365 The Effect of an Electronic Decision Support Tool on Emergency Department Prescribing of Naloxone

Abstract

The prescription of naloxone or “take-home naloxone” has been advised for patients who are at risk for opioid overdose. The provision of naloxone from the ED for these patients is a relatively new operational implementation. The purpose of this study was to evaluate the effect of an electronic health record (EHR) best practice alert (BPA) on the prescription of naloxone from the ED. This was a cross-sectional study performed at a tertiary care health system’s 2 EDs (1 urban and 1 suburban). Included were patients 18 years of age and older who were considered high-risk for opioid overdose based on: 1) the listing of methadone or fentanyl formulations in the EHR medication list, and 2) a documented diagnosis of “substance abuse” or the ICD10 equivalent diagnosis in the EHR and 3) prescription of an opioid at ED discharge. Methadone and fentanyl derivatives were chosen because they are considered high-risk opioids as identified by the Joint Commission Sentinel Event Alert on opioids. A BPA fired in the EHR for those patients who met these high-risk criteria. The BPA informed the ED provider of the patient’s high-risk status for opioid overdose and asked the ED provider if a discharge prescription for naloxone should be issued. The provider could choose to prescribe naloxone via the BPA or disregard the prompt. Data on naloxone prescribing rates at ED discharge were collected from 5/1/2016 after the BPA was implemented through 10/31/2017. There were no prescriptions given at discharge for naloxone from the ED during the 18 months prior to the implementation of the BPA. During the 18 months after the BPA was implemented, there were 596 ED discharges for which the BPA was triggered. Among these, 22 (3.7%) received a prescription at discharge for naloxone. An additional 73 patients received a prescription without receiving a BPA. The majority of patients were male (55.3%) and were between 35-64 years of age (69.2%). Visits with a naloxone discharge prescription were more likely to have a higher acuity ESI score of 1 or 2 compared to those without a prescription (15.4% versus 5.8%, p<0.001). Visits where a naloxone prescription were more likely to be given compared to visits where it was not given were those with an ESI score of 1 or 2 (15.4% versus 5.8%, p<0.001). Among visits with and without a naloxone prescription, there were no differences among sex, age group or patients with multiple visits (p’s>0.05). Implementation of a BPA in the EHR used to identify patients considered high-risk of opioid overdose resulted in a small increase in the rate of naloxone prescriptions issued for this population and a higher increase in prescribing among those not considered high-risk for overdose. Reasons for higher rates of naloxone prescribing among those patients for whom the BPA did not activate in the EHR may be attributed to general increased provider awareness of the availability of the antidote. Overall, more focus is needed to facilitate prescribing of naloxone from the ED among high-risk patients.