Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Medical Research Statistics (analyze-stats)

Turns a clinical dataset into an approved analysis plan, reproducible Python/R code, journal-formatted tables and figures, and a draft Results paragraph.

Data & AnalyticsAdvanced★ 269⑂ 65AI score 9/10Last updated: Aug 24, 2026

What it does

  • Profiles your data file first (shape, inferred types, missingness, previews) and pins down the analysis unit — patient, exam, lesion, image, or reader.
  • Proposes an analysis plan across Table 1, diagnostic accuracy (Se/Sp/PPV/NPV/AUC), inter-rater agreement (ICC, kappa, Bland-Altman), survival (KM/Cox), linear and logistic regression, propensity score (PSM/IPTW/overlap), repeated measures (LMM/GEE), complex-survey data (KNHANES/NHANES), and meta-analysis/DTA meta-analysis — then waits for your approval before running.
  • Enforces publication reporting rules: exact p-values, 95% CIs for every primary estimand, effect sizes, assumption tests, and multiple-comparison correction.
  • Pre-empts classic failures: separation screening before logistic models, strata disjointness checks before trend tests, proportion-CI lower-bound clamping, and calibration alongside AUC for prediction models.
  • Lints its own generated scripts for missing seeds, hand-typed data literals, absolute paths, and in-place overwrites of raw data.
  • Emits tables as CSV + console markdown + R gtsummary code using per-journal profiles (Radiology, JAMA, NEJM, Lancet), and figures as PDF (vector) + PNG (300 DPI).
  • Finishes with a manuscript-ready Results paragraph, captions, and a short Methods snippet, plus an output manifest for downstream figure/paper skills.

Who it's for

  • Clinicians, residents, and graduate students writing clinical, radiology, or public-health papers.
  • Researchers who lose hours reformatting statistics to journal house style.
  • Teams working with weighted survey data or claims/ICD-10 cohorts that need design-aware estimates.

Example uses

  1. "Build Table 1 and compare the two arms in cohort.csv" → data profile → normality/variance checks pick t-test vs Mann-Whitney → AMA-styled table CSV plus a Results paragraph.
  2. "Does the AI score add value over the clinical model?" → paired ΔAUC with DeLong CI, category and continuous NRI, IDI with bootstrap CIs, and net benefit at a prespecified threshold.
  3. "Compute reader agreement" → Cohen's kappa or ICC with the model/type stated, Bland-Altman bias and limits of agreement, bootstrap CIs, and interpretation labels.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/Aperivue/medsci-skills/HEAD/skills/analyze-stats/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/analyze-stats folder from the GitHub repo Aperivue/medsci-skills into my ~/.claude/skills/analyze-stats/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Aperivue/medsci-skills.git && mkdir -p ~/.claude/skills && cp -r medsci-skills/skills/analyze-stats ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal (Terminal on macOS/Linux, Git Bash or WSL on Windows).
  2. Clone the repository: git clone https://github.com/Aperivue/medsci-skills.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill in: cp -r medsci-skills/skills/analyze-stats ~/.claude/skills/
    • Confirm the references/ and scripts/ subfolders came along; the templates, journal profiles, and check scripts live there and the skill leans on them heavily.
  5. Install the analysis stack: pip install pandas numpy scipy statsmodels scikit-learn matplotlib pingouin lifelines. For meta-analysis or gtsummary tables, in R run install.packages(c("meta","metafor","mada","gtsummary","survival")).
  6. Restart Claude Code, open the folder containing your data, and ask it to "run the statistical analysis on this dataset" or invoke /analyze-stats.
  7. De-identify patient data first — the skill will warn you and prefers any *_deidentified.* file it finds, and it refuses to print raw identifiers.
View source on GitHub ↗License: MIT