Development of a Large Colonoscopy-Based Longitudinal Cohort for Integrated Research of Colorectal Cancer: Partners Colonoscopy Cohort.

BACKGROUND AND AIMS: Conventional adenomas (CAs) and serrated polyps (SPs) are precursors to colorectal cancer (CRC). Understanding metachronous cancer risk is poor due to lack of accurate large-volume datasets. We outline the use of natural language processing (NLP) in forming the Partners Colonoscopy Cohort, an integrated longitudinal cohort of patients undergoing colonoscopies. METHODS: We identified endoscopy quality data from endoscopy reports for colonoscopies performed from 2007 to 2018 in a large integrated healthcare system, Mass General Brigham). Through modification of an established NLP pipeline, we extracted histopathological data (polyp location, histology and dysplasia) from corresponding pathology reports. Pathology and endoscopy data were merged by polyp location using a four-stage algorithm. NLP and merging procedures were validated by manual review of 500 pathology reports. RESULTS: 305,656 colonoscopies in 213,924 patients were identified. After merging, 76,137 patients had matched polyp data for 334,750 polyps. CAs and SPs were present in 86,707 (28.5%) and 55,373 (18.2%) colonoscopies. Among patients with polyps at index screening colonoscopy, 14,931 (33.4%) had follow-up colonoscopy (median 46.4, interquartile range 33.8-62.4 months); 91 (0.2%) and 1127 (2.5%) patients developed metachronous CRC and high-risk polyps (polyps ≥ 10 mm or CAs having high-grade dysplasia/villous/tublovillous histology or SPs with dysplasia). Genetic data were available for 23,787 (31.7%) patients with polyps from the Partners Biobank. The validation study showed a positive predictive value of 100% for polyp histology and locations. CONCLUSION: We created the Partners Colonoscopy Cohort providing essential infrastructure for future studies to better understand the natural history of CRC and improve screening and post-polypectomy strategies.

Long-term Risk of Colorectal Cancer After Removal of Conventional Adenomas and Serrated Polyps.

BACKGROUND & AIMS: Endoscopic screening reduces incidence and mortality of colorectal cancer (CRC) because precursor lesions, such as conventional adenomas or serrated polyps, are removed. Individuals with polypectomies are advised to undergo colonoscopy surveillance to prevent CRC. However, guidelines for surveillance intervals after diagnosis of a precursor lesion, particularly for individuals with serrated polyps, vary widely, and lack sufficient supporting evidence. Consequently, some high-risk patients do not receive enough surveillance and lower-risk subjects receive excessive surveillance. METHODS: We examined the association between findings from first endoscopy and CRC risk among 122,899 participants who underwent flexible sigmoidoscopy or colonoscopy in the Nurses' Health Study 1 (1990-2012), Nurses' Health Study 2 (1989-2013), or the Health Professionals Follow-up Study (1990-2012). Endoscopic findings were categorized as no polyp, conventional adenoma, or serrated polyp (hyperplastic polyp, traditional serrated adenoma, or sessile serrated adenoma, with or without cytological dysplasia). Conventional adenomas were classified as advanced (≥10 mm, high-grade dysplasia, or tubulovillous or villous histology) or nonadvanced, and serrated polyps were assigned to categories of large (≥10 mm) or small (<10 mm). We used a Cox proportional hazards regression model to calculate the hazard ratios (HRs) of CRC incidence, after adjusting for various potential risk factors. RESULTS: After a median follow-up period of 10 years, we documented 491 incident cases of CRC: 51 occurred in 6161 participants with conventional adenomas, 24 in 5918 participants with serrated polyps, and 427 in 112,107 participants with no polyp. Compared with participants with no polyp detected during initial endoscopy, the multivariable HR for incident CRC in individuals with an advanced adenoma was 4.07 (95% confidence interval [CI] 2.89-5.72) and the HR for CRC in individuals with a large serrated polyp was 3.35 (95% CI 1.37-8.15). In contrast, there was no significant increase in risk of CRC in patients with nonadvanced adenomas (HR 1.21; 95% CI 0.68-2.16, P = .52) or small serrated polyps (HR 1.25; 95% CI 0.76-2.08; P = .38). CONCLUSIONS: These findings provide support for guidelines that recommend repeat lower endoscopy within 3 years of a diagnosis of advanced adenoma and large serrated polyps. In contrast, patients with nonadvanced adenoma or small serrated polyps may not require more intensive surveillance than patients without polyps.

