Sa1403 Bowel Preparation and Hospital Are Important Factors Influencing Quality of Colonoscopy As Measured by Cecum Intubation and Adenoma Detection

Sa1403 Bowel Preparation and Hospital Are Important Factors Influencing Quality of Colonoscopy As Measured by Cecum Intubation and Adenoma Detection Elisabeth J. Grobbee*, Tim D. Belderbos, Maarten Meijssen, Rob J. Ouwendijk, Thjon J. Tang, Frank Ter Borg, Peter J. Van Der Schaar, Doris M. Le Fevre, Merijn Stouten, Onno Van Der GalieN, Theo J. Hiemstra, Wouter H. De Vos Tot Nederveen Cappel, Pieter Ter Borg, Leon M. Moons, Ernst J. Kuipers, Peter D. Siersema Gastroenterology and Hepatology, Erasmus MC, Rotterdam, Netherlands; Gastroenterology and Hepatology, University Medical Center Utrecht, Utrecht, Netherlands; Gastroenterology and Hepatology, Isala Hospital, Zwolle, Netherlands; Gastroenterology and Hepatology, Ikazia Hospital, Rotterdam, Netherlands; Gastroenterology and Hepatology, IJsseland Hospital, Capelle aan den IJssel, Netherlands; Gastroenterology and Hepatology, Deventer Hospital, Deventer, Netherlands; Gastroenterology and Hepatology, Antonius Hospital, Nieuwegein, Netherlands; Achmea Health Care, Leiden, Netherlands; Gupta Strategists, Ophemert, Netherlands Background: As more colorectal cancer (CRC) screening programs are implemented worldwide, adequate measurement of the quality of colonoscopy becomes increasingly important. Cecal intubation rate (CIR) and adenoma detection rate (ADR) are used as quality indicators and have been related to the occurrence of post-colonoscopy CRC. The aim of this study was to compare the quality of routine colonoscopy between seven hospitals in the Netherlands to determine to what extent possible differences in CIR and ADR are attributable to casemix versus proceduraland hospital-related factors. Methods: We prospectively registered all colonoscopies performed between November 2012 and January 2013 in two academic and five general hospitals in the Netherlands. Colonoscopies in patients with inflammatory bowel disease (IBD) or hereditary CRC syndromes were excluded. Multivariate analyses were adjusted for age, gender, ASA score and indication. Results: A total of 3,145 patients were included (55% female; mean age 59 16 years). The majority of procedures (87%) were performed in adequately prepared colons, defined as a Boston Bowel Preparation Scale (BBPS)R6. Mean CIR was 94%, ranging from 89% to 99%, and was significantly different between hospitals (p! 0.001). We found no significant differences in CIR between gastroenterologists (94%) fellows (94%) and nurse endoscopists (96%) (pZ0.235). The overall ADR was 32% and was significantly higher in men vs. women (39% vs. 27%; p!0.001). ADR varied between hospitals, ranging from 23% to 43% (p!0.001) and between type of endoscopists, with nurse endoscopists (37%) performing slightly better than gastroenterologists (31%) and fellows (34%; p!0.05). In multivariate analysis, independent predictors for CIR were use of sedation (OR 2.49, 95%CI 1.57 3.94), BBPS R6 (OR 13.0, 95%CI 8.69 19.4) and hospital (p!0.001), with OR’s ranging from 0.81 (95%CI 0.44-1.51) to 4.91 (95%CI 2.01-12.01) (largest hospital as reference). In multivariate analysis, independent predictors for ADR were with BBPSR6 (OR 0.55, 95%CI 0.40 0.74), incomplete colonoscopy (OR 0.615, 95%CI 0.40 0.94) and hospital (p!0.001), with OR’s ranging from 0.59 (95%CI 0.44-0.79) to 1.62 (95%CI 1.00-2.64) (largest hospital as reference). Conclusion: CIR and ADR as quality indicators for colonoscopy are able to detect differences between hospitals. As both CIR and ADR are not dependent on casemix, they objectively provide incentives for quality improvement.