Table 2.
The results of subgroup analyses and metaregression analyses on the association between night shift work and the risk of cancers.
| Subgroup | No. of studies | Weight (%) | OR (95% CI) | I 2 | P for heterogeneity | P ∗ for interaction |
|---|---|---|---|---|---|---|
| Shift schedulea | 0.570 | |||||
| Rotating shift | 29 | 46.97 | 1.14 (1.04–1.24) | 68.7% | <0.001 | |
| Fixed shift | 9 | 11.19 | 1.09 (0.90–1.31) | 51.1% | 0.037 | |
| Mixed shift | 27 | 41.84 | 1.20 (0.82–1.77) | 80.7% | <0.001 | |
| Type of cancerb | 0.298 | |||||
| Digestive system cancer | 11 | 15.72 | 1.15 (1.01–1.32) | 40.2% | 0.081 | |
| Hematological system cancer | 5 | 9.12 | 1.08 (0.99–1.17) | 54.7% | 0.066 | |
| Prostate cancer | 11 | 16.10 | 1.26 (1.05–1.52) | 73.2% | <0.001 | |
| Breast cancer | 37 | 39.62 | 1.22 (1.08–1.38) | 81.2% | <0.001 | |
| Reproductive system cancer | 6 | 7.99 | 1.06 (0.85–1.32) | 49.5% | 0.078 | |
| Lung cancer | 5 | 7.53 | 1.08 (0.87–1.35) | 53.4% | 0.073 | |
| Skin cancer | 3 | 3.92 | 0.93 (0.50–1.74) | 74.9% | 0.019 | |
| Region | 0.298 | |||||
| Australia | 3 | 5.03 | 0.91 (0.77–1.06) | 0.0% | 0.728 | |
| Europe | 27 | 48.92 | 1.18 (1.10–1.28) | 75.1% | <0.001 | |
| Asia | 11 | 14.29 | 1.11 (0.88–1.39) | 78.3% | <0.001 | |
| North America | 17 | 31.75 | 1.16 (1.04–1.31) | 76.1% | <0.001 | |
| Occupation | 0.795 | |||||
| Unclassified occupation | 35 | 61.45 | 1.17 (1.10–1.25) | 69.7% | <0.001 | |
| Industry | 9 | 12.09 | 1.00 (0.81–1.24) | 72.8% | <0.001 | |
| Nurses | 14 | 26.46 | 1.17 (1.02–1.35) | 80.6% | <0.001 | |
| Study design | 0.845 | |||||
| Case-control study | 21 | 32.23 | 1.28 (1.15–1.42) | 66.5% | <0.001 | |
| Nested case-control study | 6 | 9.08 | 1.30 (0.89–1.90) | 88.0% | <0.001 | |
| Cohort study | 31 | 58.69 | 1.07 (1.00–1.15) | 70.9% | <0.001 | |
| Exposure assessment | 0.075 | |||||
| Questionnaire | 28 | 53.99 | 1.08 (1.00–1.17) | 77.1% | <0.001 | |
| Interview | 24 | 34.52 | 1.32 (1.17–1.49) | 66.3% | <0.001 | |
| Database | 6 | 9.06 | 1.00 (0.85–1.18) | 77.1% | 0.004 | |
| Number of adjusted variables | 0.926 | |||||
| ≤4 | 15 | 22.84 | 1.13 (0.99–1.28) | 82.7% | <0.001 | |
| >4 | 43 | 77.16 | 1.16 (1.08–1.24) | 72.8% | <0.001 | |
| Study score | 0.585 | |||||
| Low quality | 17 | 25.20 | 1.16 (0.98–1.37) | 86.9% | <0.001 | |
| High quality | 41 | 74.80 | 1.14 (1.08–1.21) | 61.6% | <0.001 |
a,bFive studies report their studies including different kinds of cancer; nine articles report their studies including different types of shift schedules. ∗ P values for metaregression.