Adherence to Colorectal Cancer Screening Among Culturally and Linguistically Diverse Low-Income Patients – Does Patient-Provider Language Concordance Matter?

Abstract

Introduction: Research studies have shown that colorectal cancer (CRC) screening among primary care patients can be increased by using different types of patient-oriented behavioral interventions. However the impact of such interventions on CRC screening disparities has not been studied. Aim: We report the differential response of white and African American primary care patients to behavioral interventions intended to increase CRC screening. Methods: A randomized trial was conducted with 1,546 patients who were 50-74 years of age and were eligible for CRC screening. Participants were randomized to one of 4 groups: Control (Usual Care), Standard Intervention (SI), Tailored Intervention (TI), or Tailored Intervention plus Phone (TIP). The SI Group received a targeted intervention by mail. The TI Group received the targeted intervention with tailored message pages. The TIP Group received the targeted intervention, tailored message pages, and a telephone reminder. Screening was measured completion of any guidelines-recommended test following randomization. Multivariable analyses were completed to assess intervention impact on screening and racerelated screening disparity. Results: For 1,430 study participants with complete data (578 whites and 852 African Americans), CRC screening rates were significantly higher (p<0.001) in the SI, TI, and TIP groups (47%, 44%, and 50%) versus control group patients (32%). Although, screening rates for whites and African Americans were comparable in usual care (33% vs. 32%, respectively; OR=1.01, 95% CI=0.64-1.61), whites exhibited significantly higher rates than African Americans in response to any intervention (53% and 43%, respectively; OR=1.44, 95% CI=1.12-1.86). This effect is largely accounted for by exposure to the SI Group mailed print intervention. Exposure to mailed materials in the SI Group resulted in more screening among whites than African Americans (55% and 41% respectively, OR = 1.68; 95% CI=1.10, 2.58) (Table 1). Conclusions: Findings from this study suggest that whites and African Americans may respond differently to interventions intended to increase CRC screening. Mailed interventions may serve to increase CRC screening, but may also produce unintended disparities in screening adherence. Table 1. Multivariable Analysis of CRC Screening by Study Group and Race (n=1,430)