Software for estimating patient-centered causal effects in pragmatic trials. For the most updated version of the software, please see the linked github repositories.
Case study 1: Coronary Drug Project (CDP)
The SAS code (version March 2018) used to estimate the relationship between adherence and mortality in the CDP placebo arm is available here and at github .
Questions: Contact Eleanor Murray.
Case study 2: Strategic Timing of AntiRetroviral Treatment Trial (START)
A SAS macro for estimating the per-protocol effect from a randomized trial using the parametric g-formula is available on the Causal Inference Software page and at github .
Requires the external SAS macro gformula3 located on the Causal Inference Software page.
Case study 3: Norwegian Colorectal Cancer Prevention Trial (NORCCAP)
A SAS macro for estimating bounds on the counterfactual risks and the average treatment effect using instrumental variables is available here and on github.
Questions: Contact Sonja Swanson
Reference:
Swanson SA, Hernán MA, Miller M, Robins JM, Richardson T. Partial identification of the average treatment effect using instrumental variables. JASA 2018; 113(522):933-947.
Case study 4: Candesartan in Heart Failure Morbidity and Mortality Trial (CHARM)
SAS code (version June 2019) for estimating the relationship between adherence and mortality in the CHARM trial, and the per-protocol effect of candesartan on all-cause mortality. To download software and documentation is available here and on github.
Case study 5: Lipid Research Clinics Coronary Primary Prevention Trial (LRC-CPPT)
R code for adjusting for adherence in the placebo arm of the LRC-CPPT. To download software and documentation (version May 2019) is available here.