Skip to main content
. 2015 Jun 2;15:447. doi: 10.1186/s12885-015-1448-x

Table 7.

Subgroup analysis for disease-specific survival

No. of included articles No. of cases Pooled HR (95 % CI) Chi2 (p value) I2 Ph*
Publication year 0.0517
2003–2010 4 375 1.61 (1.09–2.38) 7.42 (0.06) 60 %
2011–2014 6 2591 1.51 (1.16–1.97) 12.22 (0.03) 59 %
Region 0.3206
USA 3 1979 1.57 (1.10–2.22) 11.45 (0.003) 83 %
Europe 5 1679 1.54 (1.02–2.31) 14.19 (0.007) 72 %
Asia 2 261 1.61 (1.08–2.39) 0.46 (0.5) 0 %
No. of patients 0.0015
<100 5 331 1.40 (0.82–2.40) 12.35 (0.01) 68 %
≥100 5 2635 1.55 (1.23–1.95) 14.06 (0.007) 72 %
Median follow-up* 0.0017
<36 months 4 1643 1.85 (1.53–2.24) 2.58 (0.46) 0 %
≥36 months 5 544 1.20 (0.91–1.59) 7.43 (0.11) 46 %
Analysis results 0.1626
Not significant 4 382 1.14 (0.69–1.87) 7.37 (0.06) 59 %
Significant 6 2584 1.69 (1.31–2.17) 18.98 (0.002) 74 %
Quality scale 0.0233
≤4 7 2361 1.60 (1.26–2.03) 16.99 (0.009) 65 %
>4 3 605 1.08 (0.51–2.30) 8.91 (0.01) 78 %

HR: hazard ratio, CI: confidence interval

Ph* for heterogeneity between subgroups with meta-regression analysis

*One study was excluded because the duration of follow-up was not available (Morgan [26])