Abstract
Introduction
Completion of colorectal cancer (CRC) screening is essential to realizing the benefits of early detection and prevention; however, many individuals do not complete screening after a test is provided. The association between screening completion and subsequent CRC incidence remains poorly understood. We evaluated the association between multitarget stool DNA (mt-sDNA) test kit return and CRC incidence among average-risk adults.Methods
This retrospective cohort study, linking claims data from Komodo Health with Abbott laboratory data, included adults aged 45-85 years who received an mt-sDNA kit between January 1, 2017, and December 31, 2019 (based on linked laboratory and claims data). The primary outcome was incident CRC identified using claims-based diagnosis codes. Propensity score-matched (PSM) and inverse probability treatment-weighted (IPTW) analyses assessed robustness to confounding. Adjusted incidence rate ratios (IRRs) were estimated using Poisson regression and hazard ratios (HRs) using Cox proportional hazards models.Results
The cohort included 2,574,759 adults (60.5% female, 38.5% male; 68.2% White, 8.2% Black/African American, 6.5% Hispanic, 2.6% Asian/Pacific Islander, 14.4% other/unknown race), who contributed over 14 million person-years of follow-up. CRC incidence was lower among kit returners than non-returners (248.95 vs 301.89 per 100,000 person-years). In adjusted analyses, mt-sDNA kit return was associated with 55% lower CRC incidence versus non-return [IRR 0.45; 95% confidence interval (CI) 0.45-0.46; p Conclusions
In this large retrospective cohort study, mt-sDNA screening completion was consistently associated with lower CRC incidence across unmatched, PSM, and IPTW analyses. These findings support strategies to improve stool-based CRC screening completion to maximize the population-level benefits of CRC screening programs.