Automated Time-Released Reminders Improve Patient Access to Colonoscopy.

Unscheduled colonoscopy orders lead to missed opportunities for early diagnosis and screening. The aim of this study was to evaluate the effect of an automated time-released reminder program on conversion of colonoscopy orders to scheduled cases. In this prospective study, we compared patients ordered for a colonoscopy who were enrolled in an automated reminder program (intervention) with a historical cohort of patients ordered for a colonoscopy who did not receive scheduling reminders (control). The intervention group received automated text message and email reminders using a software platform at 1, 7, and 14 days after a colonoscopy order was placed. The percentage of colonoscopies scheduled within 14 days of order placement improved from 66.0% in the control group to 73.4% in the intervention group (p = .001). The percentage of colonoscopies scheduled within 30 days improved from 73.6% to 90.0% (p < .0001). For colonoscopies ordered by a nongastroenterologist, the percentage of cases scheduled within 30 days of order placement improved from 65.8% in the control group to 90.0% in the intervention group (p < .0001). There was a 10% decrease in phone calls with endoscopy staff for the intervention group relative to the control group. Automated reminders for colonoscopy scheduling improve efficiency in colonoscopy scheduling.

Interval Colorectal Cancer in Inflammatory Bowel Disease: The Role of Guideline Adherence.

BACKGROUND: Factors associated with interval colorectal cancer (CRC) development in the inflammatory bowel disease (IBD) population remain unclear. AIMS: Among a cohort of patients with interval CRC, we aimed to evaluate IBD characteristics, colonoscopy quality indicators, and surveillance guideline adherence. METHODS: We performed a retrospective review of IBD- and non-IBD-associated interval CRCs diagnosed between January 2007 and December 2014 within a large US healthcare system. We evaluated risk factors for CRC among patients with IBD. We assessed adherence to surveillance guidelines according to the American Society for Gastrointestinal Endoscopy (IBD surveillance) and the US Multi-Society Task Force on Colorectal Cancer (polyp surveillance). We compared colonoscopy quality measures between patients with and without IBD. RESULTS: Among 5345 cases of colonic adenocarcinoma, we detected 15 IBD-associated cases of interval CRC and 230 non-IBD-associated cases of interval CRC. Compared to patients without IBD, IBD patients were younger (54.5 vs. 70.4 years; p < 0.0001) and experienced a shorter interval between index colonoscopy and CRC diagnosis (20.7 vs. 35.1 months; p = 0.0009). Fifty three percent (8/15) of interval CRCs in IBD patients were detected within surveillance guidelines. All IBD patients with interval CRC detected after guideline surveillance interval had high-risk features, including active inflammation, previous low-grade or indefinite dysplasia, multiple pseudopolyps on index colonoscopy, or a first-degree relative with CRC. There were no differences in colonoscopy quality measures between patients with and without IBD. CONCLUSIONS: This study stresses the importance of strict short-interval surveillance for IBD patients with high-risk features, including active inflammation on index colonoscopy.

A Multicenter Study Into Causes of Severe Acute Liver Injury.

The differential diagnosis of an increase in alanine aminotransferase (ALT) level and/or aspartate aminotransferase (AST) level of ≥1000 IU/L often is stated to include 3 main etiologies: ischemic hepatitis, acute viral hepatitis (typically hepatitis A and hepatitis B), and drug-induced (more specifically, acetaminophen/paracetamol) liver injury (DILI).1 Unfortunately, there are a paucity of studies examining the most common causes of acute liver injury (ALI) and those that have been published have been small,2 single-center,2 or examined less severe increases in ALT or AST levels.3,4 We conducted a multicenter study of all patients with an ALT and/or AST level ≥1000 IU/L. Our study had 3 main goals: (1) to determine the most common causes of an ALT and/or AST level ≥1000 IU/L, along with their relative frequencies; (2) to determine differences in etiology based on hospital type (liver transplant center, community hospital, Veterans Affairs hospital); and (3) to confirm or disprove the differential heuristic that ischemic hepatitis, acute viral hepatitis, and acetaminophen toxicity are the most common etiologies.

Clinicopathologic and genetic characteristics of interval colorectal carcinomas favor origin from missed or incompletely excised precursors.

