2014 Finalist: UC San Diego Health System

San Diego, California
EHR “Best Practice Alerts” Improve Medication Safety in Kidney Patients

Patients with acute kidney injury and chronic kidney disease are at risk for frequent adverse medical events due to improper medication dosing. An interprofessional, pharmacist-led team at the University of California San Diego (UCSD), which was comprised of specialists in pharmacy and medical informatics, sought to design and implement a computerized decision support tool to improve the rate of appropriate medication prescriptions in patients with kidney impairment.  

UCSD promotes a culture of patient safety in all aspects of medication use. Senior UCSD leadership supported and funded the creation of the decision support tool.

The UCSD team used the “Best Practice Alert” function in Epic’s electronic health record software to create the Decision Support Tool. Twenty medications that require dosage adjustment for patients with kidney dysfunction were identified. Alerts for these medications were displayed during the order-entry process in the inpatient and ambulatory care settings. Additionally, the decision support tool was designed to include “look-backs” in which the system evaluated patients’ medication profiles and provided alerts on active medications when kidney function changed.

Physicians at UCSD were randomized to receive either standard of care (pharmacist review after order signing) or alerts in addition to standard of care. In both groups, the team prospectively reviewed opportunities for potential medication order adjustment to determine the proportion of study prescriptions that were appropriately adjusted for kidney function. Analysis used a multivariate logistic regression model that adjusted for glomerular filtration rate (GFR), gender, age, hospitalized status, length of stay, type of alert, interaction with time and clustering within the prescribing provider.

A total of 4,113 opportunities for potential drug dosage adjustment or discontinuation occurred in 1,649 unique patients. In this data set, 1,592 opportunities occurred in the intervention arm and 2,521 in the control arm.

The primary outcome—drug discontinuation or appropriate dosage adjustment—was 17 percent vs. 5.6 percent in the intervention and control arm, respectively (OR 2.66 [95%CI 2-3.53], p<0.0001). The effect of the intervention was sustained over time. In reviewing the effectiveness by type of alert, prospective alerting had a greater impact than “look-backs.” Also, drug dosage adjustment alerts had a greater impact than alerts on drug discontinuation in the setting of a contraindicated medication.  

A user group survey was conducted to assess alert acceptance. Survey results confirmed that providers felt the intervention to be of high value, non-intrusive and accurate. Study results were disseminated at the American Society of Nephrology annual meeting and the Epic Pharmacy Advisory Conference.  Feedback from these meetings, along with user survey results, was used to streamline the decision support tool. Future initiatives will expand renal decision support tools for pharmacists.

This UCSD project, conceived and led by pharmacists, illustrates the benefits of interdepartmental collaboration. Inappropriate drug prescribing in kidney impairment remains a patient safety concern. The results of this initiative show a decision support tool can improve drug prescribing, particularly when guiding physicians on new medication orders that require dosage adjustments for renal impairment.  The design of UCSD’s decision support tool can be readily adapted by other institutions using Epic or other comprehensive electronic health record systems to allow the tool’s benefits to be spread broadly within healthcare.

Interprofessional Team
Linda Awdishu, Pharm.D., MAS, Team Chair
Assistant Clinical Professor of Pharmacy

Carrie R. Coates, Pharm.D.
EMR Medication Process Manager

Charles Daniels, R.Ph., Ph.D.
Professor of Clinical Pharmacy
Associate Dean for Professional Practice

Robert El-Kareh, M.D., M.P.H., M.S.
Assistant Clinical Professor of Medicine

Adam Lyddane, Pharm.D.
Chief Applications Officer