Factors That Affect Adequacy of Colon Cleansing for Colonoscopy in Hospitalized Patients

Abstract

Background & aims: Hospitalization is associated with inadequate colon cleansing before colonoscopy. We aimed to identify factors associated to inadequate colon cleansing among inpatients, and to derive and validate a model to identify inpatients with inadequate cleansing.

Methods: We performed a prospective observational study at 12 hospitals in Italy. Consecutive adult inpatients scheduled for colonoscopy for any indication were enrolled from February through May 2019 (derivation cohort, n = 1016) and from June through August 2019 (validation cohort, n = 508). Inadequate cleansing was defined as Boston bowel preparation scale scores below 2 in any colon segment. We performed multivariate logistic regression to identify factors associated with inadequate cleansing.

Results: In the combined cohorts, 1032 patients (68%) had adequate colon cleansing. Physicians' meetings to optimize bowel preparation (odds ratio [OR], 0.42; 95% CI, 0.27-0.65), written and oral instructions to patients (OR, 0.48; 95% CI, 0.36-0.65), admission to gastroenterology unit (OR, 0.71; 95% CI, 0.51-0.98), split-dose regimens (OR, 0.27; 95% CI, 0.20-0.35), a 1-liter polyethylene glycol-based bowel purge (OR, 0.39; 95% CI, 0.23-0.65), and 75% or more intake of bowel preparation (OR, 0.09; 95% CI, 0.05-0.15) significantly reduced odds of inadequate colon cleansing. Alternatively, bedridden status (OR, 2.14; 95% CI, 1.55-2.98), constipation (OR, 2.16; 95% CI, 1.55-3.0), diabetes mellitus (OR, 1.61; 95% CI, 1.18-2.20), use of anti-psychotic drugs (OR, 3.26; 95% CI, 1.62-6.56), and 7 or more days of hospitalization (OR, 1.02; 95% CI, 1.00-1.04) increased risk of inadequate colon cleansing. We developed a model to identify patients with inadequate cleaning using data from patients in the derivation cohort and tested it in the validation cohort. Calibration values were P = .218 for the discrimination cohort and P = .232 for the validation cohort. Discrimination values were c-statistic, 0.78 (95% CI, 0.74-0.81) for the discrimination cohort and c-statistic, 0.73 (95% CI, 0.69-0.78) for the validation cohort. We developed app for use by clinicians.

Conclusions: In a prospective observational study, we identified setting-, patient- and preparation-related factors that affect colon cleansing among inpatients. We derived and validated a model to identify patients with inadequate preparation and developed an app for clinicians. ClinicalTrials.gov no: NCT03925506.

Keywords: Bowel Preparation; Colon Cleansing; Colonoscopy; Hospitalized Patients; Inpatients; Predictive Model.