Standardizing inpatient colonoscopy preparations improves quality and provider satisfaction.

Abstract

Purpose: Inpatient colonoscopy bowel preparation quality is frequently suboptimal. This quality improvement (QI) intervention is focused on regimenting this process to impact important outcomes.

Design/methodology/approach: Define, Measure, Analyze, Improve and Control (DMAIC) methodology was employed, including generating a root-cause analysis to identify factors associated with inpatient bowel quality. These findings motivated the creation of a standardized electronic health record (EHR)-based order set with consistent instructions and anticipatory guidance for administering providers.

Findings: There were 264 inpatient colonoscopies evaluated, including 198 procedures pre-intervention and 66 post-intervention. The intervention significantly improved the adequacy of right colon bowel preparations (75.0 percent vs 86.9 percent, p = 0.04) but not overall preparation quality (73.7 percent vs 80.3 percent, p = 0.22). The intervention led to numerical improvement in the proportion of procedures in which the preparation quality interfered with making a diagnosis (10 percent-6 percent, p = 0.29) or resulted in an aborted procedure (3.5 percent-1.5 percent, p = 0.39). After the intervention, provider satisfaction with the ordering process significantly increased (23.3 percent vs 61.1 percent, p < 0.001).

Practical implications: The QI intervention significantly reduced the number of inpatient colonoscopies with inadequate preparation in the right colon, while also modestly improving the diagnostic yield and proportion of aborted procedures. Importantly, the standardized EHR order set substantially improved provider satisfaction, which should justify broader use of such tools.

Originality/value: Novel clinical outcomes such as ability to answer diagnostic questions were improved using this intervention. The results align with strategic goals to enhance provider experience and continuously improve quality of patient care.