Implementation of a user-centered clinical decision support tool did not increase the rate of patients receiving buprenorphine for treatment of opioid use disorder in the emergency department (ED) when compared to usual care.
When the tool was used, it was associated with high rates of buprenorphine initiation, and the tool increased the number of physicians who provided ED-initiated buprenorphine and naloxone prescriptions.
The findings of EMBED were published in the BMJ on June 27.
From November 2019 to May 2021, researchers conducted EMBED, an NIH Pragmatic Trials Collaboratory Trial, in 18 clusters across 5 healthcare systems in 5 states, representing the Northeast, Southeast, and Western regions of the United States.
The 18 clusters included 21 EDs, with 3 sites paired with a second ED within the same healthcare system due to low patient volume or high physician crossover rates between the 2 sites. Each cluster received either the intervention or maintained usual care.
The study sample was made up of all patient visits to the ED during the study period. Eligible patient encounters were identified using criteria in the electronic health record. For patients with multiple ED visits during the trial, only the first ED visit was included in the analysis.
ED attending physicians caring for patients who met the trial’s inclusion criteria were the primary study subjects, but all ED clinicians practicing at intervention sites had access to the EMBED tool for all patients in their care at that site. A total of 5,047 patients (2,787 intervention and 2,260 usual care) were eligible for analysis along with the 599 physicians who cared for them (340 intervention, 259 usual care).
The intervention did not affect the patient-level rate of ED-initiated buprenorphine (which was 12.5% in the intervention arm and 12.0% in the usual care arm). The proportion of physicians who initiated buprenorphine in the ED at least once was higher at intervention sites compared to usual care sites (44.4% vs 34.0%).
In the intervention arm, the EMBED tool was launched by physicians for 261 of 2,787 eligible patients. When the tool was launched, buprenorphine was ordered for 161 patients (67% of 261 encounters).
“We built tools that work within the electronic health record that make it easier for clinicians to treat patients with addiction, using medications that physicians may initially find complex and unfamiliar,” said Dr. Edward Melnick of Yale University, co–principal investigator of the study. “We’ve tried to make that easier for clinicians to allow this practice to be adopted into routine care.”
“EMBED offers a major opportunity to accelerate adoption by offering a simple, automated, rapid solution to initiating treatment and referral for people suffering from opioid use disorder,” said Dr. Gail D’Onofrio of Yale University, co–principal investigator of the study. “Additional prompts and interventions will be needed to increase more universal implementation into the routine care of ED patients with opioid use disorder.”
Watch Dr. Melnick discuss the EMBED project in this NIH Pragmatic Trials Collaboratory interview from April 2022.
EMBED was supported within the NIH Pragmatic Trials Collaboratory by a cooperative agreement from the National Institute on Drug Abuse and by the NIH Common Fund through a cooperative agreement from the Office of Strategic Coordination within the Office of the NIH Director.