Table 3.
Multivariable logistic regression model for csPCa detection rate on target biopsy (csCDR-T), per-centre and per-operator
| csCDR-T-multivariable regression | Centre 1 | Centre 2 | Operator 1 | Operator 2 | Operator 3 | Operator 4b |
|---|---|---|---|---|---|---|
| Age | ||||||
| OR | 1.082 | 1.081 | 1.045 | 1.176 | 1.004 | 1.332 |
| p | 0.001 | 0.001 | 0.48 | 0.003 | 0.95 | 0.042 |
| Centre experience | ||||||
| OR | 1.002 | 1.004 | - | - | - | - |
| p | 0.088 | 0.001 | ||||
| Operator experience | ||||||
| OR | - | - | 0.999 | 1.023 | 0.984 | 1.037 |
| p | 0.984 | 0.075 | 0.427 | 0.094 | ||
| PSA density [ng/mL/cc] | ||||||
| OR | 222.952 | 43.973 | 0.722 | 3.120.352 | 46.474 | 2.840.006 |
| p | 0.002 | 0.001 | 0.926 | 0.043 | 0.237 | 0.057 |
| Prostate volume [cc] | ||||||
| OR | 0.979 | 0.98 | 0.946 | 0.99 | 0.97 | 0.955 |
| p | 0.001 | 0.024 | 0.019 | 0.407 | 0.402 | 0.266 |
| Positive DRE | ||||||
| OR | 2.376 | 2.954 | 5.297 | 5.643 | 1.961 | 26.384 |
| p | 0.002 | 0.001 | 0.039 | 0.013 | 0.555 | 0.027 |
| PI-RADS 4 lesiona | ||||||
| OR | 3.861 | 5.224 | 1.73 | 5.383 | 3.709 | 3.997 |
| P | 0.001 | 0.001 | 0.633 | 0.05 | 0.314 | 0.451 |
| PI-RADS 5 lesiona | ||||||
| OR | 3.314 | 8.444 | 71.576 | 3.931 | 2.084 | 237.558 |
| p | 0.025 | 0.001 | 0.037 | 0.213 | 0.671 | 0.014 |
| Target diameter [mm] | ||||||
| OR | 1.157 | 1.009 | 0.837 | 1.245 | 1.087 | 0.887 |
| p | 0.001 | 0.752 | 0.161 | 0.01 | 0.353 | 0.294 |
| Positive family history | ||||||
| OR | 0.812 | 3.822 | 0.211 | 0.825 | - | 61.396 |
| p | 0.587 | 0.023 | 0.224 | 0.809 | 0.008 |
The 95% confidence intervals for this table are shown in Supplementary Table 6B
Significant p values (below 0.05) are in bold
DRE digital rectal examination, PI-RADS Prostate Index Reporting and Data System score, OR odds ratio
aReference is PI-RADS 3 lesion
bFirth procedure for logistic regression used for Operator 4