Take 2: The maintenance of a colorectal cancer screening program in intervention clinics and roll out in usual care clinics during the second year of a pragmatic trial.

Background In the United States, patients in rural settings screen for colorectal cancer (CRC) at a lower rate than urban patients. SMARTER CRC, a pragmatic trial to increase CRC screening in rural clinics in Oregon, implemented a screening program that included 1) mailing fecal immunochemical tests (FIT tests) in partnership with Medicaid health plans and 2) training clinics in navigating patients to complete colonoscopy follow-up to abnormal FIT testing. We describe the second year of the SMARTER CRC project, which includes the second year of implementation for the intervention clinics, and the first year of program rollout for the usual care clinics. Methods SMARTER CRC partnered with three Medicaid health plans and 26 affiliated rural clinics. Our implementation outcomes in this mixed methods analysis were the proportion of eligible enrollees who were mailed a FIT and completed any CRC screening or FIT screening in 6 months or 12 months. We also measured if Year 1 clinics were more likely to continue CRC screening (overall and via FIT) at the same level or to increase screening in Year 2. Our qualitative evaluation contextualizes quantitative findings. Results In Year 1, Medicaid health plans identified 5,614 enrollees due for screening. After clinic staff review, FIT mailings were sent to 1,489 enrollees in intervention clinics. In Year 2, health plans identified 9,805 enrollees due for CRC screening and mailings were sent to 2,764 in the intervention clinics and 3,505 in the usual care clinics. Overall CRC screening rates and FIT completion rates among eligible patients in the intervention clinics dropped from Year 1 to Year 2 (CRC 19.9% to 12.0%, p value = < 0.00; and FIT 15.7% to 6.6%, p value = < 0.00). In usual care clinics, overall screening increased (9.3% in Year 1 to 15% in Year 2, p-value = < 0.00, and FIT completion increased from 4.9% to 9.6%, p-value = < 0.00). Conclusions CRC screening rates dropped in the second-year implementation in intervention clinics yet increased in intervention-naïve usual care clinics. Sustaining impacts of effective implementation of FIT outreach interventions over time may require ongoing effort and support.