Time Trends in Colorectal Cancer Incidence Rates by Income and Age at Diagnosis in Canada From 1992 to 2016

IMPORTANCE Colorectal cancer (CRC) incidence rates among individuals younger than 50 years have been increasing in many countries, including Canada. Whether changes in CRC incidence rates over time are uniform across income and age groups remains unknown. OBJECTIVE To examine time trends in CRC incidence rates in Canada by area-level average household income by age group from 1992 to 2016. DESIGN, SETTING, AND PARTICIPANTS A population-based, retrospective cohort study was conducted including individuals aged 20 years or older diagnosed with CRC in Canada (excluding Quebec) from 1992 to 2016. Data analysis was performed from February 27 to September 28, 2020. EXPOSURES Average household income was determined by linking an individual's postal code at diagnosis from the Canadian Cancer Registry to Canadian Census data. Average household income was then categorized into quintiles from Q1, the lowest income group, to Q5, the highest income group. MAIN OUTCOMES AND MEASURES Canadian Cancer Registry data were used to determine CRC incidence rates from 1992 to 2016. RESULTS There were 340 790 cases of CRC diagnosed from 1992 to 2016; 11 790 cases (3.5%) were diagnosed in individuals aged 20 to 44 years. Colorectal cancer incidence rates for individuals aged 20 to 44 years increased for all income quintiles, with higher incidence rates in the lower income quintiles. The ratio between the maximum and minimum CRC incidence rates was highest among the 20- to 29-year age group from 1992 to 1996 (ratio, 2.67; 95% CI, 1.47-4.83) and 2012 to 2016 (ratio, 2.00; 95% CI, 1.29-3.10). For individuals aged 45 to 49 years, CRC incidence rates increased only for individuals in the lower income quintiles (eg, incidence in Q1, 1992-1996 vs 2012-2016: 4.54 per 100 000 [95% CI, 4.05-5.03] vs 5.37 per 100 000 [95% CI, 4.91-5.83]), with higher incidence rates in the high income quintiles (eg, incidence rate for Q5 in 1992-1996: 5.92 per 100 000 [95% CI, 5.366.48]). For those aged 50 to 54 years, CRC incidence rates were stable for all income quintiles, with less variability between income quintiles. For individuals aged 55 to 74 years, CRC incidence rates were stable or decreased for all income quintiles (eg, incidence rate for age 55-59 years in Q5, 19921992 vs 2012-2016: 17.97 per 100 000 [95% CI, 16.76-19.18] vs 14.56 per 100 000 [95% CI, 13.80-15.32]), and there was less variability in the rates by income quintile, particularly from 2012 to 2016. After age 75 years, CRC incidence rates were stable or decreased (eg, incidence rate for age 75-79 years, 1992-1996 vs 2012-2016: 66.43 per 100 000 [65.00-67.87] vs 57.34 per 100 000 [56.24-58.45]), were highest for the lower income quintiles, and variability between income quintiles increased relative to younger age groups. CONCLUSIONS AND RELEVANCE In this cohort study, trends in CRC incidence rates in Canada differed by age group and income quintile. These results suggest that, although population-based screening can reduce income disparities, targeted interventions and further research are needed to address the increasing CRC incidence rate among younger individuals in Canada, particularly in the lower income quintiles.Y