Interval colorectal cancers may arise from missed or incompletely excised precursors or from a unique rapid progression pathway. We compared the clinicopathologic and molecular profiles of interval and matched non-interval colorectal cancer to determine whether interval colorectal cancers harbor any unique genetic characteristics. Fifty one of 982 colorectal cancer (5.2%) were categorized as interval colorectal cancer, defined as colorectal cancer detected in a diagnostic examination prior to the next recommended colonoscopy and at least 1 year after the last colonoscopy. Clinicopathologic characteristics of interval colorectal cancer were compared to non-interval colorectal cancer matched 1:1 on age, gender, and tumor location. Molecular profile of a subset of interval colorectal cancer (n = 20) and matched (1:2) non-interval colorectal cancer (n = 40) were evaluated using next generation sequencing. Interval colorectal cancer were more likely to occur in the right colon (55% vs. 35%; p = 0.02) and in patients > 70 years of age (55% vs. 34%; p = 0.002). Clinicopathologic features and aberrant DNA mismatch repair protein expression were not significantly different between interval and matched non-interval colorectal cancer. The frequency and spectrum of genetic alterations was also similar in interval and matched non-interval colorectal cancer. Similar findings were seen when analysis was restricted to interval colorectal cancer diagnosed <5 years after last colonoscopy (n = 42). Interval and non-interval colorectal cancers share similar clinicopathologic and genetic profiles when matched for tumor location. Interval colorectal cancers and are more likely to develop from missed or incompletely excised precursors rather than a unique rapid progression pathway.

Impact of Automated Time-released Reminders on Patient Preparedness for Colonoscopy.

GOALS: To evaluate the effect of an automated time-released colonoscopy reminder program on preparation quality and the rates of canceled procedures. BACKGROUND: Patients missing colonoscopy appointments or arriving with inadequate bowel preparations result in delays in care and increase in health care costs. STUDY: We initiated a quality improvement program which included switching to a split-dose preparation and delivering preparation instructions and appointment reminders via text messages and emails sent with an automated time-released HIPAA-compliant software platform. All patients scheduled for a colonoscopy from March through June 2017 were enrolled unless mobile phone, email, or preparation type data was missing (nonintervention cohort). Primary outcome metrics were the rate of adequate bowel preparation and the rate of canceled procedures. Outcomes of the intervention group were compared with outcomes from colonoscopies performed from March through June 2016 (baseline cohort). RESULTS: There were 1497 patients (40.7% male, mean age 56.4 y) enrolled in the automated reminder program. Compared with the baseline cohort, the rate of adequate bowel preparation increased from 88.5% to 96.2% (P<0.0001). The rate of canceled procedures decreased from 6.1% to 4.3% (P=0.02). On multivariable analysis, the automated reminder program improved adequate preparation quality 2.85-fold (95% confidence interval, 2.03-3.99; P<0.0001). Sensitivity analysis comparing the intervention to nonintervention cohorts showed that improvement in preparation quality was attributable to the automated reminder program. CONCLUSIONS: Implementation of automated time-released colonoscopy preparation reminders via text messages and emails improved patient preparedness for colonoscopy, with significantly improved bowel preparation quality and fewer canceled procedures.

Development and validation of the SIMPLE endoscopic classification of diminutive and small colorectal polyps.

BACKGROUND: Prediction of histology of small polyps facilitates colonoscopic treatment. The aims of this study were: 1) to develop a simplified polyp classification, 2) to evaluate its performance in predicting polyp histology, and 3) to evaluate the reproducibility of the classification by trainees using multiplatform endoscopic systems. METHODS: In phase 1, a new simplified endoscopic classification for polyps - Simplified Identification Method for Polyp Labeling during Endoscopy (SIMPLE) - was created, using the new I-SCAN OE system (Pentax, Tokyo, Japan), by eight international experts. In phase 2, the accuracy, level of confidence, and interobserver agreement to predict polyp histology before and after training, and univariable/multivariable analysis of the endoscopic features, were performed. In phase 3, the reproducibility of SIMPLE by trainees using different endoscopy platforms was evaluated. RESULTS: Using the SIMPLE classification, the accuracy of experts in predicting polyps was 83 % (95 % confidence interval [CI] 77 % - 88 %) before and 94 % (95 %CI 89 % - 97 %) after training (P â€Š= 0.002). The sensitivity, specificity, positive predictive value, and negative predictive value after training were 97 %, 88 %, 95 %, and 91 %. The interobserver agreement of polyp diagnosis improved from 0.46 (95 %CI 0.30 - 0.64) before to 0.66 (95 %CI 0.48 - 0.82) after training. The trainees demonstrated that the SIMPLE classification is applicable across endoscopy platforms, with similar post-training accuracies for narrow-band imaging NBI classification (0.69; 95 %CI 0.64 - 0.73) and SIMPLE (0.71; 95 %CI 0.67 - 0.75). CONCLUSIONS: Using the I-SCAN OE system, the new SIMPLE classification demonstrated a high degree of accuracy for adenoma diagnosis, meeting the ASGE PIVI recommendations. We demonstrated that SIMPLE may be used with either I-SCAN OE or NBI.

