Survival Outcome — Full Analysis & Manuscript Generation

Table of Contents

• Scope
• Workflow
• Additional Inputs
• Output Structure
• Key References (read before generation)
• Reporting Standards
• Cross-Skill Interface

Open-source skill. Read reference/specs/output-variation-protocol.md before every generation — apply all variation layers.

Scope

Right-censored survival data only. Does not handle left-censoring, interval-censoring, or competing risks. Supports time-varying predictors (covariates that change during follow-up) and recurring/recurrent events (same subject experiences the event multiple times).

Workflow

Continues from where vera-data-survival-reviewing stopped (PART 0-2 done).

Step	Responsibility	Executor	Document	Input	Output
Additional tests	Run Additional Tests	Main Agent	workflow/step04-run-additional-tests.md	Prior step output	PART 3 code + prose
Subgroup	Analyze Subgroups	Main Agent	workflow/step05-analyze-subgroups.md	Prior step output	PART 4 code + prose
Modeling	Fit Models	Main Agent	workflow/step06-fit-models.md	Prior step output	PART 5 code + prose
Comparison	Compare Models	Main Agent	workflow/step07-compare-models.md	Prior step output	PART 6 code + prose
Manuscript	Generate Manuscript	Main Agent	workflow/step08-generate-manuscript.md	Prior step output	methods.md + results.md

Additional Inputs

Collect if not already provided:

• Target discipline (for reporting conventions)
• Target journal or style (APA 7th, STROBE, CONSORT, etc.)
• Research question / hypothesis
• Subgroup variable (if subgroup analysis desired)

Output Structure

output/
├── methods.md
├── results.md
├── tables/             ← Markdown + CSV per table
├── figures/            ← PNGs, 300 DPI
├── references.bib
├── code.R              ← Style-varied
└── code.py             ← Style-varied

Key References (read before generation)

File	Purpose
reference/specs/output-variation-protocol.md	Output quality variation layers
reference/specs/code-style-variation.md	Seven-dimension code style diversity
reference/patterns/sentence-bank.md	4-6 phrasings per result type
reference/rules/reporting-standards.md	Hard rules for statistical reporting

Reporting Standards

Same as vera-data-survival-reviewing, plus:

• HR: always with 95% CI, "HR = X.XX, 95% CI [X.XX, X.XX]"
• Time ratio (AFT): "TR = X.XX, 95% CI [X.XX, X.XX]" with direction interpretation
• Median survival: always with 95% CI
• Survival rates at landmarks: with 95% CI
• Censoring: always report % censored overall and by group
• Log-rank: chi-sq(df) = X.XX, p = .XXX
• Cox global test: LR chi-sq, Wald chi-sq, Score chi-sq
• Concordance: C = X.XX (note in-sample if no validation)
• PH assumption: report Schoenfeld test per predictor and globally
• Coefficients: HR with SE for Cox; TR with SE for AFT
• Tree-based with small N: frame as "exploratory"; never claim predictive validity
• p-value rules same as other skills

Cross-Skill Interface

Method Unit Contract:
├── code_r           → .R script (style-varied)
├── code_python      → .py script (style-varied)
├── methods_md       → methods.md (varied structure)
├── results_md       → results.md (varied phrasing)
├── tables/          → Markdown + CSV
├── figures/         → PNGs 300 DPI (varied layout)
├── references_bib   → .bib with cited references
└── comparison       → cross-method narrative (in results.md)

Invoked directly after vera-data-survival-reviewing or orchestrated by vera-data-application-pipelining.