%0 Journal Article
%K Humans
%K *Practice Patterns, Physicians'/statistics & numerical data
%K *Confounding Factors, Epidemiologic
%K Bias
%K Linear Models
%K Least-Squares Analysis
%K United Kingdom
%K Computer Simulation
%K instrumental variables
%K noncollapsibility
%K physician's prescribing preferences
%K two-stage prediction substitution
%K two-stage residual inclusion
%K unmeasured confounding
%B J Comp Eff Res
%D 2024
%G eng
%M 38567965
%P e230085
%R 10.57264/cer-2023-0085
%T Comparing the performance of two-stage residual inclusion methods when using physician's prescribing preference as an instrumental variable: unmeasured confounding and noncollapsibility
%V 13
%2 PMC11036961
%7 2024/04/03
%@ 2042-6305 (Print)2042-6305