摘要：There is controversial evidence from cross-sectional, ecological, cohort and case-control studies thatarsenic exposure in drinking water is a risk factor for stillbirth. The objectives of this meta-analysiswere to evaluate the association of stillbirth and arsenic exposure in drinking water and to improve theprecision of estimates. A database search was conducted to identify relevant studies. The odds ratios(ORs) extracted from each study were pooled by using Mantel-Haenszel fixed effect model, whilesubgroup analysis was conducted by using random effect model. Heterogeneity of ORs in the includedstudies was analysed using I2 statistics. Publication bias was tested by Begg’s test. Eight studiesinvolving 51,476 participants met the inclusion criteria. The pooled ORs of studies was 1.54 (95% CI:1.32-1.79) with a small heterogeneity (I² = 25%, P =0.23) across studies and no publication bias wasevident. This meta-analysis provided evidence that chronic arsenic exposure above 50 µg.L-1 indrinking water is a significant risk factor for stillbirth.