Natural Language Processing Accurately Calculates Adenoma and Sessile Serrated Polyp Detection Rates.

BACKGROUND: ADR is a widely used colonoscopy quality indicator. Calculation of ADR is labor-intensive and cumbersome using current electronic medical databases. Natural language processing (NLP) is a method used to extract meaning from unstructured or free text data. AIMS: (1) To develop and validate an accurate automated process for calculation of adenoma detection rate (ADR) and serrated polyp detection rate (SDR) on data stored in widely used electronic health record systems, specifically Epic electronic health record system, Provation® endoscopy reporting system, and Sunquest PowerPath pathology reporting system. METHODS: Screening colonoscopies performed between June 2010 and August 2015 were identified using the Provation® reporting tool. An NLP pipeline was developed to identify adenomas and sessile serrated polyps (SSPs) on pathology reports corresponding to these colonoscopy reports. The pipeline was validated using a manual search. Precision, recall, and effectiveness of the natural language processing pipeline were calculated. ADR and SDR were then calculated. RESULTS: We identified 8032 screening colonoscopies that were linked to 3821 pathology reports (47.6%). The NLP pipeline had an accuracy of 100% for adenomas and 100% for SSPs. Mean total ADR was 29.3% (range 14.7-53.3%); mean male ADR was 35.7% (range 19.7-62.9%); and mean female ADR was 24.9% (range 9.1-51.0%). Mean total SDR was 4.0% (0-9.6%). CONCLUSIONS: We developed and validated an NLP pipeline that accurately and automatically calculates ADRs and SDRs using data stored in Epic, Provation® and Sunquest PowerPath. This NLP pipeline can be used to evaluate colonoscopy quality parameters at both individual and practice levels.

Timing of upper endoscopy influences outcomes in patients with acute nonvariceal upper GI bleeding.

BACKGROUND AND AIMS: Current guidelines advise that upper endoscopy be performed within 24 hours of presentation in patients with acute nonvariceal upper GI bleeding (UGIB). However, the role of urgent endoscopy (<12 hours) is controversial. Our aim was to assess whether patients admitted with acute nonvariceal UGIB with lower-risk versus high-risk bleeding have different outcomes with urgent compared with nonurgent endoscopy. METHODS: A retrospective cohort study was conducted of patients admitted to an academic hospital with nonvariceal UGIB. The primary outcome was a composite of inpatient death from any cause, inpatient rebleeding, need for surgical or interventional radiologic intervention, or endoscopic reintervention. The Glasgow-Blatchford score (GBS) was calculated; lower risk was defined as a GBS < 12, and high risk was defined as a GBS ≥ 12. RESULTS: Of 361 patients, 37 patients (10%) experienced the primary outcome. Patients who underwent urgent endoscopy had a greater than 5-fold increased risk of reaching the composite outcome (unadjusted odds ratio [OR], 5.6; 95% confidence interval [CI], 2.8-11.4; P < .001). Lower-risk patients who were taken urgently to endoscopy were more likely to reach the composite outcome (adjusted OR, 0.71 per 6 hours; 95% CI, 0.55-0.91; P = .008). However, in the high-risk patients, time to endoscopy was not a significant predictor of the primary outcome (adjusted OR, 0.93 per 6 hours; 95% CI, 0.77-1.13; P = .47; adjusted P for interaction = .039). CONCLUSION: Urgent endoscopy is a predictor of worse outcomes in select patients with acute nonvariceal UGIB.

Association between an increase in blood urea nitrogen at 24 hours and worse outcomes in acute nonvariceal upper GI bleeding.

