Characterizing patient attrition across the colorectal cancer screening pathway: Insights from a large, urban direct-access colonoscopy navigation program.

Abstract

86 Background: Colorectal cancer (CRC) screening can significantly reduce both cancer-related incidence and mortality, yet disparities in screening colonoscopy (SC) completion persist. The CHOICE program was developed at a large, urban academic health system to improve CRC screening through centralized referral and direct-access colonoscopy (DAC), eliminating the need for pre-procedure GI consults for SCs. This study evaluates real-world patient and system-level predictors of screening progression and identifies key points of attrition across each step of the care cascade, seeking to characterize those who were less likely to complete age appropriate CRC screening despite referral. Methods: We analyzed electronic health record data from 15,720 patients referred to the CHOICE program between 2023–2024 using R. Demographics, insurance, language preference, and referral characteristics were examined. Patient progression was tracked through three sequential stages: approval for DAC, scheduling, and procedure completion. Descriptive statistics were used to characterize attrition and subgroup differences. Results: Of all referred patients, 95.8% (11,788) were approved for DAC. Among those approved, 67.4% (7,945) were scheduled, and 90.8% (7,218) of scheduled patients completed colonoscopy. Women were less likely to be scheduled (54.3% vs. 45.7%) and less likely to complete once scheduled (61.8% vs. 46.5%; p < 0.001). Black patients represented a greater proportion of the not-approved group (30.1%) compared to the approved group (24.7%), and Hispanic patients had higher rates of non-completion (40.7% vs. 31.0%, p < 0.001). Spanish language preference was associated with a higher non-completion rate (14.0% vs. 8.1% among completers), and Medicaid patients had higher non-completion (26.5% vs. 12.3%). The median time from referral to procedure completion was 70 days. Conclusions: DAC pathways may improve access to CRC screening, evidenced by high overall rates of colonoscopy completion and meaningful reduction in lead time compared to national estimates. However, there are critical bottlenecks throughout this process, with the most substantial drop-off occurred after patients were approved, as a substantial portion did not ultimately schedule a colonoscopy. Socioeconomic and demographic disparities were evident in approval and completion, but were less predictive of scheduling itself, suggesting the potential influence of further latent logistical, structural, and behavioral barriers. These findings support the need for targeted interventions to reduce stepwise attrition and further advance equitable CRC screening.