Table 4. Logistic and Cox regression analysis for identification of factors associated with ERG+ status.

Factor	OR	95% CI	P-value
Logistic regression—univariate
Age quartile 1	1.81	1.35–2.44	0.000*
Age quartile 2	1.18	0.91–1.56	0.260
Age quartile 3	0.90	0.71–1.20	0.460
Age quartile 4	0.53	0.40–0.70	0.000*
PSA range <4	1.65	1.23–2.22	0.001*
PSA range 4–10	0.68	0.52–0.91	0.007*
PSA range ⩾10	0.87	0.59–1.30	0.501
GS group <7	0.91	0.71–0.11	0.446
GS group =7	1.31	1.02–1.69	0.030*
GS group >7	0.65	0.44–0.96	0.290
Logistic regression—multivariate step 1
GS group >7	1.00	-------------	0.27
GS group <7	1.13	0.62–2.07	0.69
GS group =7	1.45	0.84–2.51	0.18
PV, ml	0.99	0.97–1.00	0.05
PSA range ⩾10	1.00		0.58
PSA range <4	0.77	0.40–1.49	0.43
PSA range 4–10	0.72	0.39–1.33	0.30
Age quartile 4	1.00	-------------	0.05
Age quartile 1	2.03	1.17–3.50	0.01*
Age quartile 2	1.41	0.87–2.28	0.17
Age quartile 3	1.07	0.66–1.75	0.78
fPSA%	0.99	0.96–1.02	0.65
Logistic regression—multivariate final backward step 4
Age quartile 4	1.00	-------------	0.036*
Age quartile 1	2.05	1.22–3.44	0.007*
Age quartile 2	1.42	0.89–2.28	0.146
Age quartile 3	1.01	0.67–1.76	0.673
PV, ml	0.99	0.99–1.00	0.031*
Cox regression (inclusion model)
	RR	95% CI	P-value
ERG+	1.25	1.09–1.44	0.001*
PSA	0.96	0.94–0.97	<0.001*

Abbreviations: CI, confidence interval; GS, gleason score; OR, odds ratio; PV, prostate volume.

*P<0.05.

Multivariate logistic regression was performed using a stepwise backward elimination model. OR values of 1.00 indicate reference values with no CI computable. Data of the first and the last steps are shown. For Cox regression the inclusion model was applied.