Impact of a quarterly report card on colonoscopy quality measures

Background Colonoscopy quality is operator-dependent. Studies assessing the effect of interventions to decrease variation in colonoscopy quality have shown inconsistent results. Since 2009, endoscopists at our university-affiliated, Veterans Affairs medical center have received a quarterly “report card” summarizing individual colonoscopy quality indicators as part of an ongoing quality assurance program. Objective To determine the effect of the quality report card intervention on colonoscopy performance. Design Retrospective study. Setting Tertiary-care, academic, university-affiliated, Veterans Affairs medical center in Indianapolis, Indiana. Patients Data from 6 endoscopists practicing at the Roudebush Veterans Affairs Medical Center were included. Patients were average-risk, aged 50 years or older, undergoing their first screening colonoscopy. Intervention Quarterly report card. The study time frame was July 1, 2008 to December 31, 2008 (before-intervention) and April 1, 2009 to March 31, 2011 (intervention). Main Outcome Measurements The primary outcomes were cecal intubation and adenoma detection rates (ADR), adjusted for physician, patient age, and sex. Multivariable logistic regression was performed to determine factors associated with adenoma detection. Results A total of 928 patients (male 93%, white 78%) were included (before-intervention 336; intervention 592). There were no significant differences in patient age, sex, smoking status, body mass index, bowel preparation quality, colonoscope model, and proportion of colonoscopies performed with a trainee between the before-intervention and intervention phases. In the intervention phase, the adjusted adenoma detection and cecal intubation rates were significantly higher: 53.9% (95% confidence interval [CI], 49.7%-58.1%) vs 44.7% (95% CI, 39.1%-50.4%); P = .013 and 98.1% (95% CI, 96.7%-99.0%) vs 95.6% (95% CI, 92.5%-97.5%); P = .027, respectively. A higher ADR trend in the intervention phase was found for 5 of the 6 physicians. The increment in ADR was due mostly to increased detection of proximal adenomas. There were no significant changes in serrated polyp detection, advanced neoplasm detection, number of adenomas detected per colonoscopy, and mean size of adenomas after implementation of the intervention. The report card intervention remained significantly associated with higher ADRs after adjustment for patient age, sex, and physician (odds ratio 1.45; 95% CI, 1.08-1.94). Limitations Single cent