Your trial, adjusted

Adjust a randomized trial for what you knew at baseline. The estimand stays the same and the interval gets narrower. See how many patients that is worth, and why a wrong model cannot bias it.

Take it to your trial. The SAP builder turns this lesson into a deliverable: your estimand, draft SAP text for the adjusted analysis, the patients it saves, simulated operating characteristics, and R code.

Further reading: FDA (2023), Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biological Products; Tsiatis, Davidian, Zhang and Lu (2008), Covariate adjustment for two-sample treatment comparisons in randomized clinical trials, Statistics in Medicine; Benkeser, Díaz, Luedtke, Segal, Scharfstein and Rosenblum (2021), Improving precision and power in randomized trials for COVID-19 treatments using covariate adjustment, Biometrics; EMA (2022), Qualification opinion for Prognostic Covariate Adjustment (PROCOVA). The trial, the device and its outcome scale are a teaching world. Exact values and seeded simulations are labelled beside each figure.