Improving the Quality of Bowel Preparation for Colonoscopy: Populations to Target in an Urban Tertiary Care Setting

Abstract

Inadequate bowel preparation is encountered in approximately 17%–27% of outpatient colonoscopies performed in the United States. Poor or incomplete visualization of the colon reduces the yield of screening colonoscopies and increases health care costs due to longer procedure times and aborted procedures. We performed a study to identify factors impacting the quality of bowel preparation, thereby enabling the identification of a population amenable to quality improvement initiatives. We performed a cross-sectional questionnaire survey in patients undergoing outpatient colonoscopy at 2 tertiary care hospitals within the Johns Hopkins Health System. Data on demographic factors, education, health literacy, income, and functional status were recorded. Colonoscopy records were reviewed retrospectively to identify the Boston Bowel Preparation Scale (BBPS) score. A BBPS score of <5 was considered inadequate. Fisher’s exact, χ2 tests, and univariate and multivariate binary logistic regression were performed to assess the strength of the association for selected factors with a BBPS <5. A total of 467 patients completed the survey between August 2017 and October 2017. Most colonoscopies were performed for colorectal cancer (CRC) screening or polyp surveillance (52%). Fifty-eight (12.4%) patients had a family or personal history of CRC. The median BBPS score was 7, and 81.4% of patients had adequate bowel preparation (BBPS ≥5). Using Fisher’s exact or χ2 tests, there was significant association between bowel preparation adequacy and level of education (P = .019), confidence in filling out forms (P = .001), requirement of help with reading hospital materials (P = .011), difficulty learning about medical condition (P = .019), ability to walk 3 city blocks (P = .002), need for help with activities of daily living (P = .001), income (P = .004), and number of daily medications (P = .001). There was no significant association with age, sex, previous successful bowel preparation, bowel preparation agent, patient perception of completion of bowel preparation, and mode of delivery of bowel preparation instructions (letter, phone, or video). When adjusted for functional status, health literacy, education level, income, and number of prescriptions, poor form confidence (odds ratio, 24.1; P < .001) and being on multiple prescription medications (odds ratio, 12.49; P = .02) had the highest odds of predicting inadequate bowel preparation (Table 1). Inadequate bowel preparation is a significant problem, accounting for reduced yield of screening and increased health care costs. Poor health literacy, functional status, and high number of daily medications are significant factors predicting inadequate bowel preparation in the tertiary care setting. Future measures to improve the quality of bowel preparation should target these populations for maximal impact.Table 1Multivariate Logistic Regression for Factors Predicting Inadequate Bowel Preparation (Boston Bowel Preparation Scale Score <5)FactorOR95% CIP valueAOR95% CIP valueForm confidence: How confident are you filling out forms by yourself? Extremely1.00——1.00—— Quite a bit2.170.99–4.78.051.490.53–4.22.45 Somewhat2.370.64–8.80.201.030.11–9.77.98 A little bit6.331.74–23.05.012.200.26–18.82.47 Not at all5.541.56–19.63.0124.142.98–195.70.00Functional status: Because of physical, mental, or emotional problems, do you need the help of any other persons in handling routine needs, such as everyday household chores, doing necessary business, shopping, or getting around for other purposes? No1.00——1.00—— Yes3.371.69–6.74.002.821.08–7.33.03Level of education Elementary/middle school3.141.06–9.26.041.130.17–7.39.90 GED2.161.07–4.37.032.260.83–6.20.11 College or higher1.00——1.00——Income <$20,0003.031.28–7.17.010.730.21–2.47.61 $20,000–$30,0002.640.94–7.44.071.330.39–4.51.65 $30,000–$50,0001.080.36–3.29.890.400.10–1.63.20 $50,000–$100,0000.500.16–1.64.250.190.05–0.81.03 ≥$100,0001.00——1.00——No. of daily prescription medications None1.00——1.00—— ≤32.460.53–11.41.252.560.28–23.10.40 4–55.911.30–26.79.0212.491.49–105.09.02 6–88.451.78–40.10.0110.021.10–91.48.04 >87.641.53–38.04.016.170.62–61.15.12AOR, adjusted odds ratio; CI, confidence interval; GED, general educational development; OR, odds ratio. Open table in a new tab