Nurse practitioners were no more likely than physicians to inappropriately prescribe medications to older patients, researchers found.
In an analysis of Medicare Part D claims, the mean rates of inappropriate prescribing by NPs and primary care doctors were “virtually identical” at adjusted rates of 1.66 and 1.68 per 100 prescriptions, David Studdert, LLB, ScD, of Stanford University, and colleagues reported in Annals of Internal Medicine.
Michelle Keller, PhD, MPH, and Catherine Sarkisian, MD, MSHS, of the University of California Los Angeles, wrote in an accompanying editorial that the study is “to some extent reassuring.”
“NPs are no worse than primary care physicians when it comes to inappropriate prescribing for older adults,” they observed.
However, “the longer answer is that [the authors] provide further concerning evidence that there remains not only persistently unacceptably high rates but also substantial variation in potentially inappropriate prescribing among clinicians of all stripes.”
For their study, Studdert and colleagues evaluated claims for Medicare Part D beneficiaries age 65 and up in 2013 to 2019, in the 29 states that had granted NPs prescribing authority by the end of that time period. They assessed inappropriate prescribing rates for 23,669 NPs and 50,060 primary care doctors. The majority of NPs were women (89%) while 39% of physicians were women, and 40% of NPs and 52% of physicians practiced in large metropolitan areas.
Inappropriate prescriptions were defined using the American Geriatrics Society’s Beers Criteria as drugs that carry an unfavorable risk-benefit ratio for adults age 65 and up.
Overall, they found that crude rates of inappropriate prescribing between NPs and primary care physicians were very similar, at 1.63 versus 1.69 per 100 prescriptions, and in further analyses, the odds of inappropriate prescribing were “virtually identical” for both groups (adjusted OR 0.99, 95% CI 0.97-1.01).
They noted that the adjusted averages “mask substantial underlying differences in inappropriate prescribing patterns,” noting that NPs were overrepresented among clinicians with the highest and lowest rates of inappropriate prescribing. For instance, NPs represented 32.1% of the study sample, but accounted for 51.8% of clinicians in the lowest decile of inappropriate prescribing and 48.8% of those in the highest decile.
Studdert and colleagues noted that the variations in inappropriate prescribing across states were larger than discrepancies between these clinicians within states. Utah and Virginia had the highest rates of inappropriate prescribing for both NPs and physicians, they said, noting that they were nearly two times greater than the corresponding rates in Hawaii and the District of Columbia, which had among the lowest.
By contrast, the between-clinician differences were generally “modest,” they wrote, with two-thirds of states having ratios of NP-to-physician rates of inappropriate prescribing between 0.9 and 1.1.
The study was limited because it may not be generalizable to younger patients, and because Beers Criteria has several weaknesses including the fact that it measures “potentially inappropriate prescribing,” Studdert said. Other limitations include the lack of patient-specific data and the inability to discriminate between refills and new prescriptions.
Still, the authors concluded that the study “adds to growing evidence indicating that when prescriptive authority is expanded to include NPs, these new prescribers do not perform worse than physicians.”
Keller and Sarkisian similarly concluded that the findings “add to a long list of empirical work showing that NPs provide equal or better quality of care when compared with their physician colleagues in primary care.”
Studdert and colleagues noted that the U.S. is in the midst of a primary care shortage and there have long been debates over “restrictive scope of practice laws” around the safety and quality of care provided by non-physicians. Their study, they said, suggests that a better way to cut inappropriate prescribing is to identify states that have relatively high rates of prescribing and try to determine what’s driving those behaviors.
“That seems a lot more productive than trying to … focus on whether nurse practitioners should be prescribing or not,” Studdert said, “because that’s not the big source of variation.”
The study was supported by fellowship grants from the Robert Wood Johnson Foundation and the National Science Foundation.
The researchers reported no conflicts of interest.
Sarkisian reported relationships with NIH. Keller had no disclosures.
Annals of Internal Medicine
Source Reference: Huynh J, et al “Inappropriate prescribing to older patients by nurse practitioners and primary care physicians” Ann Intern Med 2023; DOI: 10.7326/M23-0827.
Annals of Internal Medicine
Source Reference: Keller MS, Sarkisian CA “Beyond Titles: The Need to Reduce Prescribing Variation of Potentially Inappropriate Medications Among All Clinicians” Ann Intern Med 2023; DOI: 10.7326/M23-2556.