Impact of Prospective, System-Wide Intervention to Influence Opioid Prescribing Practices Among Patients with Back...

In response to the opioid epidemic, our multidisciplinary team designed and integrated an alert-based, clinical-decision support intervention which identifies patients at risk of opioid misuse based on five evidence-based risk factors (early refill of opioids/benzodiazepines; >2 ED/Urgent Care visits with onsite opioids; >3 prescriptions of opioids/benzodiazepines; prior overdose; and positive toxicology screen).

To evaluate the impact of the intervention on prescribing decisions for back pain by measuring the percent of opioid prescriptions modified in response to the alert.

A total of 93,192 adult patients presenting to the emergency department with complaints of back pain from 2017-2021 were included in this prospective, observational study. We calculated rates of "decision influenced" (modifying or canceling prescriptions) in response to the PRIMUM intervention and characterized patients, encounters, and prescriptions in this population.

The 30.2% of back pain patients received an opioid prescription. Among patients prescribed opioids, 18.6% had a risk factor. An alert fired in 6,501 (19.8%) encounters, and positive toxicology was the most common risk factor (52.1%). The prescriber decision was influenced in 430 of these encounters overall (6.6%) and was highest for three or more prescriptions in the past month (11.8%) and early refill (9.1%). Chronic patients were more likely to receive opioids.

Roughly 1 in 3 patients presenting to the emergency department for back pain received an opioid. A clinical decision support intervention to identify patients at risk of opioid use disorder had a minimal influence on opioid prescribing decisions in this population.