S888 Optimizing Proton Pump Inhibitor Use at a Safety-Net Provider Hospital

Introduction: In 2022, the American Gastroenterological Association released a clinical practice update regarding de-prescribing proton pump inhibitors (PPI). Anywhere from 7-15% of patients use PPIs at any time. Patients over age 70 have an estimated prevalence of 40% for PPI use. PPI use in the long term has been associated with several adverse effects including chronic kidney disease, electrolyte imbalances, and dementia. These associations are from observational studies and no direct causality can be created between the 2 groups. Nonetheless, the long-term safety of PPI use is brought into question and further evaluation is needed. In this study, we sought to evaluate this at LBJ Hospital in Houston to evaluate the extent of PPI overprescription and apply interventions to reduce overprescription. Methods: Patients discharged from LBJ Hospital from July 1, 2024, to December 31, 2024, were included in the study. Institutional review board approval was obtained. Baseline characteristics were collected along with data surrounding the PPI prescription, type, duration, dose, and indication. Follow-up data was also collected including any esophagogastroduodenoscopies (EGD), adverse events, or reactions. Protected health information was protected using a linking log and all information was securely stored on one device with an encrypted file. Results: Of the 260 patients discharged with PPIs, 73 (28.0%) of those patients did not have an appropriate indication. Notably, 155 (59.6%) patients did not have an appropriate duration for PPI. Most commonly, patients were given a 90-day prescription though the average prescription length for patients was 122 days. The most common diagnosis associated happened to be gastroesophageal reflux disease with 61 patients having the diagnosis code. One hundred three (39.6%) patients did not have provider follow-up on the PPI for discontinuation or adjustment. Seventy-three (28.0%) underwent EGD near time of diagnosis with 28 (10.8%) patients having esophagitis, ulcers, erosions, or Barrett’s esophagus that would warrant therapy. However, only 2 patients qualified for prolonged treatment. Conclusion: We were able to demonstrate a significant portion of patients who both had an inadequate diagnosis associated with prescription as well as those with an inappropriate duration of prescription. This is likely due to the limited access to follow up at the institution as well as a mechanism of the default prescription. Additionally, discussion of reflux is often ignored in hospitalization where not as much time is able to be dedicated to thorough prescription monitoring.