Interview with Dr. Joanne Spetz, PhD, Professor in the School of Nursing at UCSF

About Dr. Joanne Spetz, PhD: Dr. Spetz is a Professor in the Institute for Health Policy Studies, the Department of Family and Community Medicine, and the School of Nursing at the University of California, San Francisco (UCSF). She is the Director of the UCSF Health Workforce Research Center on Long-Term Care, and serves as the Associate Director for Research at Healthforce Center at UCSF. Her areas of expertise include labor economics, econometrics, public finance, and their connection to the accessibility and quality of patient care across different demographics. Dr. Spetz has conducted extensive research on the health care workforce, health care technology and its impact, and youth substance abuse. In addition to her work in health care and economics research, Dr. Spetz has served as a consultant to the Institute of Medicine Committee on the Future of Nursing, and was a member of the Institute of Medicine Standing Committee on Credentialing Research in Nursing.

Dr. Spetz earned her M.A. and her PhD in Economics from Stanford University in 1993 and 1996, respectively. She earned her S.B. in Economics in 1990 from the Massachusetts Institute of Technology, where she majored in Economics. In 2011 she became an Honorary Fellow of the American Academy of Nursing, and in 2016 she received the Mentorship Award from AcademyHealth’s Interdisciplinary Research Group on Nursing Issues.

Interview Questions

[] In your research, what were some of your principal findings in terms of the impact of expanded APRN scope of practice on patient access to health care and patient health outcomes?

[Dr. Joanne Spetz] There have been two projects I’ve completed that speak to APRN scope of practice most directly. One was a study looking at advanced practice nursing care at retail health clinics, and which built off of a previous study that one of my colleagues did that looked into whether retail health clinics actually save money. My colleague’s study looked at insurance claims data to see if, after seeing a clinician at a retail health clinic, a patient also had to go to the hospital or the doctor’s office because their needs weren’t met. In other words, they were seeing if retail clinics were essentially making patients pay twice for the same service. And what he found was this was not the case: if you have a condition that you go to a retail health clinic for, you will generally get what you need there, and will not show up at the ER or your doctor’s office.

And I remember our talking with him about his findings, and I said, “I wonder if health care services costs at these clinics differ when nurse practitioners are under restricted scope of practice, versus when they are able to operate independently.” Because when NPs are restricted, they must have physician supervision when working with patients, which may entail physician review of a certain number of their charts, obtaining approval for certain prescriptions or treatments, and other steps that bring with them additional costs. So we added that variable to the analysis, and found that indeed, in states with restricted APRN scope of practice, cost savings at retail health clinics were a lot smaller. We published our findings in Health Affairs, and it was one of the few papers at that point that had really looked at whether these scope of practice regulations were costing us more money.

Before then, people typically talked about access to care and professional standards for advanced practice nurses. And I’m an economist, and I believe that money talks. If you’re telling insurance companies that you’re spending more money because NPs have to have a collaborator or a supervisor, then maybe that will get the insurance industry to advocate for giving NPs full practice authority as well.

I was a little surprised at the result, and I thought it was equally likely that we would not see any difference between the clinics in restricted practice states versus full practice states. We still don’t quite know what was in the black box–by that I mean we don’t know how the cost increases came across, or what the cost breakdowns were. We didn’t know whether it was because the retail clinic had to pay the physician to supervise the NP, and therefore raised the price of the clinic’s overall services, or whether NPs had to complete procedures as part of their collaborative agreements with physicians that ended up costing the retail clinic more. We don’t know exactly why it costs more, we just know it costs more.

Another study that we did was with the Office of the Assistant Secretary for Planning and Evaluation, which is part of the U.S. Department of Health and Human Services. The agency is known as ASPE, and they put out a request for proposal to study the impact of APRN scope of practice on access to care. And for that, some colleagues we had worked with before said that they were intending to put in a proposal, and they contacted me as I’ve worked with them on the National Sample Survey of Nurse Practitioners. At that point, I had already been contacted by David Auerbach at RAND, and from a few of my former colleagues at Georgetown University. So we all decided that, rather than compete with each other, we all put in a joint proposal to ASPE, and to work together. And that was really fun. We did analysis of Medicare claims data, we did some analyses of the National Sample Survey of Nurse Practitioners, and then we did site visits in four states about the practice environment for nurse practitioners, APRN scope of practice, and patient access to care issues. And there were a number of things we found.

We discovered that when scope of practice is less restricted, NPs are more likely to practice in rural areas, which was not surprising, but nevertheless was nice to document and to put out there. We also found from our interviews that there are a lot of other policies that can impede nurse practitioners from practicing to the full extent of their knowledge. For example, at the time of our study, NPs in several of the states we surveyed were not allowed to prescribe durable medical equipment (DME). So they couldn’t recommend a hearing aid or a walker or a wheelchair, for example, without consulting with and receiving approval from a doctor. That, I understand, has since been changed.

I also did the site visit in New Mexico, and I heard from the leadership and medical practitioners in New Mexico that NPs had issues prescribing medication-assisted therapy for opioid abuse. Nurse practitioners until just last year were not allowed to prescribe medication-assisted treatment, which was a huge problem in New Mexico, where opioid abuse had reached crisis levels. And these were federal policies–so even though we were visiting states with independence for NPs, such as New Mexico and Washington State, we were still hearing that these federal laws were in the way. I have since heard that NPs are now allowed to prescribe medication-assisted therapy, thanks to the Comprehensive Addiction Treatment and Recovery Act, which was passed last year.

We also found that the local culture and the culture of the medical establishment matter. There are a number of hospitals and medical centers that for no rational reason that anyone could really articulate to me, do not allow NPs to admit patients, or to round on patients. Furthermore, I had someone in New Mexico tell me that one medical facility allowed a physician position to stay open for five years, waiting for a doctor to fill it, rather than hire an NP to fill that need, because that was what people expected there–physician care. There were also rural communities we visited with medical facilities that would not let nurse midwives deliver babies there.

The results illustrated the fact that you can have independent scope of practice, but it doesn’t mean that federal and local community forces do not get in your way. Local organizations, stereotypes, and policies can inhibit NPs and physician assistants from doing what they need to do to meet the needs of their population. And one of the principal conclusions we drew from the findings described above was that you have to have the state laws in place, but that that’s not enough to guarantee that the population will have sufficient access to the care they need.

Two of the other studies I have done were literature reviews. One was looking at payment regulations around nurse practitioners. The Medicare regulation is that NPs get paid 85 percent of what a doctor gets paid. Why is that the case? We don’t know, we could not find any document or anyone we could interview who could tell us where that 85 percent number came from. One of the great mysteries of policy.

We also found that even in states with policies that aim to allow APRNs to practice to the full extent of their knowledge, insurance can pose another practice barrier for NPs. Medicaid bills differently for NPs and other APRNs versus physicians. And insurance plans are all over the map, whether they reimburse NPs and PAs directly or not. An APRN could technically be allowed to practice without physician assistance or oversight, but if you can’t bill independently, then that doesn’t really do you a lot of good. So that was another factor that illustrated how state practice laws are not the only thing NPs need to combat in order to truly practice autonomously.

[] In your own research and your review of the existing literature on APRN scope of practice, have you encountered any findings that have indicated a qualitative difference between the quality of care when it is administered by an APRN practicing independently versus a physician or a physician-supervised APRN?

[Dr. Joanne Spetz] I’ll start by asking a really narrow question: “Has any study shown that physician oversight of APRNs improves quality of care and yields better safety or quality?” No, not at all. There is no evidence that these scope of practice regulations improve quality or safety for patients.

Now, in terms of whether an NP can provide better quality care than a doctor, in general, the research would indicate that the quality of care is pretty equal. You can find different studies that have slightly different results, and the devil is always in the details, and this is the challenge with research, but I would say that on average, almost every study finds high quality of care, high adherence to clinical guidelines, high evidence of appropriate prescribing, and higher patient satisfaction with NPs relative to physicians.

