All corrections
LessWrong April 6, 2026 at 07:47 PM

www.lesswrong.com/posts/6npSCaALdFaoeFx59/it-takes-two-paracetamol

2 corrections found

1
Claim
This study only contains 29 participants, where only 14 of them took paracetamol at a dose of 500mg.
Correction

This misstates the Dahlöf 1996 study design and sample. Dahlöf 1996 was a five-period crossover trial in which each participant treated separate headache attacks with each test medication, so the study was not limited to 14 people on paracetamol 500 mg.

Full reasoning

The cited Dahlöf 1996 trial was not a parallel-group study where different subsets of participants were assigned to only one dose. According to the PubMed record for the paper, it was a "five-period, within-patient comparative trial" in which "Each patient had to treat five attacks of episodic tension-type headache with a single dose of each of the tested medications"—that is, ketoprofen 25 mg, ketoprofen 50 mg, paracetamol 500 mg, paracetamol 1,000 mg, and placebo.

The same abstract also reports: "Altogether 30 patients treated 5 attacks and 2, 3, 1 and 4 patients treated 4, 3, 2 and 1 attack, respectively." That means the study involved data from more than 29 participants overall, and because it was a crossover study, it is incorrect to say that only 14 participants "took paracetamol at a dose of 500mg." The paper's design specifically required participants to take each tested medication across separate headache attacks.

So the sentence is inaccurate in two ways:

  1. it understates the participant count; and
  2. it incorrectly treats the paracetamol-500-mg condition as if only a small subgroup received it, instead of recognizing the study's within-patient crossover design.
2 sources
2
Claim
Pain-free at one hour: The proportion of attacks/participants who were pain-free or had only mild pain at one hour with paracetamol.
Correction

This outcome description is incorrect. In the Cochrane review, “pain-free at one hour” means participants who were pain-free at one hour, not those who were pain-free or had only mild pain.

Full reasoning

This bullet conflates two different outcome measures used in the Cochrane review.

In the review's summary-of-findings table for paracetamol 1000 mg vs placebo, "Pain-free at 1 hour" is listed as its own outcome. Separately, the review lists "Pain-free or mild pain at 2 hours" as a different outcome. Those are not interchangeable endpoints.

So the post's wording is inaccurate because it defines "Pain-free at one hour" as if it meant pain-free or mild pain, when the review treats those as distinct measures. The correct description of that outcome is simply the proportion of participants who were pain-free at one hour.

1 source
Model: OPENAI_GPT_5 Prompt: v1.16.0