P1299 - Adherence to Recommending a 10-Year Interval After Normal Screening Colonoscopy in Average-Risk Individuals: An Analysis of the GI Quality Improvement Consortium (GiQuic) Registry

Philip Schoenfeld, MD, MSEd, MSc (Epi), FACG¹, Jennifer Holub, MPH², Aasma Shaukat, MD^3
1John D. Dingell VA Medical Center, Redwood City, CA; ²GI Quality Improvement Consortium, Bethesda, MD; ³Minneapolis VA Health Care System, Minneapolis, MN

Introduction: Repeat colonoscopy is recommended in 10 yrs after a normal CRC screening colonoscopy in average-risk individuals. This is a priority quality indicator for CMS Medical Incentive Payment System, the multi-society Choosing Wisely campaign (2013), and multi-society statements on endoscopic quality indicators (2015). Although the target is 90% adherence, data collected prior to 2014 demonstrate adherence is usually < 75%.

Methods: Retrospective analysis of GIQuIC database from 2014-19. Patient data is de-identified and exempt per Western IRB. In order to minimize possible confounders, patients were limited to: (a) sole indication is average-risk, screening colonoscopy; (b) ages 45-89 outpatients; (c) normal colonoscopy with no biopsy or polypectomy; and, (d) no adverse events. These are all required fields in the GIQuIC Colonoscopy Data Collection form. Co-Primary endpoints: (a) Adherence to 10-yr interval or “none” (i.e., no further screening) after complete colonoscopy to cecum with photo-documentation and adequate bowel prep; and, (b) adherence to < 1 yr interval for incomplete colonoscopy due to inadequate prep. Secondary endpoints: Adherence in GI vs non-GI endoscopists compared with chi-square analysis. Frequency of adherence quantified after stratifying for patient-level factors, including age, gender, race, ASA class, and geographic region. Trends in adherence over time were assessed for sites contributing data from 2014-19 with Cochran-Armitage test.

Results: For complete colonoscopy (n = 1,266,819), adherence to a 10-yr interval or stop screening was 85.9% and was higher in non-GI endoscopists vs GI (87.7% vs 85.3%, p < 0.001). (Table 1) Repeat colonoscopy in 5 yrs was the second most common recommendation (13.3%) (Figure 1a). For complete colonoscopy, adherence significantly improved over time (p < 0.001), although it has only improved by 1% from 2016-2019 (Figure 1b). For incomplete colonoscopy due to inadequate prep (n = 68,960), adherence to repeat interval of <1 yr was 40.2%, and 5 yr interval (26.2%) was the second most common recommendation. (Figure 1c) Adherence was again higher for non-GI endoscopists vs GI (47.4% vs 37.3%, p < 0.001), and adherence to < 1 yr interval with inadequate prep improved significantly at stable sites from 25.3% in 2014 to 51.2% in 2019, p < 0.001. (Figure 1d)

Discussion: Adherence to guideline recommendations are sub-optimal. Future research should assess interventions to improve adherence.


Figure: Figure 1a-1d: Frequency of Guideline Adherent Recommendations