Comparative study overview — Comparative cohort using existing data

Causal question

Specify the exposure strategies, target population, outcome and whether a causal claim is intended.

Protocol

Prespecify design, analysis, permissions and protocol availability before examining the comparative results.

Assess sources

Assess whether existing records capture both groups, baseline factors, outcomes and observation windows.

Define comparable groups and time zero

Apply comparable eligibility and align exposure assignment and time zero; avoid defining groups using future survival.

Derive outcomes

Define outcomes, assessment windows, competing events and the estimand for each comparison.

Address confounding

Use subject knowledge to identify confounders and choose adjustment; examine overlap and measured balance.

Estimate effects

Estimate the prespecified risk, rate or other contrast with compatible uncertainty methods.

Assess robustness

Examine sensitivity to definitions, missingness, residual confounding and informative censoring.

Report

Report selection, balance, absolute and relative contrasts where relevant, and limits on causal interpretation.

Comparative study overview

End flow

Protocol and decision register

Record the protocol, approvals and registration where applicable. Preserve planned decisions and amendments separately from results.

Protocol and rationale evidence

Attach the rationale and dated protocol, retaining amendments. A study registry records study details; a disease registry contains health data. The HMA–EMA Catalogue of RWD studies is one option for relevant studies. Check eligibility and applicable requirements, and record an identifier and timing if registered. Registration does not imply randomization.