Table 6. Results for diference in restricted mean survival times (RMST) at 14 days estimand in hospitalized population, when the adjusted estimator uses all six baseline variables from^{Section 4.2.3}.

Confidence intervals and hypothesis tests are Wald-based. “Effect” denotes the true estimand value; “MSE” denotes mean squared error; “Rel. Eff.” denotes relative efficiency which we approximate as the ratio of the MSE of the estimator under consideration to the MSE of the unadjusted estimator. In each block of four rows, the first two rows involve no treatment effect and the last two rows involve a benefit from treatment.

Sample Size	Estimator Type	Effect	P(reject H0)	MSE	Bias	Variance	Rel. Eff.
100	Unadjusted	0	0.011	0.781	0.018	0.780	1.000
100	Adjusted	0	0.038	0.740	0.006	0.740	0.948
100	Unadjusted	1.06	0.085	0.570	−0.265	0.500	1.000
100	Adjusted	1.06	0.160	0.546	−0.251	0.483	0.957
200	Unadjusted	0	0.048	0.481	−0.013	0.481	1.000
200	Adjusted	0	0.055	0.418	−0.013	0.418	0.869
200	Unadjusted	1.06	0.326	0.328	−0.145	0.307	1.000
200	Adjusted	1.06	0.397	0.293	−0.123	0.278	0.893
500	Unadjusted	0	0.050	0.201	−0.003	0.201	1.000
500	Adjusted	0	0.052	0.164	−0.005	0.164	0.814
500	Unadjusted	1.06	0.729	0.151	−0.070	0.146	1.000
500	Adjusted	1.06	0.810	0.129	−0.065	0.125	0.855
1000	Unadjusted	0	0.048	0.100	0.001	0.100	1.000
1000	Adjusted	0	0.045	0.079	0.001	0.079	0.793
1000	Unadjusted	1.06	0.959	0.079	−0.060	0.076	1.000
1000	Adjusted	1.06	0.986	0.065	−0.058	0.061	0.819