Impact of a Multifaceted Intervention on Prescribing of Proton Pump Inhibitors for Stress Ulcer Prophylaxis in the Critically Ill

Abstract

Background: Lack of well-supported evidence clearly defining one agent as superior to the other for use in stress ulcer prophylaxis (SUP) has led to an array of treatment strategies. Studies comparing histamine-2 receptor antagonists (H2RAs) and proton pump inhibitors (PPIs) have shown H2RAs to be non-inferior to PPIs at preventing clinically significant gastric bleeding. This, in addition to the decreased cost associated with H2RAs, has led to the adoption of H2RAs as the preferred agent for SUP at our institution. Objectives: To evaluate the impact of developing a guideline and removing PPIs from computerized order entry PowerPlans on PPI prescribing for SUP in critically ill patients. Patients and Methods: Members of the critical care service developed a guideline to direct SUP and removed pantoprazole from two critical care admission PowerPlans. In this interventional study with a historical control, adult patients admitted to the medical or surgicaltrauma ICU and prescribed a PPI were evaluated during two time periods before and after the interventions. Patients were excluded if they were receiving a PPI for a reason other than SUP. Patients were assessed daily for an indication for SUP. These included mechanical ventilation, high dose steroids, and coagulopathy. Results: A total of 92 and 60 patients were included in the pre-and post-intervention groups, respectively. PPI use for SUP was reduced from 38% in the pre-intervention group to 15.6% in the post-intervention group (P < 0.0001). PPI days adjusted for LOS were not different between the pre-and post-intervention groups (84% vs 86%, P = 0.5909). Of the total PPI days, 52.4% (95% CI 48.8 -56.1) were classified as inappropriate because there was no indication, while the remaining 47.6% (95% CI 43.9 -51.2) had an indication but should have received a H2RA. The total cost associated with guideline non-adherence was $1802. Conclusions: Our multifaceted intervention reduced the number of days in which a PPI was prescribed for SUP in the overall ICU population; however, it did not impact the duration of PPI therapy. Patients either had no indication for SUP, or had an indication where less costly H2RAs could have been used. Additional opportunity exists to improve cost-effective prescribing of SUP for critically ill patients.

The Role of Electronic Medical Records in Reducing Unwarranted Clinical Variation in Acute Health Care: Systematic Review

“…Almost all the studies were based on the implementation of an intervention (new or refined) into a clinical setting (35/36, 97%) with 1 qualitative analysis of EMRs by clinicians [36]. Most studies were quality or process improvement based (28/36, 78%) [33][34][35]37,[39][40][41][42][43][44][45][47][48][49]51,52,54,55,58,59,[61][62][63][64][65][66][67] [33][34][35][36][39][40][41][42][43][46][47][48]50,52,54,58,60,62,63,65,66,70]. Many papers addressed the minimization or elim...…”

“…Regarding overall disposition, most studies reported positive results (30/36, 83%) [33][34][35][37][38][39][40][41][42][43][45][46][47][48][49][50][51][52][53][54][58][59][60][62][63][64][65][66][67]70], while a minority reported mixed [44,55,56] or negative results [36,57,61] (3/36, 8% each). That is, the authors concluded in most studies that the EMR was used successfully as part of an initiative to address clinical variation.…”

“…However, most studies did not measure or report variation. Of the 5 codes for coding variance (Figure 3), no studies reported Type 1 or 3, half reported Type 2 (18/36, 50%) [33][34][35]38,44,[46][47][48][49][50]53,56,58,59,62,[65][66][67], some reported Type 4 (13/36, 36%) [37,39,40,43,45,52,55,57,60,61,63,64,70], and a few reported Type 5 (5/36, 14%) [36,41,42,51,54]. The studies that reported results for variation coded as Types 2 and 4 generally examined how an intervention led to changes in the average of a clinical behavior.…”

“…Regarding the quadruple aims of health care outcomes, over half of the studies addressed individual care outcomes (19/36, 53%) [33,35,37,38,40,[42][43][44][46][47][48][49]51,53,54,58,60,63,65], many examined efficiency (21/36, 58%) [33,34,37,38,[40][41][42][43]46,47,49,51,52,55,56,58,60,61,63,64,67], a handful examined clinician experience (6/36, 17%) [33,34,36,39,41,61], and none examined population health outcomes (Table 1). Some studies examined just 1 quadruple aim of health care outcome (13/36, 36%) [35,36,…”

“…1. Publication bias: most papers (30/36, 83%) reported positive results [33][34][35][37][38][39][40][41][42][43][45][46][47][48][49][50][51][52][53][54][58][59][60][62][63][64][65][66]70], with a minority reporting mixed [44,55,56] or negative results [36,57,61] (3/36, 8% for both). The completeness of results including nonsignificant effects was not always assured.…”