BACKGROUND AND AIMS: An increase in blood urea nitrogen (BUN) at 24 hours is a solitary and significant predictor of mortality in patients with acute pancreatitis, which may predict worse outcomes in the similarly resuscitation-requiring condition of acute nonvariceal upper GI bleeding (UGIB). The aim of our study was to assess whether an increase in BUN at 24 hours is predictive of worse clinical outcomes in acute nonvariceal UGIB. METHODS: A retrospective cohort study including patients admitted to an academic hospital from 2004 to 2014 was conducted. An increase in BUN was defined as an increase in BUN at 24 hours of hospitalization compared with BUN at presentation. The primary outcome was a composite of inpatient death, inpatient rebleeding, need for surgical or radiologic intervention, or endoscopic reintervention. Associations between BUN change and outcomes were assessed via the Pearson χ2 test and the Fisher exact test and via logistic regression for adjusted analyses. RESULTS: There were 357 patients included in the analysis with a mean age of 64 years; 54% were men. The mean change in BUN was -10.1 mg/dL (standard deviation, 12.7 mg/dL). Patients with an increased BUN (n = 37 [10%]) were significantly more likely to experience the composite outcome (22% vs 9%, P = .014), including an increased risk of inpatient death (8% vs 1%, P = .004), compared with patients with a decreased or unchanged BUN (n = 320 [90%]). In a logistic regression model adjusting for the AIMS65 score, an increase in BUN was independently associated with an increased risk for the composite outcome (odds ratio, 2.75; P = .026). CONCLUSION: Increasing BUN at 24 hours likely reflects under resuscitation and is a predictor of worse outcomes in patients with acute nonvariceal UGIB.

Impact of physician compliance with colonoscopy surveillance guidelines on interval colorectal cancer.

BACKGROUND AND AIMS: Interval colorectal cancer (iCRC) incidence is the criterion standard benchmark for measuring the effectiveness of colonoscopy. Colonoscopy surveillance guidelines are designed to minimize iCRC cases. Our aims were to describe characteristics of iCRC patients and to assess whether development of iCRC is related to colonoscopy surveillance guideline intervals. METHODS: We performed a retrospective cohort study of postcolonoscopy iCRC cases in a large healthcare system. Guideline-based colonoscopy intervals were calculated based on the 2012 U.S. Multi-Society Task Force for Colorectal Cancer colonoscopy surveillance guidelines. Backward stepwise linear regression was used to determine predictors of iCRC before guideline-recommended follow-up intervals. RESULTS: We identified 245 iCRC cases (mean age, 69.4 years; 56.3% male) out of 5345 colon cancers evaluated for a prevalence of 4.60%. On index colonoscopy, 75.1% had an adequate preparation, 93.0% reached the cecum, and 52.5% had polyps. iCRC developed before the guideline-recommended interval in 59.1% of patients (94/159). Independent predictive factors of this finding were inadequate preparation (OR, .012; 95% CI, .003-.06; P < .0001) and ≥3 polyps on index colonoscopy (OR, .2; 95% CI, .078-.52; P = .0009). An endoscopist-recommended follow-up interval past the guideline-recommended interval was seen in 23.9% of cases (38/159). Most (34/38, 89.5%) of these iCRCs had inadequate preparation and were diagnosed after the guideline-based follow-up interval. CONCLUSIONS: Current colonoscopy surveillance guidelines may be inadequate to prevent many iCRC cases. Physician noncompliance with guideline-based surveillance intervals may increase in iCRC cases, especially in patients with an initially inadequate bowel preparation.

Appointment-keeping behaviors and procedure day are associated with colonoscopy attendance in a patient navigator population.

