Additional results of the study showed patients who received prescriptions for long courses of the drugs – more than 30 tablets of oxycodone 5 milligrams – were five times more likely to fill additional opioid prescriptions in the next six months than those who received just a few days’ supply – 10 tablets of less. The findings, reported by researchers at Penn Medicine and published in the Annals of Emergency Medicine, showed wide geographic variability in prescribing patterns for minor injuries.
“Although opioids are not – and should not – be the first-line treatment for an ankle sprain, our study shows that opioid prescribing for these minor injuries is still common and far too variable,” said lead author Dr. M. Kit Delgado, an assistant professor of emergency medicine and epidemiology at Penn. “Given that we cannot explain this variation after adjusting for differences in patient characteristics, this study highlights opportunities to reduce the number of people exposed to prescription opioids for the first time and also to reduce the exposure to riskier high-intensity prescriptions.
In the study, researchers examined private insurance claims data from more than 30,800 patients visiting U.S. emergency departments for an ankle sprain from 2011-2015. All patients included in the study had not filled an opioid prescription within the past six months. Overall, 25 percent of patients received a prescription for an opioid pain medication such as hydrocodone or oxycodone. However, there was wide variation across states, threefold between the low vs. high prescribing states, and at the extremes it was over tenfold, with only 3 percent of patients received an opioid prescription in North Dakota, compared to 40 percent in Arkansas. The authors admit that the extreme variation between the two states could be explained by smaller sample sizes, despite the results being statistically significant. Nevertheless, the overall pattern of variation across states suggests that there is significant room to reduce unnecessary prescribing for this condition.
In total, more than 143,000 opioid tablets were prescribed for patients in the study sample who filled prescriptions. The authors noted bringing states with above-average prescribing rates down to the average prescribing rate 24.1 percent would result in 18,000 fewer opioid tablets prescribed. Similarly, reducing the number of tablets given with each prescription to the average 16 tablets would result in 32,000 fewer tablets prescribed.
“Although prescribing is decreasing overall, in 2015 nearly 20 percent of patients who presented with an ankle sprain were still given an opioid, a modest decrease from 28 percent in 2011,” Delgado said. “By drilling down on specific common indications as we did with ankle sprains, we can better develop indicators to monitor efforts to reduce excessive prescribing for acute pain.”
Furthermore, Delgado said given that the study found high-intensity prescriptions were associated with prolonged use not related to the original ankle sprain, the study supports guidelines and policies aimed at reducing the size of new, initial opioid prescriptions.
“There is a clear need for further impactful guidelines similar to the CDC guidelines that outline more specific opioid and non-opioid prescribing by diagnosis,” said senior author Dr. Jeanmarie Perrone, a professor of emergency medicine and director of medical toxicology at Penn Medicine. “Medical, surgical and subspecialty societies should convene to propose best practices similar to the popular ‘Choosing Wisely’ campaign, acknowledging that pain management for most diagnoses can be accomplished with non-opioids. And certainly, ankle sprains are a model example.”
In the end, the authors said the goal should be to maximize non-opioid alternatives for pain management of minor injuries and, if opioids are absolutely necessary, to use the lowest initial dose possible, which should be no more than 10-12 tablets of common short-acting formulations.
“Simply making these amounts the default setting electronic medical record orders could go a long way in reducing excessive prescribing as our previous work has shown,” Delgado said. “It would be great to see analyses such as ours replicated in other settings, such as post-operative prescribing, where prescriptions are higher intensity. In these settings there may be greater opportunities to decrease transitions to prolonged opioid use by reducing excessive prescribing.”
Preliminary results of the study were presented last year at the annual meeting of the Society for Academic Emergency Medicine. The National Institute on Drug Abuse and the Leonard Davis Institute Small Grants Program funded the study.