Innovative data visualization tool for tracking system failures in colorectal cancer screening.

Abstract

e13598 Background: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the U.S., yet 30% of Americans are not up to date with CRC screening. Successful screening requires a multi-step process including a patient visit, clinician order, and scheduling (for colonoscopy) or return of a test kit. Where the screening cascade most frequently breaks down is unknown. We developed a comprehensive data visualization tool to quantify where failures occur and improve screening rates. Methods: We analyzed data from the electronic health record system of a regional academic health system from January 1, 2023, to June 1, 2024, to quantify steps from CRC screening order to completion. Key process steps were defined via stakeholder interviews with primary care physicians (PCP), gastroenterologists, triage nurses, and schedulers. Using data visualization software, we developed a dynamic reporting tool that tracks completion rates across these steps with a six-month reporting lag to allow order completion. Patients were included if unscreened per USPSTF guidelines during the study period. Orders with diagnostic or treatment-related indications were excluded. Results: During the study period, 151,990 unscreened patients interacted with the health system. Of these, 19,317 attended PCP visits thus considered engaged. Among eligible patients, 6,165 (31.9%) had a screening order placed, with colonoscopy comprising 5,379 (79.5%) of orders, 69.9% of which were referred to academic center providers. Cologuard accounted for 1,287 orders (19.0%) and FIT for 102 orders (1.5%). Completion rates for community and academic center colonoscopies were 77.0% and 56.1%, respectively. Colonoscopy, major failure points included scheduling (attrition of 10.0% and 26.2% for community and academic settings, respectively) and patient arrival (attrition of 13.2% and 22.9%). Adequate bowel preparation was achieved in 92.7% of patients who arrived and underwent colonoscopy. Cologuard was completed in 65% of orders, while FIT completion was 90%. Conclusions: This tool provides key insight into the significant need of process optimization for CRC screening. Notably, while less than a third of eligible patients with PCP follow up had a CRC screening test placed, the majority of unscreened patients consistently refuse screening. Colonoscopy process failures primarily occurred at scheduling and patient arrival stages, presenting key opportunities for targeted interventions. Though Cologuard is increasingly favored for its ease of use, its completion rate was comparable to colonoscopy. A limitation of this “real world” data is that some clinics may distribute FIT but only place the order upon return, potentially inflating completion percent. These findings can guide future efforts to enhance CRC screening effectiveness across settings.