Childhood cancer (CC) is the highest mortality-causing disease for children under 15 years, and the incidence in Canada is increasing. Air pollution has been associated with the development and progression of CC, but the role of spatiotemporally varying industrial sources is understudied. We assessed the association between industrial air emissions and CC by performing a population-based matched case-control study in Alberta, Canada. Using a nationwide database, we studied 1320 children ages 0-14 with a primary CC diagnosis from 2003 to 2019. Each case was matched to six controls based on birth month, year, and sex from provincial birth records. Industrial air emissions were linked to participant postal codes using a wind-based exposure assessment. We aggregated chemicals into six groups based on reporting requirements as monthly averages beginning one year before birth until the month of diagnosis. Using conditional logistic regression models (CLR) we analyzed all cancers and subgroups for leukemias, lymphomas, CNS tumours, and other cancers. Our analysis included 1320 cases and 7920 controls. In adjusted CLR models, we found positive associations in models combining all cancers for multiple groups, notably: core substances (Q4 OR = 3.22, 95% CI: 2.60-4.00), alternate threshold substances (Q4 OR = 1.77, 95% CI: 1.44-2.17), criteria air contaminants (Q4 OR = 6.06, 95% CI: 4.87-7.55), and speciated volatile organic compounds (Q4 OR = 2.52, 95% CI: 2.04-3.10). We did not find evidence of a different effect on CC types. Emissions from industrial sources may be worth considering as additional air pollutants in CC research.
Syer et al. (Thu,) studied this question.