The multifaceted problem of sharply rising opioid prescribing and related overdose deaths, which peaked in 2009, has been further analyzed by a research group in a recently published paper PM R 2014 Jan 9.
The Stanford University team used 1997-2009 data from the large National Ambulatory and National Health Ambulatory Medical Care Surveys from the Centers for Disease Control and Prevention to study opioid-related trends in ambulatory patients. Using weighted multiple logistic regression, they found that the rapid rise in opioid prescribing took place over and above the concurrent significant increase in low back pain incidence, and was independent of insurance status and demographic shifts. Overall, patients with low back pain were the most likely to receive an opioid prescription.
“We’ve known for a while that evidence shows—and therefore guidelines recommend—the best approach to back pain is not so much strong medications, but rather conservative management,” said lead investigator Ming-Chih Jeffrey Kao, MD, PhD, clinical assistant professor, pain management and physical medicine and rehabilitation in the Departments of Orthopaedic Surgery and Anesthesiology at Stanford Hospital and Clinics, in Redwood City, Calif. “Therefore, we’re seeing more and more reliance on strong pain medications.”
Other striking findings included specialists being 35% more likely to prescribe an opioid to patients aged 60 to 90 years than to those aged 16 to 29. Findings also showed that they were 22% less likely to give women an opioid prescription than men; 23% more likely to give an opioid prescription to a Medicare patient than one with private insurance; and 30% more likely to hand such a prescription to a Medicaid patient than someone with private insurance.
A REMS Effect?
There is some evidence that the trend of increased opioid prescribing began reversing in 2009 (Figure) and continued to do so after the FDA implemented a Risk Evaluation and Mitigation Strategy (REMS) for extended-relase/long-acting prescription opioids in July 2012, said Mark Brandenburg, MD, who was not involved in the study.
“We do believe that there might be an association between the recent decrease in unintentional overdose deaths and a decrease in the prescribing of long-acting opioids; however, this has not yet been proven,” said Dr. Brandenburg, an emergency physician in Bristow, Okla., and chair of the Opioid Prescribing Guidelines for Oklahoma Workgroup.
Mehul Desai, MD, MPH, director of spine, pain medicine and research at the International Spine, Pain & Performance Center, in Washington, D.C., who also was not involved in the study, agreed that follow-up investigations are needed to assess more recent changes in prescribing patterns. He noted the study data do not indicate whether there was an increase or decrease in the number of pills or dose in each prescription.
“It is possible that when the doses are converted to equianalgesic doses, they may not represent an increase; nonetheless, the trend to 2009 is noteworthy and alarming,” Dr. Desai said. “Part of this trend may have been clinicians merely fulfilling patient desires and expectations. For example, a primary care provider may start a patient on opioids, increase the dose, then become uncomfortable and send the patient to me—and then I feel compelled to continue to write for them.”
Dr. Desai said regulators should consider restricting the prescribing of pain medications to physicians with “appropriate levels of training, as long as it is appropriate and well defined—that is, fellowship-trained, board-certified or having undergone specific, federally mandated training.”
Other main findings were:
The probability of a patient receiving an opioid prescription for any diagnosis increased every five years by 33% in the emergency department (ED), 29% in primary care clinics and 53% in specialty clinics.
After adjusting for demographics and insurance source, during every five-year period between 1997 and 2009 in the ED, primary care clinic and specialist clinic, the rate of opioid prescribing overall and for back pain in particular outstripped the rate of back pain diagnoses.
Patients aged 60 to 90 were 24% less likely to receive an opioid prescription in the ED than those aged 16 to 29, whereas in specialist clinics, individuals in the older age bracket were 35% more likely to receive opioids than those aged 16 to 29, but were less likely than individuals in the 30-to-59 age group.
In the ED and primary care clinics, but not in specialist practices, blacks, Hispanics and other races/ethnicities were significantly less likely to receive an opioid prescription than non-Hispanic whites.
Dr. Kao and his colleagues have developed an opioid REMS continuing medical education course for Stanford providers.