Skip to main content
. 2020 Jul 23;10:1207. doi: 10.3389/fonc.2020.01207

Table 4.

Results of the multi-variable logistic regression analysis.

Predictors Coeff ± dev.std. p-value OR CI (95%)
FREQUENCY—Frequency of urine loss
Endpoint: ICIQ3 >3, N = 52/148 (35%), excluded: ICIQ3 >3 at baseline
Baseline score 0.655 ± 0.174 <0.001 1.93 1.38–2.74
EQD2(0.8) [Gy] 0.075 ± 0.033 0.024 1.08 1.01–1.15
Intercept −6.964
H&L = 0.78 Slope = 1.06 R2 = 0.82 Brier score = 0.198
(optimism = 0.01)
AMOUNT—Amount of Urine Loss
Endpoint: ICIQ4 >2, N = 40/188 (21%), excluded: ICIQ4 >2 at baseline
Baseline score 0.648 ± 0.237 0.006 1.91 1.24–3.19
Intercept −2.269
H&L = 1.00 Slope = 1.00 R2 = 1.00 Brier score = 0.160
(optimism = 0.005)
OBJECTIVE—Objective
Endpoint: ICIQ3+4 >5, N = 52/148 (35%), excluded: ICIQ3+4 >5 at baseline
Baseline score 0.371 ± 0.099 <0.001 1.45 1.20–1.77
EQD2(0.8) [Gy] 0.077 ± 0.033 0.022 1.08 1.01–1.16
Intercept −7.220
H&L = 0.97 Slope = 1.11 R2 = 0.92 Brier score = 0.199
(optimism = 0.009)

The endpoints are related to frequency, amount and objective component of urinary incontinence, as reported in ICIQ-SF.

H&L, Hosmer and Lemeshow test; ICIQ-SF, International Consultation on Incontinence Modular Questionnaire Short Form.