Pickett W, King W D, Lees R E, Bienefeld M, Morrison H I, Brison R J
Department of Community Health and Epidemiology, Queen's University, Kingston, Ontario, Canada.
Am J Ind Med. 1998 Oct;34(4):364-72. doi: 10.1002/(sici)1097-0274(199810)34:4<364::aid-ajim10>3.0.co;2-0.
An exploratory, case-control study was used to investigate a new hypothesis about suicide among farm operators. This hypothesis suggested a biologically plausible link between exposures to certain pesticides and the occurrence of suicide among farm operators. These analyses were based on data from the Canadian Farm Operator Cohort.
Canadian male farm operators who committed suicide between 1971-1987 (n = 1,457) were compared with a frequency matched (by age and province) sample of control farm operators (n = 11,656) who were alive at the time of death of individual cases. Comparisons focused on past exposures to pesticides reported to the 1971 Canada Census of Agriculture.
Multivariate logistic regression analyses indicated no associations between suicide and (1) acres sprayed with herbicides, (2) acres sprayed with insecticides, and (3) the costs of agricultural chemicals purchased; after controlling for important covariates. There was, however, a suggestive increase in risk for suicide associated with herbicide and insecticide spraying among a subgroup of farm operators who were most likely to be directly exposed to pesticides: OR = 1.71 (95% CI = 1.08-2.71) for 1-48 vs. 0 acres sprayed. Additional risk factors that were identified included seasonal vs. year-round farm work (OR = 1.68; 95% CI = 1.15-2.46); and high levels of paid labor on the farm (e.g., OR = 1.61; 95% CI = 1.24-2.10, for > 13 vs. 0 weeks per year). Factors that were protective included marriage (odds ratio (OR) = 0.69; 95% confidence interval (CI) = 0.58-0.81), having more than one person resident in the farm house (e.g., two vs. one person; OR = 0.62; 95% CI = 0.42-0.92); and higher levels of education (e.g., postsecondary vs. primary; OR = 0.40; 95% CI = 0.17-0.96).
This study does not provide strong support for the main hypothesis under study, that exposure to pesticides is an important risk factor for suicide among farmers. Although secondary to the main hypothesis, a number of other risk factors for suicide were suggested. These have implications for the future study and targeting of suicide prevention programs in rural Canada.
一项探索性病例对照研究用于调查关于农场经营者自杀的新假说。该假说表明接触某些农药与农场经营者自杀事件之间存在生物学上合理的联系。这些分析基于加拿大农场经营者队列的数据。
将1971年至1987年间自杀的加拿大男性农场经营者（n = 1457）与在个别病例死亡时仍在世的对照农场经营者（n = 11656）的频率匹配样本（按年龄和省份）进行比较。比较重点是向1971年加拿大农业普查报告的过去农药接触情况。
多变量逻辑回归分析表明，在控制重要协变量后，自杀与以下因素之间无关联：（1）喷洒除草剂的英亩数，（2）喷洒杀虫剂的英亩数，以及（3）购买农业化学品的成本。然而，在最有可能直接接触农药的农场经营者亚组中，与除草剂和杀虫剂喷洒相关的自杀风险有提示性增加：喷洒1 - 48英亩与0英亩相比，比值比（OR）= 1.71（95%置信区间（CI）= 1.08 - 2.71）。确定的其他风险因素包括季节性与全年性农场工作（OR = 1.68；95% CI = 1.15 - 2.46）；以及农场高有偿劳动水平（例如，每年> 13周与0周相比，OR = 1.61；95% CI = 1.24 - 2.10）。具有保护作用的因素包括婚姻（比值比（OR）= 0.69；95%置信区间（CI）= 0.58 - 0.81）、农场房屋中有不止一人居住（例如，两人与一人相比；OR = 0.62；95% CI = 0.42 - 0.92）；以及较高的教育水平（例如，高等教育与小学教育相比；OR = 0.40；95% CI = 0.17 - 0.96）。
本研究未为所研究的主要假说提供有力支持，即接触农药是农民自杀的重要风险因素。尽管次要于主要假说，但提出了一些其他自杀风险因素。这些对加拿大农村地区未来的自杀预防项目研究和目标设定具有启示意义。
