Opioid and Benzodiazepine Co-Prescribing Not Curbed By Pharmacy Alerts

Email alerts to practitioners from pharmacists did not reduce concurrent prescribing of opioids and benzodiazepines, a randomized trial showed.

In a group of over 2,000 patients who recently received co-prescriptions, there were no differences in the number of prescribed days for opioids, benzodiazepines, or both between the email alert group and the usual-care group over 90 days, reported Adam Sacarny, PhD, of Columbia University Mailman School of Public Health in New York City, and colleagues:

• Opioids: adjusted difference 1.1 days (P=0.81)
• Benzodiazepines: adjusted difference -0.6 days (P=0.30)
• Both: adjusted difference -0.1 days (P=0.41)

These findings suggest that different evidence-based interventions should be considered when attempting to encourage adherence to the co-prescribing guidelines for patient care, the authors wrote in JAMA Health Forum.

“We have the sense now that the resources that were being devoted to these emails — a lot of pharmacists’ time and effort — could be deployed elsewhere to more valuable forms of patient care,” Sacarny told MedPage Today. “Because these emails were not changing patient care to the extent that we could see [an improvement], it kind of suggested that the emails were not useful for the doctors either. And as a result, it probably doesn’t make sense to send them.”

The secondary outcome of total prescribing by practitioners during the 90 days after their patients were enrolled also showed that the email alert did not reduce the total number of prescribed days for either drug:

• Opioids: adjusted difference -5.5 days (P=0.34)
• Benzodiazepines: adjusted difference 9.4 days (P=0.82)
• Both: adjusted difference 0.0 days (P=0.49)

The authors highlighted the ongoing concerns around concurrent prescribing, noting that “one-third to one-half of prescription opioid overdose deaths involve a benzodiazepine.”

“In 2017, more than 1 in 5 patients prescribed an opioid also received a benzodiazepine. While this rate has declined in recent years, 3 million adults still receive concurrent prescriptions (co-prescriptions) annually,” they added.

The main takeaway from this study is the continued need for rigorous, evidence-based approaches to addressing the issue, Sacarny said. He pointed out that just because an intervention might appear to have an effect, it does not mean that it is a net improvement.

“There’s a lot of evidence that the more alerts you send to clinicians, the more they tend to glance over other alerts — this idea of alert fatigue,” he added.

Sacarny said that the lack of a difference was a clear indication that this particular intervention did not achieve the desired effect, but that the outcome is still useful for researchers and practitioners.

“This is another sign of the value of doing rigorous evaluation of quality improvement work,” he said. “If we hadn’t used this randomized approach, we might have erroneously concluded that the intervention was effective.”

Other methods have been shown to reduce opioid and benzodiazepine prescribing, including letters from medical examiners notifying of a recent patient’s fatal overdose.

For this study, Sacarny and team randomized 2,237 patients with co-prescriptions from the National Capital Region of the Military Health System to email alerts (n=1,187) to their practitioners or usual care (control; n=1,050). Over half of patients were women (57%), and mean age was 48-49.

Patients received a mean of 31 days of opioids and 33 days of benzodiazepines in the 90 days before enrollment. More than half of patients had a mental health disorder diagnosis prior to enrollment, and one in nine had a substance use disorder diagnosis.

These patients had 789 practitioners eligible for emails. The email alert was sent by clinical pharmacists to practitioners with a message that said that their patient might have been concurrently prescribed an opioid and a benzodiazepine. The alert also encouraged practitioners to take immediate action to revise the patient’s treatment plan.

The majority of practitioners were physicians, most of whom were primary care physicians. Nonphysician practitioners were mostly physician assistants and nurse practitioners.