Table 1

Comparison of men with and without endpoint evaluated

N (%) / mean ± stdEnd point evaluatedOR*(95% CI)P*
NoYes
n= 5,809n= 10,181
Treatment arm
 Finasteride3,008 (52%)4,958 (49%)0.89 (0.84-0.95)0.0007
 Placebo2,801 (48%)5,223 (51%)1.0 (reference)
Age at randomization63.4 ± 5.962.9 ± 5.40.98 (0.97-0.99)<0.0001
Race
 White5,297 (91%)9,483 (93%)1.37 (1.21-1.55)<0.0001
 Other512 (9%)699 (7%)1.0 (reference)
Family history of PCA
 Yes782 (13%)1,698 (17%)1.23 (1.12-1.35)<0.0001
 No1,200 (79%)383 (77%)1.0 (reference)
PSA at randomization63.4 ± 5.962.9 ± 5.40.99 (0.94-1.04)0.59
Prior negative study biopsy
 Yes463 (8%)1,349 (13%)1.60 (1.43-1.80)<0.0001
 No5,345 (92%)8,833 (87%)1.0 (reference)<0.0001
Biopsy prompt for elevated PSA
 Yes69 (1%)803 (8%)6.80 (5.32-8.84)<0.0001
 No5,739 (99%)9,381 (92%)1.0 (reference)
Biopsy prompt for suspicious DRE
 Yes82 (1%)830 (8%)5.66 (4.52-7.18)<0.0001
 No5,726 (99%)9,352 (92%)1.0 (reference)
  • *From a multivariable logistic regression model with endpoint evaluated (yes/no) as the outcome, adjusting for other factors in Table 1. OR = odds ratio, CI = confidence interval.

  • The OR represents the difference in odds of endpoint comparing men 1-year apart in age.

  • OR for each 1-unit increase in PSA level.