Regional variations in human chemical exposures in Canada: A case study using biomonitoring data from the Canadian Health Measures Survey for the provinces of Quebec and Ontario
The Canadian Health Measures Survey (CHMS), an ongoing national health survey conducted in two-year cycles, collects extensive biomonitoring data that is used to assess the exposure of Canadians to environmental chemicals of concern. Combining data from multiple cycles of the CHMS allows for the calculation of robust regional estimates of chemical concentrations in blood and urine. The objective of this work was to compare biomarkers of exposure to several environmental chemicals for the provinces of Quebec and Ontario, two major CHMS regions, as well as the entire CHMS (representing Canada) minus Quebec (CMQ), and the entire CHMS minus Ontario (CMO), and to interpret differences between regions. Geometric means and 95th percentiles of blood and/or urinary concentrations of 45 environmental chemicals or their metabolites for Ontario, Quebec, CMQ, and CMO were calculated by combining the two most recent cycles of data available for a chemical (cycles 1 and 2, or cycles 2 and 3) from the first three cycles of the CHMS (2007-2013). Weighted one-way ANOVA was used to test the differences between regional estimates. After applying a Bonferonni-Holm adjustment for multiple comparisons, the following measures were significantly higher in Quebec as compared to Ontario and CMQ: blood lead, urinary lead and the urinary polyaromatic hydrocarbon (PAH) metabolites, 9-hydroxyfluorene, 1-hydroxyphenanthrene, 2- hydroxyphenanthrene and 3-hydroxyphenanthrene. In Quebec compared to CMQ only, urinary 2-hydroxfluorene, 3-hydroxyfluorene, 2-hydroxynaphthalene, and 4-hydroxyphenanthrene were higher. The concentration of urinary fluoride was significantly higher in Ontario as compared to Quebec and CMO. Blood manganese and urinary fluoride were significantly lower in Quebec compared to CMQ, and blood and urinary selenium were significantly lower in Ontario compared to CMO. Regional differences in tobacco use, age of dwellings and drinking water fluoridation are among the possible contributing factors to some of the observed differences. In conclusion, this is the first study where biomonitoring data from multiple cycles of CHMS were combined in order to generate robust estimates for subsets of the Canadian population. Such assessments can contribute to a regional-level prioritization of control measures to reduce the exposure of Canadians to chemicals in their environment.