nightmares, sleep quality or global improvement.

assertion by assertion

assertionverdictwhat the record says
there have been ten placebo-controlled trials of prazosin in PTSDconfirmeda 2025 meta-analysis included exactly ten placebo-controlled RCTs (648 patients total). a 2012 three-arm trial in veterans, 58% with full PTSD, fell outside its criteria. counting that one too makes eleven, so 'ten' depends on which inclusion rules you use. [1][4]
seven of those ten trials were positiveoverstatedin the 2025 review's ten-trial set, the authors described 6 trials as favoring prazosin: three small veteran/soldier trials (2003, 2007, 2013), a 2008 civilian trial, a 2014 trial in Iran, and a 2023 trial of 8 women. their footnote says the 2023 trial was not significant in their reanalysis, which leaves 5. the other four (2015 and 2016 trials in alcohol use disorder plus PTSD, a 2018 trial in suicidal patients, and the 2018 VA trial) showed no benefit over placebo. no ten-trial list we read reaches seven. [1][4]
the largest trial was negativeconfirmedthe 2018 VA cooperative trial (n=304) found no difference between prazosin and placebo. difference on the nightmare item: 0.2 (95% CI -0.3 to 0.8, p=0.38). sleep quality: 0.1 (p=0.80). global improvement: 0 (p=0.96). this one trial is bigger than all the positive trials in the 2025 review combined. [3]
pooled evidence still shows a benefit on nightmaresconfirmeda 2020 meta-analysis of 8 RCTs (286 vs 289) found a nightmare benefit (SMD -1.13, 95% CI -1.91 to -0.36), but no significant benefit for overall PTSD (SMD -0.45, CI -0.95 to 0.05) or sleep quality. the 2025 meta-analysis found benefits for nightmares (SMD -0.641, p=0.025) and insomnia (SMD -0.654, p=0.043), but not for overall PTSD (p=0.077). [1][2]

both directions

for the speaker: trials favoring prazosin do outnumber null trials in every list we read: 5 or 6 of 10 in the 2025 review, and up to 7 of 11 if you add the 2012 three-arm trial. several of the early positive trials had large effects. both meta-analyses we read still find a statistically significant pooled benefit on nightmares, so the speaker's broader point that prazosin has real positive evidence holds up.

against: no independently assembled ten-trial list we read gets to seven positives. counting trials as positive or negative treats an 8-person crossover the same as a 304-person multisite trial. the positive trials were all small, with 8 to 67 analyzed, and the largest and most rigorous trial was clearly null. pooled across trials, prazosin does not significantly beat placebo on overall PTSD symptoms, and one meta-analysis found no significant benefit on sleep quality. "positive" also shifts from trial to trial: the 2012 trial was positive on insomnia but not on nightmares, which is what prazosin is used for.

what this cannot say

the speaker named no list of trials, so we cannot tell which ten were counted or what "positive" meant (primary outcome, any sleep outcome, or a secondary outcome such as drinking). a different but defensible list could produce seven. our count rests on one 2025 review's classification and our reading of one trial it left out. we read abstracts and the summary tables, not every full trial report. vote-counting is a crude way to summarize evidence, and the pooled effect sizes say more than the tally does.

sources, each one fetched and read

  1. Factors impacting prazosin efficacy for nightmares and insomnia in PTSD patients: a systematic review and meta-regression analysis. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 2025;136:111253. doi:10.1016/j.pnpbp.2025.111253 What we read there: ten RCTs (648 patients). the table marks the 2003, 2007, 2013, 2008, 2014 and 2023 trials as favoring prazosin, and the 2015, 2016 and both 2018 trials as showing no benefit. a footnote says the 2023 trial 'shows no statistical significance for this outcome in our meta-analysis.' pooled results: insomnia SMD -0.654 (p=0.043), nightmares SMD -0.641 (p=0.025), overall PTSD SMD -0.428 (p=0.077). Fetched 2026-11-03 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC12985405
  2. Effects of prazosin on sleep disturbances in post-traumatic stress disorder: a systematic review and meta-analysis. Sleep Medicine, 2020. PMID 31972510 What we read there: eight RCTs, 286 prazosin vs 289 placebo. nightmares SMD -1.13 (95% CI -1.91 to -0.36). overall PTSD SMD -0.45 (CI -0.95 to 0.05) and sleep quality SMD -0.44 (CI -1.44 to 0.55), neither significant. Fetched 2026-11-03 (HTTP 200). https://pubmed.ncbi.nlm.nih.gov/31972510/
  3. Trial of Prazosin for Post-Traumatic Stress Disorder in Military Veterans. New England Journal of Medicine, 2018. PMID 29414272 What we read there: 304 veterans randomized 152/152. at 10 weeks, no significant difference on the nightmare item (p=0.38), sleep quality (p=0.80) or global change (p=0.96). 'prazosin did not alleviate distressing dreams or improve sleep quality.' Fetched 2026-11-03 (HTTP 200). https://pubmed.ncbi.nlm.nih.gov/29414272/
  4. Placebo-controlled comparison of prazosin and cognitive-behavioral treatments for sleep disturbances in US military veterans. Journal of Psychosomatic Research, 2012 What we read there: both active arms reduced insomnia and daytime PTSD symptoms more than placebo. 58% met full PTSD criteria and 42% had subthreshold symptoms. 'prazosin was not associated with reductions in nightmares as previously reported.' Fetched 2026-11-03 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC3267960

speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam