Medical Examiner

Can a Nurse Practitioner Replace a Physician?

Data and personal experience suggests it’s possible. The current shortage of doctors attending veterans might make it necessary.

A nurse checks a patient’s blood pressure at the Remote Area Medical health care clinic in Wise, Virginia.
A nurse checks a patient’s blood pressure at the Remote Area Medical health care clinic in Wise, Virginia.

John Moore/Getty Images

If you or your child were sick and were assigned to see a nurse practitioner instead of a physician, would you hesitate?

As a medical student in Cleveland, John was diagnosed with skin lymphoma, a rare form of cancer. He was terrified partly thanks to the word “cancer” but mainly by the uncertainty of his prognosis.

Soon after his diagnosis, he moved to Boston for his residency training at Harvard Medical School. His dermatologist devised a treatment plan and promptly referred him to a nurse practitioner named Marianne, who practiced within a physician-led team. He initially wondered whether he should switch to only seeing physicians, but those thoughts were fleeting: Marianne provided excellent and compassionate care, making him feel perfectly at ease while also paying close attention to the details of his case. Under the care of his medical team and Marianne’s watchful eye, John started to get better.

As it turns out, John’s experience having a nurse practitioner as his primary point of contact may become less and less unique—particularly for veterans.

Currently, a heated national debate is taking place centered on the Department of Veterans Affairs’ plans to grant full independent practice authority to nurse practitioners in an attempt to improve unacceptably long wait times. Although a recent RAND Corp. report found the VA meets most veterans’ needs overall, at certain care facilities, the wait to see a physician has reached an average of 41 days—resulting in an increased risk of hospitalization or death among veterans (Disclosure: One of us, John, works for RAND but was not involved in this report). The American Medical Association, the nation’s largest physician organization, strongly opposes this proposed new rule, arguing that autonomous nurse practitioners would undermine the quality of care delivered to veterans. And although the VA has been attempting to also hire more physicians, recruitment has been hampered by numerous challenges such as the lower salaries they offer compared to the private sector.

Such controversy is hardly new. Like any guild, physicians have a vested interest in protecting their professional turf and are loath to let other providers, particularly those with less experience and intensive training, assume an equal role. The nurses are equally confident in their ability to step up. A 2013 study in the New England Journal of Medicine surveying physicians and nurse practitioners found that while most physicians believe nurse practitioners provide inferior quality of care, most nurse practitioners felt they provided better quality of care than physicians.

Who’s right?

Neither, according to a study we conducted that was published in June in the Annals of Internal Medicine. We performed a 15-year analysis of nearly 29,000 U.S. patients who saw either a primary-care nurse practitioner, physician assistant, or physician, and compared the group’s rate of inappropriate medical services—such as prescribing low-quality antibiotics for colds; inappropriate and costly use of diagnostic imaging (such as CT scans and MRIs) for back pain, headache and colds; and unnecessary referrals to specialists for those same three conditions. All categories of providers made these choices at similar rates. Although our study did not include VA patients and examined a relatively narrow slice of care, it supports the idea that nurse practitioners can provide equal quality and efficiency of care compared with physicians for common illnesses across diverse primary-care settings.

So should the VA finalize its policy proposal? Before we answer, there are two major caveats to consider.

First, our study focused on relatively straightforward conditions with clearly defined guidelines (e.g., the common cold, lower-back pain), rather than complex conditions that might also come up in routine practice. Other studies (including one at the VA) showed that nurse practitioners tend to overuse costly imaging studies when dealing with conditions that might be more complicated or less guideline-based. But while nurse practitioners might actually raise costs for more complex cases, the jury is still out on their overall cost-effectiveness as they also have lower salaries than physicians.

Second, a 2014 systematic review of 24 randomized trials comparing nurse practitioners and physicians found that patients seen by nurse practitioners had similar health outcomes, and patients seen by nurse practitioners even reported better satisfaction with their care. Case closed? Not quite: Many of the trials in this review had important shortcomings. For example, most of these trials comparing nurse practitioners with physicians didn’t measure diagnostic accuracy (how often a provider got the correct diagnosis) or medical errors—both critical aspects to quality of care. They also often didn’t account for whether nurse practitioners were consulting a physician as they went, which would have implications for how the VA’s system is implemented. We need more high-quality randomized trials to provide clarity on these issues.

Given these uncertainties, rather than fully institute this policy across the board, we propose that the VA study these questions by randomly expanding the purview of nurse practitioners in some regions. The VA should also better define how nurse practitioners would fit into a team-based environment, which can improve quality of care. For instance, will they still be able to consult physicians for complex cases? Will physicians receive compensation for assisting? These are critical questions that should be answered with data-driven results.

But beyond answering these important questions, the VA still faces very real challenges in providing veterans timely access to primary-care services. Given these problems, expanding the role of nurse practitioners makes sense, as long as it’s done thoughtfully. We are not advocating that nurse practitioners replace physicians but they can, and should, be allowed to expand their scope of practice in primary care, particularly for relatively straightforward conditions with clearly defined guidelines much like the ones we studied.

Time and time again, research has shown that nurse practitioners working in collaborative settings can provide high-quality care. John, who has been cancer-free since 2011, experienced this care firsthand. The VA has a rare opportunity to expand access to care for millions of veterans in need and answer critical questions about nurse practitioners’ role in the practice of medicine—let’s hope they don’t squander it.