Skip to main content
. 2019 Jan 23;2019:7090545. doi: 10.1155/2019/7090545

Table 2.

The multivariable logistic regression for the presence of metastases at diagnosis of prostate cancer.

Pt. no. Multivariable logistic regression analysis Hazard Ratio (95% CI)
Localized With invasion With metastasis Among pt. with invasion P value Among pt. with metastasis P value
Age at diagnosis (y)
 30-50 256 13 17 1 (Reference) NA 1 (Reference) NA
 51-70 4217 185 283 0.86 (0.49-1.54) 0.62 1.01 (0.61-1. 68) 0.97
 71-85 2438 125 325 1.01 (0.56-1.81) 0.97 2.01 (1.21-3. 32) ∗∗0.01
 ≥86 152 13 62 1.68 (0.76-3.73) 0.20 6.14 (3.46-10.89) ∗∗<0.01
Race
 White 5303 238 478 1 (Reference) NA 1 (Reference) NA
 Black 1184 79 157 1.49 (1.14-1.93) ∗∗<0.01 1.47 (1.22-1. 78) ∗∗<0.01
 Asian 495 15 42 0.68 (0.40-1.15) 0.15 0.94 (0.68-1.31) 0.72
 American Indian 21 4 8 4.24 (1.45-12.46) ∗∗0.01 4.23 (1.86-9.59) ∗∗<0.01
Grade
 1 299 6 6 1 (Reference) NA 1 (Reference) NA
 2 4880 139 146 1.42 (0.62-3.24) 0.41 1.49 (0.65-3.40) 0.34
 3 1639 173 315 5.26 (2.31-11.98) ∗∗<0.01 9.58 (4.23-21.68) ∗∗<0.01
 4 26 3 16 5.75 (1.36-24.33) 0.02 30.67 (11.06-85.07) ∗∗<0.01
Number of localized tumors
 1 6218 304 648 1 (Reference) NA 1 (Reference) NA
 ≥2 845 32 39 0.81 (0.61-1.07) 0.13 0.74 (0.61-0. 91) ∗∗<0.01
PAP
 - 2878 242 127 1 (Reference) NA 1 (Reference) NA
 + 4185 94 560 1.78 (1.39-2.26) ∗∗<0.01 3.03 (2.49-3.70) ∗∗<0.01

The multivariable logistic regression was used to determine whether age and race were associated with the presence of invasion and metastases at diagnosis; other variables in the model included pathological grade and number of localized tumors. The likelihood ratio test (LRT) is used in the multivariable logistic regression analysis. Abbreviations: PAP: prostatic acid phosphatase; y: years; Pt. no.: patient number; P ≤ 0.05; ∗∗P ≤ 0.01.