A couple of studies seem to suggest that NPs might order diagnostic tests more often, although one of those studies lumped together NPs and PAs, and did not disentangle the fact that many (about two-thirds of) PAs work in specialty practice. For example, many PAs do more imaging because they work in orthopedics. I remember asking the people who published this study at a conference, “Have you analyzed the data by separating NPs from PAs?” and the guy said, “Oh yeah, and we found that more of the diagnostic testing was ordered by the PAs, but the paper had already been accepted when we did that analysis, so that is not included in the paper.” And I remember thinking, “The fact that two thirds of PAs are working in specialty care is not a trivial detail. If you’re a PA working in orthopedics, then of course you’re going to be ordering more diagnostic tests! You’re ordering X-rays all the time because your patients show up when they break a bone.” There are a few papers that try to argue or tell a story that NPs do not have the same expertise in their diagnostic skills and therefore they order more tests, but those papers have generally not been of a high enough quality to convince me.

As for studies that are being conducted to try and authoritatively determine whether health outcomes for patients who are treated primarily by an APRN are comparable or better than the health outcomes for patients who are under the care of a physician, there are a few. I believe Perri Morgan of Duke University has been doing some really neat research looking at VA databases, and investigating the quality of care given by PAs and NPs. And they have very good quality data. At present, they have done presentations at conferences, but their findings have not been published yet.

And one of my colleagues Ulrike Muench has been conducting a study where she looked at Medicare data, specifically at medication adherence, and she found similar rates of medication adherence with patients seen primarily by NPs versus those seen primarily by physicians. She also had a study where she looked at Health Care Costs Institute (HCCI) data, and she found that scope of practice regulations did not have any influence on medication adherence, which to her said we are trying to have these regulations, but we’re getting very equivalent outcomes regardless of scope of practice regulations. So there was no evidence that having these restrictions are beneficial.

One of the challenges in this field is that the data can be difficult to work with. The insurance claims data, such as the data that Ulrike used, generally don’t indicate whether it was an NP or a PA who saw the patient. And the same is true for Medicare data. For Medicare if you are billing for a service that an NP did, you can either bill it under the NP’s ID number or your can bill it under the physician’s ID number and say that what the NP did was under a doctor’s guidance of care.

Well, if you can bill 100 percent with a physician and only 85 percent with an NP or a PA, you’re going to bill it through the physician so you get your 100 percent. As a result, a lot of what NPs do gets hidden in the data. The Medicare data do not show all the care that NPs provide, because of the billing problem. There was one study however, that Jennifer Perloff did, along with a group of other researchers, and they looked at Medicare data and came up with a very complicated algorithm that in its essence was trying to do two things. Firstly, it aimed to determine when a patient was mostly managed by an NP versus when he or she was mostly managed by a physician. And based on that, it subsequently gave estimated cost trajectories over time for patients in both categories. And from what I remember of the study, what they found was cost savings and in some cases better preventative care when patients received care primarily from an NP.

Not only are the data and its analysis complicated, but the politics are also complicated. It’s very hard to find a physician who individually does not have a tremendous amount of respect for and confidence in the NPs they have worked with. In fact, the original vision for NPs was for them to manage patient care in rural areas where there were no doctors. The original vision was for them not to have supervising physicians dictating how they practiced. Many doctors have an incredible amount of trust for NPs, but for some reason, their professional organizations circle the wagons and block NPs. And I’m an economist, so I would say follow the money for the real reason. I think that there are a lot of issues around who is in professional power, who has control over the most financial resources in the health care system, etc. and so the politics and the turf protection end up coming ahead of ensuring that there are an adequate number of providers to allow patients to get the care they need. And that is really sad, that the politics play out that way.

We’ve made a lot of progress, but still have so much work left to do. For example, in California, these scope of practice issues just keep coming up, and I remember at one conference hearing the members of the California Medical Association saying, “If we gave NPs full practice authority in our state, they wouldn’t go practice in rural areas, they would go set up medi-spas in the suburbs.” And I thought, “Where are you getting that idea? There is no data to support that.”

So there is this layer of politics that really does not serve the best interest of our population, of our primary care needs for the future. And for the states that have modernized their scope of practice policies for APRNs, my hats off to them. Nevada went a long way, and while some states have had to make some compromises–for example, states like Kentucky still require NPs to work under a collaborative practice agreement for a few years before practicing independently–the trend is still net positive.

Thank you Dr. Spetz for your insight into scope of practice regulations and their impact on patients’ access to care!

About the Author: Kaitlin Louie is the Managing Editor of, and creates informational content that aims to assist students in making informed decisions about graduate programs. She earned her BA & MA in English from Stanford University.