TABLE 2.

Simulation to mimic the monitoring and management of prostate cancer: two intervention stages, two treatment options at each stage nested within the exam at each stage with 500 replications, and n=1000 or 2000. opt% show the empirical mean and standard deviation (SD) of the percentage of subjects correctly classified to their underlying true optimal treatments, estimated by the proposed method when (a) the conditional mean model and the propensity score model are both correctly specified, (b) the conditional mean model is mis-specified and the propensity score model is correctly specified, and (c) the conditional mean model is correctly specified and the propensity score model is mis-specified. Different treatment rates correspond to different proportions of patients who switch from active surveillance to curative treatment among those who have taken the biopsy test.

Treatment Rate		5%	15%	20%	25%
		opt%	opt%	opt%	opt%
N=1000	(a)	92.8(4.9)	93.8(4.8)	94.9(4.5)	95.3(4.6)
	(b)	91.6(2.1)	93.2(1.9)	93.9(1.5)	94.3(1.4)
	(c)	91.8(5.6)	93.2(5.8)	94.0(5.9)	94.6(5.5)
N=2000	(a)	93.7(4.7)	94.9(4.6)	95.7(4.4)	96.7(3.7)
	(b)	92.6(1.9)	94.0(1.4)	94.5(1.2)	94.7(1.1)
	(c)	92.2(6.5)	93.9(6.6)	94.6(6.5)	95.5(6.1)