checkedPsychopharmacology and Psychiatry Updates2026-11-08last verified 2026-11-08

did only 41% of patients in the catatonia review get a benzodiazepine before the antipsychotic?

verdict: misattributed
a real number attached to the wrong thing
“Only 41% of patients in these reports had been given a benzodiazepine prior to the antipsychotic.”

a guest, Dr. Scott Beach (psychiatrist presenting the review) said on Psychopharmacology and Psychiatry Updates, 2026-11-08, at 05:28. hear it in context → the speaker cited: a recent systematic review in European Psychiatry.

the episode: Catatonia: Should Antipsychotics Always Be Avoided? (at 05:28). we link the show's own page; we do not re-host audio.

first, the part that holds

The 41% figure is real, and it comes from the review the speaker named: a 2025 systematic review in European Psychiatry on using antipsychotics in catatonia. The review's own abstract says "only 41.1% of the patients" had a prior benzodiazepine trial, so the speaker repeated the paper's headline wording almost word for word.

what we found when we went and looked

We read the full text of the review (European Psychiatry 2025;68(1):e48). The abstract says "only 41.1% of the patients benefited from a previous benzodiazepine trial." The results section and table 1 show that the 41.1% is 72 of 175 antipsychotic trials, not a share of the 110 patients those trials happened in. The paper never reports how many patients had a benzodiazepine first. So the number is right, but the paper's own abstract attached it to patients when it counts trials, and the claim repeats that.

assertion by assertion

assertionverdictwhat the record says
The figure comes from a recent systematic review in European PsychiatryconfirmedPublished 2025 in European Psychiatry 68(1):e48, doi:10.1192/j.eurpsy.2025.9. It included 79 articles, 110 patients and 175 antipsychotic trials, almost all from case reports and case series. [1][2][3]
The review reports about 41% for benzodiazepines given before the antipsychoticconfirmedResults: "BZD were tried before antipsychotics in 72 trials (41.1%)." The abstract gives the same 41.1%. [1][2][3]
The 41% is a share of patientsmisattributedThe count behind it is trials: 72 of 175. If it were patients, 72 of 110 would be about 65%. The abstract's word 'patients' doesn't match the paper's own table, and the paper gives no count of patients who had a benzodiazepine first. [1][3]
The figure shows how often benzodiazepines are skipped in real-world catatonia careunverifiableThe authors say the data reflect "the frequency of events reported in the literature, not the frequency of occurrence in real-world practice." They also note that case reports often leave out treatment history, so a benzodiazepine trial that was never written down can't be told apart from one that never happened. [1][3]
Trials where the antipsychotic helped were more likely to have had a benzodiazepine firstconfirmedTable 1: 48 of 105 beneficial trials (45.7%) had a prior benzodiazepine, against 24 of 70 neutral or harmful trials (34.3%). Benzodiazepine given at the same time as the antipsychotic was 45.7% against 22.9%. This is descriptive case-report data and doesn't show cause. [2][3]

both directions

for the speaker: The number, the journal and the topic all match. The abstract itself says "41.1% of the patients," so the speaker read the source faithfully. The general point also holds: benzodiazepines are the standard first treatment for catatonia, yet a minority of the reported antipsychotic trials list one beforehand. The authors draw the same conclusion and recommend trying a benzodiazepine alone before starting an antipsychotic.

against: The paper's results count trials, not patients: 72 of 175 antipsychotic trials, which happened in 110 patients. Some patients had several trials, so a patient-level figure could be quite different, and the paper doesn't give one. The abstract's wording is also loose ("benefited from a previous benzodiazepine trial" for what the results call "tried before"). The source is case reports, which the authors say record what got published, not what clinicians usually do.

what this cannot say

Case reports are written up because they are unusual. They are prone to publication and recall bias and often leave out medication history, so 41.1% is the share of trials where a prior benzodiazepine was reported, not a measured rate. We could not get patient-level benzodiazepine exposure from the paper. We relied on the published text and table 1, not the underlying case files. We did not check how the full show presented the figure beyond the quoted sentence.

sources, each one fetched and read

  1. The use of antipsychotics in the treatment of catatonia: a systematic review. European Psychiatry 2025;68(1):e48. doi:10.1192/j.eurpsy.2025.9 (Cambridge Core full-text reader) What we read there: "BZD were tried before antipsychotics in 72 trials (41.1%)"; 175 antipsychotic trials in 110 patients; the data reflect "the frequency of events reported in the literature, not the frequency of occurrence in real-world practice." Fetched 2026-11-08 (HTTP 200). https://www.cambridge.org/core/product/332FE3F1201E16C3F67ECBB059DA85DD/core-reader
  2. The use of antipsychotics in the treatment of catatonia: a systematic review. European Psychiatry 2025;68(1):e48 (Cambridge Core article page) What we read there: Abstract: "Only 41.1% of the patients benefited from a previous benzodiazepine trial"; beneficial trials had a prior benzodiazepine more often (45.7% vs 34.3%) and same-time benzodiazepine more often (45.7% vs 22.9%); the authors recommend trying a benzodiazepine alone before starting an antipsychotic. Fetched 2026-11-08 (HTTP 200). https://www.cambridge.org/core/product/identifier/S0924933825000094/type/journal_article
  3. The use of antipsychotics in the treatment of catatonia: a systematic review. European Psychiatry 2025;68(1):e48. PMC12041727 What we read there: 79 articles, 110 patients, 175 trials; table 1: prior benzodiazepine in 72/175 trials (41.1%), 48/105 beneficial vs 24/70 neutral or harmful; the unit of analysis is the trial, not the patient; "data were frequently missing." Fetched 2026-11-08 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC12041727/

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