Trial Reporting (bioSkills)

A Claude Code skill that takes a clinical trial from estimand definition through Table 1, primary analysis, missing-data sensitivity and multiplicity control to a CONSORT-conformant statistical report.


Type	Claude Skill
Supplier	GPTomics bioSkills (community OSS, MIT)
Availability	GA — part of the bioSkills collection
Pricing	Free / OSS (MIT) — the Python (`pandas`, `numpy`, `statsmodels`, `scikit-learn`, `tableone`, `rpy2`) and R (`mmrm`, `rbmi`, `gMCP`, `RBesT`, `mice`, `miceforest`) packages are separately installed OSS
Capabilities	Read/Write — Claude runs the skill’s workflow locally (Python, with R via `rpy2` for confirmatory steps), not as an MCP tool
Verified	works · 2026-08-06
Security	cleared · 2026-08-06 — GPTomics/bioSkills MIT, standard OSS deps, no advisories

How to install

bioSkills is not an npm package — skills are plain markdown/code read directly by the agent. Clone the repo, then either run the installer for the whole category or copy the single skill directory.

Claude Code — clone and install via the bundled script:
```
git clone https://github.com/GPTomics/bioSkills
cd bioSkills
./install-claude.sh --categories "clinical-biostatistics"
```
The installer copies matching skills into ~/.claude/skills/ (default target). Use ./install-claude.sh --list to preview the skills first.
Claude Code / other agents — copy just this one skill:
```
cp -r bioSkills/clinical-biostatistics/trial-reporting ~/.claude/skills/
```
(run from inside your clone — the previous step left you in bioSkills/; otherwise replace bioSkills/ with the absolute path of your clone). Install the packages on first use, e.g. pip install "pandas>=2.1" "tableone>=0.9" "statsmodels>=0.14" and, for the confirmatory steps, install.packages(c("mmrm", "gMCP")) in R.

What it does

Seven stages:

Estimand definition — specify all five ICH E9(R1) attributes before selecting a method, and name the intercurrent-event strategy (treatment policy, hypothetical, composite, while-on-treatment, principal stratum).
Analysis populations — define ITT, full analysis set, per-protocol and safety populations with explicit inclusion and exclusion criteria.
Baseline characterization — Table 1 via tableone, using standardized mean differences (SMD > 0.1 flags imbalance) rather than baseline significance tests, which CONSORT advises against.
Primary analysis — MMRM under MAR for continuous endpoints (preferred), or reference-based MI under MNAR, reporting the treatment-by-visit contrast at the primary timepoint.
Missing-data sensitivity — Permutt tipping-point delta-adjustment and reference-based MI variants (J2R, CR, CIR), with the mechanism assumptions documented.
Multiplicity control — graphical procedures (Bretz–Maurer) via gMCP for co-primary and key secondary endpoints.
Regulatory reporting — CONSORT 2025 flow diagram, Item 21c (missing-data methods) and Item 4 (data and code sharing), plus ITT-versus-per-protocol reconciliation.

Software: Python tableone 0.9+ (primary), pandas 2.1+, numpy 1.26+, statsmodels 0.14+, scikit-learn 1.4+, rpy2 3.x+; R mmrm 0.3+, rbmi 1.5+, gMCP 0.8+, RBesT 1.6+ (Bayesian shrinkage for subgroup estimates), mice 3.14+, miceforest 5.x+.

Primary use cases: writing a clinical study report’s statistics sections, defining an estimand for an SAP, generating Table 1 and the CONSORT flow.

Notes

Standards cited: CONSORT 2025, SPIRIT 2025, ICH E9(R1) estimands, and the FDA 2023 covariate-adjustment guidance. The same caveat as elsewhere in this category applies — statsmodels.mixedlm has no Kenward–Roger correction and is exploratory only; confirmatory MMRM goes through R mmrm, which is why rpy2 is in the stack.

Distributed as a SKILL.md (plus reference material) in the bioSkills collection — Claude executes the workflow locally rather than as an MCP server. The upstream skill front-matter name is bio-clinical-biostatistics-trial-reporting; if invoked as a namespaced plugin command it resolves under the bioSkills plugin, not as a bare /trial-reporting. Upstream directory: clinical-biostatistics/trial-reporting.

Pairs with Adaptive Designs at the design stage and Missing Data Sensitivity when the missing-data package needs more depth than the one stage here provides. For generating the protocol document itself, see Clinical Trial Protocol; for time-to-event modelling, scikit-survival.

Sources

Installed this tool?

Share feedback — install path, OS, errors, workarounds. The form opens with this tool pre-selected and a link back to this page.