Skip to main content
. 2013 Sep 30;8(9):e75279. doi: 10.1371/journal.pone.0075279

Table 3. Summary risk estimates of the association between the number of parity and colorectal cancer risk, highest vs. lowest parity number.

No. of studies Summary RR (95% CIs) Q Statistic I 2 Value (%) P h * P h **
All studies 11 1.02 (0.89–1.17) 17.48 37.1 0.094
Subgroup analyses 0.611
High quality studies (scores ≥8) 7 1.05 (0.94–1.17) 8.78 31.6 0.186
Low quality studies (scores <8) 4 0.97 (0.72–1.31) 8.08 50.5 0.089
Duration of follow-up 0.641
<10 y 5 1.06 (0.84–1.34) 8.43 52.6 0.077
≥10 y 6 0.99 (0.85–1.14) 8.54 29.7 0.201
Number of cases 0.847
<500 5 1.00 (0.79–1.26) 6.82 41.4 0.145
≥500 6 1.03 (0.93–1.15) 10.59 43.4 0.102
Geographic location 0.339
America 7 1.08 (0.88–1.34) 11.65 48.5 0.070
Europe 2 1.06 (0.90–1.23) 4.23 52.7 0.121
Asia 2 0.85 (0.61–1.17) 0.04 0 0.833
Anatomic cancer site 0.977
Colon 8 1.04 (0.87–1.24) 10.67 34.4 0.154
Rectum 6 1.04 (0.81–1.35) 4.57 0 0.470
Cancer subsite of colon 0.881
Proximal colon 4 0.98 (0.70–1.36) 5.02 40.3 0.170
Distal colon 4 0.90 (0.64–1.27) 2.99 0 0.393
Adjustment for confounders or important risk factors
Body mass index 0.611
 Yes 7 1.05 (0.94–1.17) 8.78 31.6 0.186
 No 4 0.97 (0.72–1.31) 8.08 50.5 0.089
Diabetes mellitus 0.787
 Yes 2 1.05 (0.92–1.20) 2.38 58.0 0.123
 No 9 1.01 (0.84–1.21) 14.87 39.5 0.095
Physical activity 0.886
 Yes 6 1.04 (0.93–1.16) 8.36 40.2 0.138
 No 5 1.00 (0.84–1.20) 9.02 44.6 0.108
Cigarette smoking 0.611
 Yes 7 1.05 (0.94–1.17) 8.78 31.6 0.186
 No 4 0.97 (0.72–1.31) 8.08 50.5 0.089
Alcohol drinking 0.886
 Yes 6 1.04 (0.93–1.16) 8.36 40.2 0.138
 No 5 1.00 (0.84–1.20) 9.02 44.6 0.108
HRT use 0.616
 Yes 4 1.03 (0.89–1.19) 5.68 47.2 0.128
 No 7 1.03 (0.90–1.17) 11.80 40.7 0.107
Family history of colorectal cancer/adenomatous polyposis 0.631
 Yes 5 0.95 (0.84–1.07) 5.97 33.0 0.201
 No 6 0.95 (0.87–1.03) 10.38 42.2 0.110

RR: relative risk; CI: confidence interval; HRT: hormone replacement therapy.

*

P value for heterogeneity within each subgroup.

**

P value for heterogeneity between subgroups with meta-regression analysis.