Skip to main content
. Author manuscript; available in PMC: 2012 Oct 1.
Published in final edited form as: JAMA. 1992 Apr 22;267(16):2215–2220.

Table 2.

Observed and Predicted Rates of Change in Prostate-Specific Antigen Levels (μg/L per year)*

Cancer Cases
Time Interval Before Diagnosis, y Controls BPH Cases Local or Regional Metastatic Significance
Tests
Zero-5.0
 Observed median (observed range) 0.03 (−0.30-0.56) 0.12 (−0.32-0.53) 0.88 (−0.04-13.83) 5.35 (1.74-7.69) LR, M>C, BPH

Predicted 0.04 0.18 0.75 4.40

 5.0-10.0
Observed median (observed range) 0.01 (−0.28-0.37) 0.09 (−0.28-1.1) 0.27 (−0.04-1.78) 1.33 (−0.07-1.83) LR>C, BPH

Predicted 0.03 0.14 0.37 1.26

 10.0-15.0
Observed median (observed range) 0.02 (−0.07-0.28) 0.09 (−0.14-0.62) 0.14 (−0.01-0.37) 0.30 (0.12-1.18) LR>C; M>C, BPH

Predicted 0.02 0.10 0.13 0.42
*

Predicted from the mixed-effects regression model assuming an age at diagnosis of 75.0 years. BPH indicates benign prostatic hyperplasia; LR, local or regional cancer; M, metastatic cancer cases; and C, controls.

Results of pairwise comparisons using Mann-Whitney U tests.