Skip to main content
. Author manuscript; available in PMC: 2023 Oct 18.
Published in final edited form as: Proc Mach Learn Res. 2023 Aug;218:98–115.

Table 4:

G-formula estimation the effect of vancomycin vs. other treatment updated at each time point on 30-days mortality, in the presence of time-varying and time-fixed confounding, stratified by empiric treatment (Time 1).

Population Intervention G-formula mean Mean ratio Mean difference
All patients (N=427) 0 – Observed Treatment 0.07645 1.00 (ref) 0.0000
1 – Never treat with vancomycin or switch immediately after empiric treatment 0.14615 1.91 0.0696
2 – Always Treat with vancomycin and do not switch to another antibiotict 0.05005 0.65 −0.0264
Patients received vancomycin at T1 (N=414) 0 – Observed Treatment 0.07533 1.00 (ref) 0.0000
1 – Never treat with vancomycin or switch immediately after empiric treatment 0.16744 2.22 0.0921
2 – Always Treat with vancomycin and do not switch to another antibiotict 0.04977 0.66 −0.0255