Analysis specification | No. of studies | Study heterogeneity | Effects model | Pooled HR (95% CI) | p value | |
---|---|---|---|---|---|---|
I2 (%) | p heterogeneity | |||||
Univariate analysis (BCR) | ||||||
 Overall | 25 | 70.9 | < 0.001 | Random | 1.56 (1.46,1.66) | < 0.001 |
 Geographical region | ||||||
  Asia | 12 | 72.1 | < 0.001 | Random | 1.61 (1.43,182) | < 0.001 |
  Europe and North America | 12 | 70.8 | < 0.001 | Random | 1.50 (1.37,1.65) | < 0.001 |
 Date of publication | ||||||
  ≥ 2014 | 13 | 81.8 | < 0.001 | Random | 1.52 (1.36,1.70) | < 0.001 |
  < 2014 | 12 | 18.5 | 0.262 | Fixed | 1.61 (1.52,1.71) | < 0.001 |
 Mean age (years) | ||||||
  ≥ 64 | 9 | 84 | < 0.001 | Random | 1.62 (1.34,1.97) | < 0.001 |
  < 64 | 15 | 55.6 | 0.005 | Random | 1.54 (1.45,1.64) | < 0.001 |
 Sample size (cases) | ||||||
  ≥ 500 | 10 | 40.1 | 0.09 | Random | 1.61 (1.52,1.70) | < 0.001 |
  < 500 | 15 | 76.9 | < 0.001 | Random | 1.51 (1.33,1.71) | < 0.001 |
 Mean p-PSA (ng/ml) | ||||||
  ≥ 10 | 7 | 81 | < 0.001 | Random | 1.65 (1.38,1.97) | < 0.001 |
  < 10 | 14 | 58.5 | 0.003 | Random | 1.59 (1.48,1.71) | < 0.001 |
 Median follow-up | ||||||
  ≥ 36 months | 11 | 77.1 | < 0.001 | Random | 1.49 (1.33,1.67) | < 0.001 |
  < 36 months | 14 | 59.8 | 0.002 | Random | 1.61 (1.49,1.74) | < 0.001 |
 BCR (ng/ml) | ||||||
  Cutoff value 0.1 | 4 | 0 | 0.775 | Fixed | 1.61 (1.49,1.72) | < 0.001 |
  Cutoff value 0.2 | 20 | 72 | < 0.001 | Random | 1.58 (1.46,1.70) | < 0.001 |
  Cutoff value 0.4 | 1 | – | – | – | – | – |
Multivariate analysis (BCR) | ||||||
 Overall | 32 | 79.2 | < 0.001 | Random | 1.35 (1.27,1.43) | < 0.001 |
 Geographical region | ||||||
  Asia | 14 | 67 | < 0.001 | Random | 1.42 (1.29,1.55) | < 0.001 |
  Europe and North America | 15 | 84.7 | < 0.001 | Random | 1.31 (1.19,1.43) | < 0.001 |
  Multi-centred | 3 | 71.9 | 0.029 | Random | 1.33 (1.00,1.78) | 0.053 |
 Date of publication | ||||||
  ≥ 2014 | 16 | 82.9 | < 0.001 | Random | 1.27 (1.17,1.39) | < 0.001 |
  < 2014 | 16 | 67.2 | < 0.001 | Random | 1.44 (1.32,1.56) | < 0.001 |
 Mean age (years) | ||||||
  ≥ 64 | 8 | 62.5 | 0.009 | Random | 1.56 (1.32,1.85) | < 0.001 |
  < 64 | 22 | 81.5 | < 0.001 | Random | 1.33 (1.24,1.43) | < 0.001 |
 Sample size (cases) | ||||||
  ≥ 500 | 18 | 77.1 | < 0.001 | Random | 1.40 (1.32,1.49) | < 0.001 |
  < 500 | 14 | 76.8 | < 0.001 | Random | 1.28 (1.12,1.47) | < 0.001 |
 Mean p-PSA (ng/ml) | ||||||
  ≥ 10 | 7 | 80.8 | < 0.001 | Random | 1.36 (1.22,1.57) | < 0.001 |
  < 10 | 19 | 79 | < 0.001 | Random | 1.35 (1.24,1.48) | < 0.001 |
 Median follow-up | ||||||
  ≥ 36 months | 16 | 79.6 | < 0.001 | Random | 1.36 (1.24,1.46) | < 0.001 |
  < 36 months | 15 | 79.8 | < 0.001 | Random | 1.34 (1.21,1.47) | < 0.001 |
 BCR (ng/ml) | ||||||
  Cutoff value 0.1 | 5 | 87.7 | < 0.001 | Random | 1.22 (1.01,1.48) | 0.044 |
  Cutoff value 0.2 | 23 | 71.3 | < 0.001 | Random | 1.39 (1.30,1.48) | < 0.001 |
  Cutoff value 0.4 | 4 | 82.2 | 0.001 | Random | 1.34 (1.15,1.57) | < 0.001 |