A study came out last week that compared emergency department physicians on their opioid prescribing patterns. They found that different physicians, even within the same hospital, could have wildly different rates of opioid prescribing patterns, with those who were considered high-intensity prescribers being more likely to have patients with long-term use of opioids.
The study was published last week in the New England Journal of Medicine, and looked at 377,000 Medicare recipients who had an emergency department visit between 2008 and 2011, without a prescription for opioids in the previous 6 months. Physicians were categorized as either high-intensity or low-intensity based on their behaviour relative to peers in the same hospital, and they then followed up patients to see how many had 6 months of use in the a year after their visit. They also controlled for other factors, such as diagnosis, that might influence use of opioids.
Now here’s where things get interesting: differences in your chances of being given opioids varied significantly between providers. While the high-intensity group prescribed opiods 24.1% of the time, the low-intensity group only prescribed them 7.3% of the time; almost a three-fold difference in prescribing patterns. On top of this was the finding that prescribing patterns were remarkably consistent regardless of patient characteristics; the high-intensity group always prescribed more opioids than the low-intensity group.
They then followed patients to see how many continued to use opioids for at least 6 months in the next year following their visit. Among the high-intensity prescribers, 1.51% of patients were considered long-term opioid users. On the other hand, the rate among the low-intensity prescribers was 1.16%; almost a 30% difference. Put another way, if this is a true causal relationship, this means that for every 48 patients treated with opioids, there is one extra long-term opioid user (for those who are interested, the authors go into extensive detail as to how they evaluated for potential selection bias, as well as a discussion about casuality).
All of this adds to a growing body of literature about the opioid crisis. The patients under study were relatively old (mean age of 68 years), and so opioids may have been appropriate, as pain management is tougher among older than younger patients. However, this is then complicated by the side effects of opioids, which include kidney and blood pressure problems. In addition, the authors are quick to point out that this study is not meant to point blame at ED physicians, as they are only part of the care journey that patients experience. As they point out:
Of course, prescriptions provided by other physicians in the months after an emergency department visit are necessary for long-term opioid use to take hold. Conversion to long-term use may be driven partly by clinical “inertia” leading outpatient clinicians to continue providing previous prescriptions. Such clinical inertia may affect only a narrow segment of the population; this could explain why rates of initial opioid prescribing may vary by a factor of three, whereas long-term use varies by only approximately 30%.
The wide variation in prescribing patterns suggests that this is a potential area for intervention. In some situations, especially in the pressure situation of the emergency departments, physicians may be prescribing opioids without realizing they are doing it. Understanding the differences in prescribing patterns in the low-intensity and high-intensity group can help pinpoint where and how researchers and practitioners can intervene to ensure that when opioids are used, they are used effectively. Deeper understanding of the physician decision-making process could be a potential avenue for reducing the prescribing of opioids to these patients.
Barnett, M L, et al. “Opioid-Prescribing Patterns of Emergency Physicians and Risk of Long-Term Use.” NEJM, vol. 376, 16 Feb. 2017, pp. 663–673., doi:10.1056/NEJMsa1610524. Accessed 20 Feb. 2017. Available online at http://www.nejm.org/doi/full/10.1056/NEJMsa1610524#t=articleResults