[Pesticides and cancer in a Quebec rural farming population: a geographical interpretation].

Data on the use of pesticides in agriculture and three types of cancer among the rural farm population were compiled for 34 drainage basins of the province of Québec for 1982 and 1983. The basins were divided following three categories of exposure (low, intermediate, high) on the basis of the level of pesticides sales. Results at the scale of the three categories show that among women living in rural farm areas, the relative risk (RR) for cancers of the lymphatic tissues is of 1.62 when highly exposed basins were compared to low exposed basins (95% Confidence Interval = 1.16; 2.25), and of 1.63 (95% CI = 1.08; 2.45) when they were compared to basins with intermediate exposure. For women of 35-64 age group, RR values are respectively of 3.15 (95% CI = 1.67; 5.93) in the first case, and 3.73 (95% CI = 1.54; 9.01) in the second case. By another way, at the level of the 34 drainage basins, correlation analysis have shown statistically significant positive associations between several variables representing pesticides utilization in agriculture and geographic variations of some cancers, according to sex and age groups, especially for the following: men, cancer of the brain (1-14 and 15-34 age groups), cancers of the lymphatic tissues (34-64) and leukaemia (1-14); for women, cancer of the brain (35-64), the lymphatic tissues (35-64, 65 and more, all ages) and leukaemia (15-34, 35-64, 65 and more and all ages). After that, principal components analysis has been performed using variables representing pesticides utilization and socio-economic status by basin. Height factors have been produced, including two socio-economic and six pesticides-factors. Factor scores have been introduced in multiple regression analysis to see the role of each factor in the explanation of spatial distribution of those cancers who had been identified in the correlation analysis before. Results have shown that in most of the cases, only pesticides-factors have significant relationship with those cancers. Socio-economic factors play a significant role only for the following cancers: men, lymphatic tissues, 65 years old and more; women, cancer of the brain all ages, leukaemia 1-14 and 65 years old and more. However, these results must be interpreted with caution for the goal of the study is to describe potential relationships.(ABSTRACT TRUNCATED AT 400 WORDS)