JAMA Netw Open. 2026 Jul 1;9(7):e2623089. doi: 10.1001/jamanetworkopen.2026.23089.
ABSTRACT
IMPORTANCE: Childhood cancer is a leading cause of disease-related mortality, with increasing incidence in many high-income countries. Socioeconomic disparities in childhood cancer survival are well documented, but less is known about incidence.
OBJECTIVE: To examine associations between area-based deprivation and childhood cancer incidence in a universal health care system.
DESIGN, SETTING, AND PARTICIPANTS: This population-based cross-sectional study linked data from the population-based Pediatric Oncology Group of Ontario Networked Information System cancer registry with Statistics Canada’s postal code and Ontario Marginalization Index (ON-Marg) databases to identify pediatric cancer cases diagnosed in Ontario from January 1, 1999, to December 31, 2023, among approximately 2.3 million children aged 0 to 14 years living in Ontario, Canada. Data analysis was performed from October 2025 to February 2026.
EXPOSURES: ON-Marg dimensions: material resources (income and educational attainment), households and dwellings (residential instability), age and labor force (dependents and adults not in the labor force), and racialized and newcomer populations.
MAIN OUTCOMES AND MEASURES: Crude and age-standardized incidence rates for each ON-Marg dimension; incidence rate ratios were estimated using multivariable Poisson regression models adjusted for age, sex, and diagnosis period for the overall population and stratified by cancer subtype (ie, leukemia and lymphoma, central nervous system tumors, and other non-central nervous system solid tumors).
RESULTS: Of the 9063 pediatric cancer cases (4981 [55.0%] male; median [IQR] age at diagnosis, 5 [2-10] years), the age-standardized incidence rate was 162.7 cases per million (95% CI, 159.3-166.1). Most cancer cases were found in areas with the greatest concentration of racialized and newcomer populations; however, after accounting for population size, there was no difference in cancer incidence across quintiles. A key finding of this study was that the least marginalized quintiles of material resources (incidence rate ratio, 1.15; 95% CI, 1.07-1.22) and households and dwellings (incidence rate ratio, 1.08; 95% CI, 1.01-1.16) had significantly higher incidence rates compared with the most marginalized quintiles. These findings were driven primarily by hematologic and other solid cancer types.
CONCLUSION AND RELEVANCE: This large population-based cross-sectional study with virtually complete ascertainment of pediatric cancer cases in a Canadian province with a universal health care system found higher cancer incidence among children who lived in affluent and residentially stable neighborhoods. These findings underscore the importance of considering sociodemographic factors when examining pediatric cancer epidemiology; more research is needed to understand the mechanisms driving these differences, their impact on outcomes, and how the findings may inform targeted prevention and intervention strategies.
PMID:42455571 | DOI:10.1001/jamanetworkopen.2026.23089
