Maximizing Colonoscopy Completion Rates and Quality: Effect of Direct Primary Care Provider versus Gastroenterology Referral in Colorectal Cancer Screening and Surveillance

Abstract

Introduction: Colorectal cancer (CRC) is the 2nd leading cause of cancer death in the US and the 4th worldwide. In the US, incidence has declined as a result of screening strategies, since early diagnosis is the most important prognostic factor. CRC screening with colonoscopy provides visualization of precancerous polyps and the opportunity for immediate removal. The effect in CRC mortality is stronger when colonoscopies are of good quality. A good bowel preparation increases diagnostic accuracy and therapeutic safety. Methods: A retrospective review of CRC screening rates in our inner-city community health clinic identified our screening rate was 37%, significantly below the citywide average of 70%. To improve our screening rate, direct colonoscopy referral by Primary Care Provider (PCP) was added to our EMR's order set. We aimed to determine if referral origin, PCP vs Gastroenterologist (GI), impacts completion rates and quality of bowel preparation. We reviewed all referral orders for screening or surveillance colonoscopy at our facility from June 2015 to September 2016. Patients >50 years old were included and those on which colonoscopies were performed for diagnostic purposes were excluded. Results: We identified 313 patients referred for CRC screening or surveillance colonoscopy. Of these, 50.5% were referred by GI and 49.5% by PCP. Baseline characteristics were compared among groups (Table 1). Older patients and those with diabetes or liver disease were more likely to be referred by GI (Table 2). 60% of our patients (N=188) completed their colonoscopy. Patients referred by GI were more likely to complete their colonoscopy compared to those with direct PCP referral, 72.8% vs 47.1% respectively (p < 0.0001). Also, the GI referral group was more likely to have better bowel preparation compared to the direct PCP referral group (p=0.0017).Table: Table. Patient baseline characteristics and characteristics per referral typeTable: Table. Odds of GI referral for colonoscopy evaluation per patients' factorsConclusion: This study is part of the quality improvement endeavors undertaken at our clinic to improve our 37% CRC screening rate. Despite providing direct PCP referrals for colonoscopy, patients referred by GI were more likely to complete their colonoscopy and to have better bowel preparation. We theorize that multiple factors contributed to our findings, including PCP's knowledge of procedure's details and confidence answering questions. Hence, the approach of our PCPs towards CRC screening and surveillance should be explored. Further studies are needed to evaluate interventions as ours and their impact in improving CRC screening rates.