Everything Hertz 195: Living meta-analysis
Jan 14, 2026
They explore living meta-analyses that are continuously updated to keep evidence current and reduce research waste. They debate what counts as “recent” in scientific literature and citation practices. They warn about risks of synthetic research participants and discuss verification, business incentives, and threats to research integrity.
AI Snips
Chapters
Transcript
Episode notes
What "Recent Study" Actually Means In Medicine
- The phrase "recent study" varies wildly across specialties, with dentistry showing a median lag of 14 years in the sampled papers.
- Dan Quintana reports a broader median of 1–2 years across medicine and notes critical care and infectious disease are the fastest to call something recent.
Don't Call Papers Recent Unless Within A Year
- Avoid labeling literature as "recent" unless it's within the past year to prevent misleading readers.
- Dan Quintana enforces this with his students, rejecting the term unless the cited work is within the last calendar year.
Synthetic Survey Panels Are Becoming A Real Offer
- Qualtrics and similar services are offering synthetic panels that generate fake participants to complete surveys, provoking widespread concern.
- James Heathers and Dan Quintana discuss limited legitimate uses like testing analysis pipelines but warn of major ethical and validity problems.