BACKGROUND: Patient navigator programs (PNP) have been shown to improve colonoscopy completion with demonstrated cost-effectiveness. Despite additional resources available to these patients, many still do not attend their colonoscopies. The aim of this study was to determine factors associated with colonoscopy attendance amongst patients in whom logistical barriers to attendance have been minimized through enrollment in a PNP. METHODS: Retrospective case-control study of patients enrolled in a PNP for colonoscopy performed at a tertiary endoscopy center from 2009 to 2014. Cases were defined as patients who did not attend their first scheduled colonoscopy after PNP enrollment. Age- and gender-matched controls completed their first scheduled colonoscopy after PNP enrollment. RESULTS: 514 subjects (257 cases, mean age 57.1years, 36.6% males) were included. Patients who attended their colonoscopy were less likely to be Spanish-speaking (64.6% vs 78.2%, p=0.0003) and uninsured (0.4% vs 3.9%, p=0.006). Attendance rates were significantly lower for screening colonoscopies compared to an indication of surveillance or diagnostic (45.5% vs 65.3%, p<0.0001). Fewer patients attended colonoscopies scheduled on Monday (39.2% vs 52.1%, p=0.04) and in December (10.7% vs 52.3%, p<0.0001). On multivariate analysis, poor appointment-keeping behaviors, including a prior missed colonoscopy (OR 0.20, 95% CI 0.10-0.39) or missed office visit (OR 0.44, 95% CI 0.26-0.73) and procedures scheduled on Mondays (OR 0.51, 95% CI 0.27-0.94) were negatively associated with attendance. CONCLUSIONS: Appointment-keeping behaviors, in addition to insurance-status, language-barriers and medical comorbidities, influence colonoscopy attendance in a PNP population. Patients scheduled for colonoscopies on Mondays or in December may require more resources to ensure attendance.

Endoscopic Localization of Colon Cancer Is Frequently Inaccurate.

BACKGROUND: Colonoscopic location of a tumor can influence both the surgical procedure choice and overall treatment strategy. AIMS: To determine the accuracy of colonoscopy in determining the location of colon cancer compared to surgical localization and to elucidate factors that predict discordant colon cancer localization. METHODS: We conducted a retrospective cross-sectional study of colon cancers diagnosed on colonoscopy at two academic tertiary-care hospitals and two affiliated community hospitals from 2012 to 2014. Colon cancer location was obtained from the endoscopic and surgical pathology reports and characterized by colon segment. We collected data on patient demographics, tumor characteristics, endoscopic procedure characteristics, surgery planned, and surgery performed. Univariate analyses using Chi-squared test and multivariate analysis using forward stepwise logistic regression were performed to determine factors that predict discordant colon cancer localization. RESULTS: There were 110 colon cancer cases identified during the study period. Inaccurate endoscopic colon cancer localization was found in 29% (32/110) of cases. These included 14 cases (12.7%) that were discordant by more than one colonic segment and three cases where the presurgical planned procedure was significantly changed at the time of surgery. On univariate analyses, right-sided colon lesions were associated with increased inaccuracy (43.8 vs 24.4%, p = 0.04). On multivariate analysis, right-sided colon lesions remained independently associated with inaccuracy (OR 1.74, 95% CI 1.03-2.93, p = 0.04). CONCLUSIONS: Colon cancer location as determined by colonoscopy is often inaccurate, which can result in intraoperative changes to surgical management, particularly in the right colon.

Comparison of Colonoscopy Quality Measures Across Various Practice Settings and the Impact of Performance Scorecards.

BACKGROUND: Quality performance measures for screening colonoscopy vary among endoscopists. The impact of practice setting is unknown. AIMS: We aimed to (1) compare screening colonoscopy performance measures among three different US practice settings; (2) evaluate factors associated with adenoma detection; and (3) assess a scorecard intervention on performance metrics. METHODS: This multi-center prospective study compared patient, endoscopist, and colonoscopy characteristics performed at a tertiary care hospital (TCH), community-based hospital (CBH), and private practice group (PPG). Withdrawal times (WT), cecal intubation, and adenoma detection rates (ADR) were compared by site at baseline and 12 weeks following scorecard distribution. Generalized linear mixed models identified factors associated with adenoma detection. RESULTS: Twenty-eight endoscopists performed colonoscopies on 1987 asymptomatic, average-risk individuals ≥50 years. Endoscopist and patient characteristics were similar across sites. The PPG screened more men (TCH: 42.8%, CBH: 45.0%, PPG: 54.2%; p < 0.0001). Preparation quality varied with good/excellent results in 70.6, 88.3, and 92% of TCH, CBH, and PPG cases, respectively (p < 0.0001). Male ADRs, cecal intubation, and WT exceeded recommended benchmarks despite variable results at each site; female ADRs were <15% at the PPG which screened the fewest females. Performance remained unchanged following scorecard distribution. Adenoma detection was associated with increasing patient age, male gender, WT, adequate preparation, but not practice setting. CONCLUSIONS: Each practice performed high-quality screening colonoscopy. Scorecards did not improve performance metrics. Preparation quality varies among practice settings and can be modified to improve adenoma detection.