Van Maele-Fabry G, Willems J L
Department of Public Health, Ghent University, De Pintelaan 185, B-9000 Ghent, Belgium.
Occup Environ Med. 2003 Sep;60(9):634-42. doi: 10.1136/oem.60.9.634.
To summarise recent literature on the risk of prostate cancer in pesticide related occupations, to calculate the meta-rate ratio, and to compare it to data from meta-analyses previously published.
A meta-analysis of 22 epidemiological studies, published between 1995 and 2001, was conducted in order to pool their rate ratio estimates. Studies were summarised and evaluated for homogeneity and publication bias.
The meta-rate ratio estimate, based on 25 estimators of relative risk from 22 studies, was 1.13 (95% CI 1.04 to 1.22). Significant heterogeneity of rate ratios existed among the different studies. Therefore, a stratified analysis was carried out. Major sources of heterogeneity identified were geographic location, study design, and healthy worker effect. Overall, pooled risk estimates for studies derived from Europe were lower than those derived from the USA/Canada. A significant increase in rate ratio was observed for the occupation category of pesticide applicators, whereas no significant increase was observed for farmers. There was no evidence of publication bias.
This increased meta-rate ratio for prostate cancer in different pesticide related occupations, including farmers, is very similar to three, previously published, meta-rate ratios for prostate cancer in farmers calculated from studies published before 1995. Although the underlying data do not identify pesticide exposure as an independent cause for prostate cancer, the fact that an increased meta-rate ratio is again obtained points to occupational exposure to pesticides as a possible factor. Future epidemiological studies should focus, as far as possible, on reliable methods to estimate actual exposure.
总结近期关于从事与农药相关职业患前列腺癌风险的文献，计算合并率比，并与之前发表的荟萃分析数据进行比较。
对1995年至2001年间发表的22项流行病学研究进行荟萃分析，以汇总其率比估计值。对研究进行总结并评估其同质性和发表偏倚。
基于22项研究的25个相对风险估计值得出的合并率比估计值为1.13（95%可信区间为1.04至1.22）。不同研究之间存在显著的率比异质性。因此，进行了分层分析。确定的异质性主要来源包括地理位置、研究设计和健康工人效应。总体而言，来自欧洲的研究的合并风险估计值低于来自美国/加拿大的研究。农药施用者职业类别的率比显著增加，而农民的率比未观察到显著增加。没有发表偏倚的证据。
在包括农民在内的不同与农药相关职业中，前列腺癌的合并率比增加，这与之前发表的三项根据1995年之前发表的研究计算得出的农民前列腺癌合并率比非常相似。尽管基础数据并未将接触农药确定为前列腺癌的独立病因，但再次获得合并率比增加这一事实表明职业接触农药可能是一个因素。未来的流行病学研究应尽可能关注估计实际接触的可靠方法。
