Disparities in Screening for Colorectal Cancer Based on Limited Language Proficiency

Abstract

Background This study analyzes the association between limited language proficiency and screening for colorectal cancer. Methods This is a retrospective cohort study from the 2015 sample of the National Health Interview Survey database utilizing univariate and multivariate regression analysis. The study population includes subjects between 50 and 75 years of age. The main outcome analyzed was rates of screening colonoscopies between limited English-language proficiency (LEP) subjects and those fluent in English. Secondary outcomes included analysis of baseline, socioeconomic, access to health care variables, and other modalities for colorectal cancer screening between the groups. Results Incidence of limited language proficiency was 4.8% (n = 1978, count = 4 453 599). They reported lower rates of screening colonoscopies (61% vs 34%, P < .001), less physician recommendation for a colonoscopy (87 vs 60%, P < .001), fewer polyps removed in the previous 3 years (24% vs 9.1%; P < .001), and fewer fecal occult blood samples overall (P < .001). Additionally, Hispanic non-LEP subjects have higher rates of colonoscopies compared to those with language barriers (50% vs 33%, P < .001). On multivariate analysis, LEP was associated with a lower likelihood to have a screening colonoscopy (OR .67 95% CI .49-.91). A second regression model with “Spanish language” and “other language” variables included, associated Spanish speakers with a lower likelihood for a screening colonoscopy (OR .71 95% CI .52-.97) when controlling for baseline, socioeconomic, and access to health care covariates. Discussion Patients with limited English-language proficiency are associated with lower rates of screening for colorectal cancer, in particular the Spanish speaking subgroup.