Patient and provider determinants of breast cancer screening among Ontario women aged 40-49: a population-based retrospective cohort study

Purpose — Canadian breast cancer screening guidelines state that mammography screening for women 40–49 should be individualized based on risk assessment and preferences. This retrospective cohort study describes the frequency of screening in women aged 40–49 and identifies patient and provider-level associations with screening.

Methods — Administrative databases were linked. The overall cohort included Ontario women aged 40–49 between April 1, 2009 and March 31, 2019. Subgroups were created: the “screen” group included women who received a mammogram defined as screening (using a set of exclusion criteria) and the “routine screen” group included women with three or more screening mammograms. A multivariable multilevel logistic regression model accounting for patient and provider characteristics was fit to determine characteristics associated with routine screening. The intracluster correlation co-efficient was used to quantify the degree of variation across providers.

Results — Of approximately 2 million eligible women, there were 532,596 (25.5%) in the screen group and 90,651 (4.3%) the routine screen group. There was an average of 0.30 and 0.52 screening mammograms per woman year, in the screen and routine screen groups, respectively. Routine screening was associated with periodic health exams (OR 1.21, 95% CI 1.20–1.22), receiving pap smears (OR 1.38, 95% CI 1.37–1.39), and fee-for-service models of care (OR 1.32, 95% CI 1.27–1.36). Over 20% of the variation in screening was due to systematic between-provider differences.

Conclusions — Approximately 4.3% of women aged 40–49 in Ontario received routine breast cancer screening with substantial variation across providers. Routine screening is associated with periodic health examinations, receipt of pap smears, and fee-for-service models of care.