MedPea curates the tools of the medical-research workflow — literature, references, journals, public cohort data and AI — and layers a cross-cohort variable index of 519,143 variables on top. Free, no account, fully static.
Why MedPea
Most research delays aren't analytical — they're logistical. MedPea removes the lookup layer between you and the data.
69 tools hand-picked by working researchers and organized by workflow stage — from PubMed and Zotero to MIMIC and openFDA. Every entry reviewed, every link live.
One query across CHARLS, HRS, ELSA, SHARE and six more aging cohorts: 59 harmonized core variables with real coverage, plus full-text search over 519K raw variables.
Click any variable and get pandas and Stata snippets for pulling that variable out of the merged cohort table — waves, keys and subsetting included.
No account, no tracking, nothing stored on our side. Your saved tools and theme live in your browser, and the whole app is a set of static files served from the edge.
Product
Six stages of the medical-research workflow, each with its own curated toolkit. Open any stage to jump straight in.
About
MedPea started as an internal bookmark file that kept growing. We are a small distributed team of public-health researchers and engineers who spent years doing cohort studies — and lost entirely too much time rediscovering the same thirty browser bookmarks, re-checking which cohort has which variable, and re-writing the same extraction snippets.
So we turned that scattered knowledge into infrastructure: a curated workbench for the whole workflow, and a variable index that answers "which cohort, how many waves, how many people" before you download a single dataset.
The core product is free and runs entirely in your browser. We're working on deeper harmonization tooling for multi-cohort studies — if that's your world, we'd love to hear from you.
Questions, feedback, or partnership ideas — we read everything.
founder@xiaoxiaof.com