K-Dense-AI/scientific-agent-skills / skills/clinical-decision-support
clinical-decision-support
VerifiedPrepares and validates research-only clinical decision-support evaluation, evidence-profile, cohort, survival, biomarker/model, privacy, and governance artifacts. Supports aggregate or synthetic research documentation and traceability, excluding patient care and live clinical operation.
assets/survival_analysis_plan_template.json
View on GitHub{
"schema_version": "1.0",
"metadata": {
"plan_id": "SYNTHETIC-SURVIVAL-PLAN-001",
"title": "Synthetic Aggregate Time-to-Event Analysis Plan",
"version": "0.1-draft",
"status": "draft",
"purpose": "Demonstrate estimand-led planning without time-to-event records.",
"data_cut_date": "2026-07-23",
"data_level": "plan_only",
"raw_rows_supplied": false,
"patient_care_use": false,
"human_review_required": true
},
"intended_use": "Research analysis planning only; not for patient care or live clinical use.",
"prohibited_uses": [
"No individual prognosis or classification.",
"No diagnosis, treatment recommendation, dosing, triage, alarm, or autonomous decision.",
"No claim of statistical approval, clinical validity, authorization, or compliance."
],
"limitations": [
"The validator does not inspect data or fit a model.",
"Methods require review against the actual design and estimand.",
"Synthetic placeholders must be replaced before an approved research analysis."
],
"estimand": {
"population": "Synthetic eligible research cohort",
"condition_or_comparison": "Synthetic exposure A versus synthetic exposure B",
"endpoint_variable": "Time from defined synthetic index to defined synthetic event",
"population_summary": "Hazard contrast plus restricted mean survival time difference at the prespecified horizon",
"time_horizon": "24 synthetic time units",
"rationale": "The hazard contrast addresses relative event rate; restricted mean survival time supports interpretation if proportional hazards is doubtful.",
"intercurrent_events": [
{
"event": "Synthetic exposure discontinuation",
"strategy": "Treatment-policy strategy in the primary estimand",
"rationale": "Preserves the prespecified research question."
},
{
"event": "Synthetic competing event",
"strategy": "Competing-event strategy with cumulative incidence",
"rationale": "The competing event precludes the primary event."
}
]
},
"endpoint": {
"time_zero": "Synthetic index date defined before analysis",
"event_definition": "Prespecified synthetic event with blinded aggregate adjudication",
"time_scale": "Synthetic months",
"ascertainment": "Prespecified uniform synthetic schedule",
"follow_up_end": "Earliest of event, loss to follow-up, or administrative data cut",
"delayed_entry": "Assess and use left-truncation methods if eligibility begins after time zero",
"same_time_rules": "Apply prespecified deterministic ordering and document ties",
"censoring_rules": [
"Administrative censoring at the fixed data cut",
"Censor loss to follow-up at last verified observation with informative-censoring sensitivity analysis"
],
"competing_events": [
"Synthetic competing event"
]
},
"analysis": {
"primary_method": "cox_proportional_hazards",
"effect_measure": "Hazard ratio with 95% confidence interval; restricted mean survival time difference at 24 units if proportional hazards is not supported",
"analysis_population": "Prespecified synthetic analysis population",
"covariate_strategy": "Prespecified covariates based on subject-matter rationale; no univariable p-value selection",
"missing_data": "Describe patterns; use justified multiple-imputation or model-based approach with sensitivity analysis",
"multiplicity": "Prespecify primary endpoint and control or transparently label secondary and exploratory analyses",
"uncertainty": "Two-sided 95% confidence intervals and assumption-focused sensitivity analyses",
"software_and_version": "Record locked statistical software, packages, code, and versions before execution",
"model_diagnostics": "Functional forms, influential observations, residual checks, convergence, and calibration where predictive",
"proportional_hazards_assessment": "Prespecified graphical assessment and time-interaction/residual-based assessment",
"non_proportional_hazards_strategy": "Report time-varying effects and restricted mean survival time at the prespecified horizon",
"competing_risk_method": "cumulative_incidence"
},
"bias_controls": {
"informative_censoring": "Compare follow-up patterns and run a justified sensitivity analysis.",
"immortal_time": "Align eligibility, assignment, and time zero; model time-varying exposure if required.",
"time_dependent_confounding": "Specify whether g-methods are required for the causal estimand.",
"outcome_misclassification": "Document ascertainment validity and conduct plausible sensitivity analyses.",
"informative_visits": "Assess observation schedules and potential outcome-dependent measurement."
},
"sensitivity_analyses": [
"Alternative censoring assumptions tied to plausible loss-to-follow-up mechanisms",
"Alternative effect summary when proportional hazards is not supported",
"Competing-risk analysis aligned with the absolute-incidence question"
],
"subgroups": {
"prespecification": "List scientifically justified subgroups before results.",
"interaction_testing": "Use formal interaction estimates with uncertainty.",
"multiplicity": "Control or label exploratory subgroup analyses.",
"precision_and_disclosure": "Report aggregate sizes/events and suppress unstable or disclosive outputs."
},
"validation": {
"internal_validation": "Use design-appropriate resampling if a prediction model is involved.",
"external_validation": "Plan independent temporal or geographic validation before transportability claims.",
"calibration": "Evaluate calibration at prespecified horizons for prediction models.",
"subgroup_performance": "Report performance and uncertainty across prespecified groups."
},
"human_review": {
"roles": [
"survival-analysis statistician",
"epidemiologist",
"domain expert"
],
"completed": false,
"approval_boundary": "An accountable research organization must approve the final plan; approval never permits patient-care use."
},
"governance": {
"owner": "research analysis governance owner",
"change_control": "Version the estimand, endpoint, data cut, code, software, deviations, and sensitivity analyses.",
"auditability": "Retain the approved plan, amendments, execution logs, aggregate outputs, and reviewer decisions.",
"monitoring": "Review assumption violations, missingness, censoring, subgroup precision, and external validation.",
"retirement": "Supersede or retire the plan when the question, data source, endpoint, or method changes materially."
},
"source_ids": [
"ICH-E9R1",
"RMST"
],
"sources": [
{
"id": "ICH-E9R1",
"citation": "ICH E9(R1), Addendum on Estimands and Sensitivity Analysis in Clinical Trials, 2019.",
"url": "https://database.ich.org/sites/default/files/E9-R1_Step4_Guideline_2019_1203.pdf",
"accessed": "2026-07-23"
},
{
"id": "RMST",
"citation": "Royston P, Parmar MKB. Restricted mean survival time. BMC Medical Research Methodology. 2013;13:152.",
"url": "https://pubmed.ncbi.nlm.nih.gov/24314264/",
"accessed": "2026-07-23"
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |