The Role of Physician Knowledge and the Inappropriate Initiation of PPI Therapy for Stress Ulcer Prophylaxis in the ICU: Presidential Poster

Abstract

Purpose: Our objective was to calculate the rate of inappropriate PPI initiation in an academic ICU and the contributing role of the prescribing physicians' knowledge of stress ulcer prophylaxis (SUP), as established by the American Society of Health-System Pharmacists (ASHP). Secondary aims were to determine the subsequent rate of continuation of a PPI on hospital discharge, the rate of missed indications for starting SUP and the rate of gastrointestinal bleeding in patients not started on a PPI. Methods: A retrospective chart review of medical ICU patients on no acid suppression was conducted from January, 2012 to December 31, 2012. Demographics including age, sex, ethnicity, comorbidities, and ICU length of stay were collected. The rate of inappropriate PPI initiation, the subsequent rate of continuation of PPI on hospital discharge and the rate of bleeding in patients not started on a PPI were determined. A 10-question questionnaire about SUP based on ASHP guidelines was administered to prescribing physicians. Results: Of 477 total patients, 212 patients were excluded due to concurrent acid suppression therapy. One hundred seventy-seven patients were started on a PPI for SUP and 88 patients were not. The average age was 54 years, 56% of the patients were male, 54% were Caucasian, 39% were African-American, with an average ICU length of stay of 6.7 days. Of the 177 patients, 101 patients (57%) were inappropriately started on a PPI for SUP. However, only three of these patients were subsequently discharged from the hospital on a PPI. Of the 88 patients not begun on a PPI, no indications for SUP were missed and no incidents of gastrointestinal bleeding were reported. Fifty residents who staff the ICU responded correctly to only 42% of the questions for a knowledge deficit of 58% for SUP. Conclusion: Our study demonstrates that 57% of patients started on SUP were done so inappropriately. The questionnaire suggested that the major determinant was a lack of knowledge of ASHP guidelines. We believe that the inclusion of a PPI in the ICU admission order set leads to reflexive ordering. The rate of patient discharge on inappropriate PPI prescription was very low, likely due to a medication reconciliation program that requires documentation of an appropriate indication. Given the implications of PPI use on Clostridium difficile infection, pneumonia, and the associated incremental costs, we have proposed targeting this knowledge gap with each resident receiving laminated cards with ASHP guidelines, posting these on all ICU computers, and removing PPIs from the admission order set. Based on the outcomes of our intervention we hope to implement such targeted measures in other areas.