Predictors of nonadherence to breast cancer screening guidelines in a United States urban comprehensive cancer center

Abstract

AbstractBackgroundThis study aimed to identify predictors of nonadherence to breast cancer screening guidelines in an urban screening clinic among high‐ and average‐risk women in the United States.MethodsWe reviewed records of 6090 women who received ≥2 screening mammograms over 2 years at the Karmanos Cancer Institute to examine how breast cancer risk and breast density were associated with guideline‐concordant screening. Incongruent screening was defined as receiving supplemental imaging between screening mammograms for average‐risk women, and as not receiving recommended supplemental imaging for high‐risk women. We used t‐tests and chi‐square tests to examine bivariate associations with guideline‐congruent screening, and probit regression to regress guideline‐congruence unto breast cancer risk, breast density, and their interaction, controlling for age and race.ResultsIncongruent screening was more likely among high‐ versus average‐risk women (97.7% vs. 0.9%, p < 0.01). Among average‐risk women, incongruent screening was more likely among those with dense versus nondense breasts (2.0% vs. 0.1%, p < 0.01). Among high‐risk women, incongruent screening was more likely among those with nondense versus dense breasts (99.5% vs. 95.2%, p < 0.01). The significant main effects of density and high‐risk on increased incongruent screening were qualified by a density by high‐risk interaction, showing a weaker association between risk and incongruent screening among women with dense breasts (simple slope = 3.71, p < 0.01) versus nondense breasts (simple slope = 5.79, p < 0.01). Age and race were not associated with incongruent screening.ConclusionsLack of adherence to evidence‐based screening guidelines has led to underutilization of supplementary imaging for high‐risk women and potential overutilization for women with dense breasts without other risk factors.