Skill1k repo starsupdated 12d ago

clinical

This Claude Code skill provides a structured reference for clinical research methodology, covering study design hierarchies, statistical analysis techniques including survival analysis and meta-analysis, sample size calculation principles, and regulatory compliance requirements such as IRB approval and ClinicalTrials.gov registration. Use it when designing medical studies, selecting appropriate statistical approaches, ensuring regulatory adherence, or following standard reporting guidelines like CONSORT and STROBE.

View source Repository: ScienceClaw

Install in Claude Code

Copy

git clone --depth 1 https://github.com/beita6969/ScienceClaw /tmp/clinical && cp -r /tmp/clinical/skills/clinical ~/.claude/skills/clinical

Then start a new Claude Code session; the skill loads automatically.

Definition

SKILL.md

# Clinical Research

## Overview
Clinical study design, statistical analysis, and regulatory compliance for medical research.

## Study Designs
| Design | Level of Evidence | Best For |
|--------|------------------|----------|
| RCT | I | Treatment efficacy |
| Cohort (prospective) | II | Risk factors, prognosis |
| Cohort (retrospective) | III | Exposure-outcome associations |
| Case-control | III | Rare diseases, risk factors |
| Cross-sectional | IV | Prevalence, correlations |
| Case report/series | V | Novel observations |

## Common Analyses
- **Survival analysis**: Kaplan-Meier curves, Log-rank test, Cox regression
- **Diagnostic accuracy**: Sensitivity, specificity, ROC curve, AUC
- **Meta-analysis**: Fixed/random effects, forest plot, heterogeneity (I-squared)
- **Propensity score matching**: For observational study confounding
- **Nomogram**: Predictive model visualization for clinical use

## Sample Size Calculation
- Define: alpha (usually 0.05), power (usually 0.80), effect size, outcome type
- Tools: G*Power, R `pwr` package, Python `statsmodels`
- Report: formula used, assumptions, expected dropout rate

## Regulatory Compliance
- **IRB/Ethics committee**: Required for all human subjects research
- **Informed consent**: Written, voluntary, comprehensive
- **ClinicalTrials.gov**: Registration before enrollment (ICMJE requirement)
- **GDPR/HIPAA**: Data privacy for patient information
- **GCP (Good Clinical Practice)**: ICH E6 guidelines

## Reporting Guidelines
| Guideline | Study Type |
|-----------|-----------|
| CONSORT | Randomized controlled trials |
| STROBE | Observational studies |
| PRISMA | Systematic reviews / meta-analyses |
| STARD | Diagnostic accuracy |
| TRIPOD | Prediction models |
| SPIRIT | Study protocols |