Table 1.

PSA screening and the diagnosis of prostate cancer in the SEER 9 population aged 50–84 years during 1985–2000, as predicted by the three models^*

Group	Item	MISCAN	FHCRC	UMich
A	No. of screening tests	7 919 110	7 769 666	7 433 518
B	No. of men diagnosed with PC†	238 720	243 565	230 449
C	No. of screen-detected cancers	106 061	103 058	86 975
	Percentage of group A	1.3	1.3	1.2
	Percentage of group B	44.4	42.3	37.8
D	No. of overdiagnosed cancers	44 499	28 874	19 872
	Percentage of group B	18.6	11.9	8.6
	Percentage of group C	42.0	28.0	22.9
E	Lead time, y
	Non-overdiagnosed, mean	6.9	5.9	5.4
	Censored, mean	7.8	5.9	5.7
	Uncensored, mean	10.0 (median)	7.2	8.8

^*PSA = prostate-specific antigen; PC = prostate cancer; SEER 9 = the nine core catchment areas in the Surveillance, Epidemiology, and End Results program of the National Cancer Institute; MISCAN = the microsimulation screening analysis model based on ERSPC Rotterdam, calibrated to SEER 9 incidence; FHCRC = the microsimulation model developed at the Fred Hutchinson Cancer Research Center, explicitly linking PSA levels and prostate cancer development; UMich = the analytic model developed by Dr Tsodikov (University of Michigan); ERSPC = European Randomized Study of Screening for Prostate Cancer.

^†Observed number of men diagnosed with prostate cancer = 235 112.