Cancer and the physicochemical quality of drinking water in Quebec

The science of Medical Geography permits the delineation of maps of illness using different scales to compare the factors of incidence of a disease. The identification of differences in risk according to population and region facilitates the elaboration of ecological hypotheses relative to the cause of disease. Cancer in humans is an illness of high frequency in our society. According to the most recent Canadian statistics it is in Quebec that the risk of being a cancer victim is the greatest in this country. During the years 1970–1972 Quebec males registered the highest frequency for the ensemble of seats of cancer, while Quebec females ranked directly after those of Nova Scotia. In addition, cancer remains one of the illnesses where environmental agents may exercise a determinant role. Certain experts consider that the direct or indirect dependence upon environmental factors applies in 80 to 90% of cancer cases.

To establish the geographic location of cancers in Quebec we have used figures supplied by Statistics Canada for census regions (R.R.I of Quebec. We have taken the number of new cases of malignant tumors observed during the year for each type of tumor figuring in the international classification of illnesses for a period of 5 years that is from 1970–1975. These figures indicate the rates of incidence of a specific cancer observed during the period in question and allow us to calculate the ratio of incidence standardized according to age (SIR) in order to analyze the incidence of malignant tumors in the various census regions.

The objective of this study was to discover the possible relation between the quality of drinking water and cancer. More particularly, we think that certain cancers are more prevalent in regions where the water is considered “soft”. To establish this relation we grouped the R.R. by sites of cancer in two groups: one where the SIR was high and one where the SIR was low. The R.R. of less than 10 cases of cancer per year were eliminated. We then compared the R.R. of high and low rates of incidence according to physicochemical parameters of drinking water as per 1974–1976 observational data. The analysis of correlations and multiple regression were used to evaluate the significant associations.