ModusLab
Staring at a paper for journal club knowing you should appraise it properly, but not where to start?
A ChatGPT prompt for critically appraising a paper
Free · lite version A simpler, single-purpose version of the full toolkit prompt — see the difference ↓Critical appraisal is a skill that rewards structure. The same questions apply to almost any clinical paper — what was the question, was the design right for it, where could bias creep in, do the results support the conclusions — but holding all of them in your head while reading is hard.
This prompt turns the paper into a structured appraisal you can pressure-test and learn from. Paste the abstract and methods (or the full text), and it works through the appraisal with you rather than just summarising.
You are helping a UK doctor critically appraise a research paper. Base your appraisal ONLY on the text I provide; if something isn't reported, say it isn't reported rather than guessing. Work through: 1. The research question (PICO where applicable) 2. Study design — and whether it suits the question 3. Risk of bias — selection, performance, detection, attrition, reporting 4. The results — effect size, precision (confidence intervals), and whether they're clinically meaningful, not just statistically significant 5. Internal and external validity — does it generalise to UK practice? 6. Bottom line — what this paper does and does not let us conclude End with three sharp questions a journal club could debate. Here is the paper: [PASTE THE ABSTRACT, METHODS, AND RESULTS HERE]
Anonymise before you paste. Outputs are drafts — review and verify before use. Clinical responsibility stays with you.
What this prompt does
- Works through design, bias, and validity in a fixed structure
- Separates statistical significance from clinical meaning
- Ends with debate-ready journal-club questions
Want the rest of the set — free?
Get the Doctor’s AI Starter Kit: a clean PDF of our most-used prompts you can keep, plus the occasional new one as we add them. No spam.
Email capture activates at launch. (Set the Buttondown username in src/lib/newsletter.ts.)
What you’re getting — and what you’re not
- Works through design, bias, and validity in a fixed structure
- Separates statistical significance from clinical meaning
- Ends with debate-ready journal-club questions
The comprehensive “Single Paper — Deep Analysis + Learning”, plus 8 more prompts
This free version appraises one paper. The Academic toolkit also synthesises across multiple papers into a mind map, maps research gaps and next studies, and turns lectures or podcasts into structured notes.
Get Academic & Research — £19 £69 70% off One-time payment · Instant PDF · ChatGPT, Claude or Gemini