Skill: Clinical Trial Statistical Analysis Planning
Plan rigorous statistical analyses for clinical trial protocols
What You Can Do
This skill guides you through designing comprehensive statistical analysis plans (SAPs) for clinical trials, ensuring regulatory compliance and methodological rigor. You'll define primary and secondary endpoints, determine appropriate sample sizes, specify statistical tests, and document pre-specified analyses that minimize bias and maximize scientific validity. The skill helps you translate clinical hypotheses into precise statistical frameworks that withstand peer review and regulatory scrutiny.
Features
Systematically define primary, secondary, and exploratory endpoints with clear efficacy criteria, timing windows, and outcome measures aligned to trial objectives.
Calculate required sample sizes based on effect sizes, statistical power, significance levels, and assumptions, with sensitivity analyses for different scenarios.
Recommend appropriate statistical tests for your trial design (parallel, crossover, adaptive) accounting for data distributions, missing data patterns, and multiplicity.
Specify handling approaches for dropout and missing values including sensitivity analyses and imputation methods aligned with regulatory guidance (FDA, EMA).
Design multiple comparison corrections and interim analysis plans that maintain family-wise error rates while protecting trial integrity.
Pre-specify subgroup analyses, interaction tests, and handling criteria to prevent spurious findings while exploring clinically relevant heterogeneity.
Define safety endpoints, adverse event categorization, clinical significance thresholds, and statistical monitoring rules for patient protection.
Generate structured analysis plan sections ready for adaptation into your protocol supplement with rationales, assumptions, and contingency plans.
Example Output
Sample Size Calculation:
- Primary endpoint: Time to disease progression
- Hazard ratio: 0.67 (33% risk reduction)
- Power: 90%, α: 0.05 (two-sided)
- Expected median survival: Control 12 months, Treatment 18 months
- Required events: 187 progressions
- Estimated sample size: 240 patients (accounting for 10% early withdrawal)
Primary Analysis Set:
- Population: Intent-to-treat (ITT) — all randomized patients
- Test: Kaplan-Meier curves with log-rank test stratified by disease stage
- Censoring: Patients lost to follow-up or alive at study end
- Subgroups: Disease stage, ECOG performance status
Missing Data Plan:
- Up to 5% dropout acceptable before triggering sensitivity analyses
- Primary: Multiple imputation (chained equations)
- Sensitivity: Last observation carried forward (LOCF), worst-case imputation
What's Included
- Analysis Plan Development Workflow: Step-by-step guidance covering trial design assessment, statistical framework selection, and pre-specification of all analyses.
- Regulatory Compliance Checklist: Requirements aligned with FDA, EMA, ICH-GCP, and health authority-specific statistical guidance to ensure approvability.
- Statistical Power Calculator Reference: Formulas and assumptions for common trial designs (parallel, crossover, adaptive) with sensitivity tables for effect size variations.
- Sample Analysis Plan Template: Customizable SAP sections including study objectives, populations, variables, and statistical method specifications ready for protocol integration.
- Risk Assessment Framework: Identify statistical vulnerabilities (multiplicity, missing data, confounding) and pre-specify mitigation strategies to strengthen scientific evidence.
Who It's For
- Biostatisticians and statisticians designing clinical trials
- Clinical researchers writing study protocols and statistical sections
- Regulatory scientists preparing submissions to health authorities
- Pharmaceutical and biotechnology program managers ensuring analytical rigor
- Academic researchers conducting NIH-funded or investigator-initiated trials
Best For
- Designing statistical analysis plans for Phase II and Phase III trials
- Determining sample sizes and power calculations
- Planning subgroup and sensitivity analyses
- Specifying missing data and interim analysis strategies
- Documenting pre-specified analyses for regulatory submissions
You might also like
You can rigorously evaluate scientific research by assessing methodology, experimental design, and statistical validity using established frameworks like GRADE and Cochrane Risk of Bias. This skill helps you identify confounding variables, detect biases, and determine the overall quality of evidence in research papers, enabling you to critically challenge scientific claims and distinguish robust findings from flawed conclusions.
by k-dense-ai
You'll design robust SCADA system architectures, configure industrial protocols (Modbus, Profibus, OPC-UA, DNP3), and diagnose connectivity and performance issues across distributed control networks. Get step-by-step configuration guidance, integration workflows, and troubleshooting decision trees tailored to your specific hardware and protocol stack.
by skillslib
You can establish systematic color grading workflows that accelerate project delivery, ensure visual consistency across episodes and projects, and maintain broadcast-quality standards. Claude generates reusable templates, quality control checklists, and grading decision frameworks tailored to your project's color science and deliverable requirements.
by skillslib
You'll build complete activation funnels by analyzing user behavior, identifying conversion bottlenecks, and designing targeted onboarding campaigns. Claude creates customer journey maps, funnel stage definitions, engagement messaging frameworks, and A/B testing strategies tailored to your product and audience.
by skillslib
Use Claude to automate your entire collection workflow, from first touch to recovery, with legally sound escalation protocols. You'll recover more money in less time, reduce bad debt write-offs by 25-40%, and build defensible collection records that withstand legal scrutiny. Claude handles customized demand letters, compliance validation, debtor profiling, and escalation strategies tailored to account age, debtor behavior, and jurisdiction.
by skillslib
You can conduct comprehensive active transportation network audits that identify missing links, connectivity gaps, and underserved communities. Claude helps you prioritize infrastructure investments against limited budgets using equity screening, mode-share modeling, and cost-benefit analysis—producing client-ready documentation and stakeholder presentations that justify infrastructure decisions with quantitative evidence rather than assumptions.
by skillslib
Turn project descriptions, timelines, and structured data into professional draw.io diagrams instantly. You can generate flowcharts, org charts, Gantt charts, swimlane diagrams, and process flows by describing what you need. Claude creates draw.io XML that you can open directly in the editor or modify further.
by skillslib
This skill helps you troubleshoot Wwise integration issues, diagnose audio performance bottlenecks, and generate comprehensive documentation for interactive audio specifications. You'll get step-by-step fixes for common implementation problems, performance optimization strategies, and best practices for audio systems across different game engines and platforms.
by skillslib