Fig. 5From: Analysis approaches to address treatment nonadherence in pragmatic trials with point-treatment settings: a simulation studyBias versus the nonadherence rate for the null and non-null effect for simulation setting 3. The 2SLS and 2SRI methods share the same line, and the baseline-adjusted PP and IP-weighted PP share the same line. The bias is out of the bound [-0.6,0.6] for the 2SLS and 2SRI methods for severe violation of the exclusion restriction. Abbreviations: PP: per-protocol, IP-weighted PP: inverse probability-weighted per-protocol, 2SLS: two-stage least square, 2SRI: two-stage residual inclusionBack to article page