One of my New Year’s resolutions for 2014 for the blog, besides looking for talented bloggers to add to our pool of awesome bloggers, was to try to look at areas of science-based medicine that we don’t often cover (or haven’t covered before), such as the delivery of health care. Fear not, I’ll certainly do enough posts on the usual topics, but I am going to make a conscious effort to diversify a bit, if only for my own personal growth. Ironically enough, in the couple of months before the end of 2013, just such an issue came up in my state. Uncharacteristically (for SBM at least) the topic I’m going to take a look at has nothing to do with the infiltration of the pseudo-medicine known as “complementary and alternative medicine” (CAM) or “integrative medicine” into academia or CAM practitioners like naturopaths or chiropractors trying to lobby state legislatures for greater scope of practice to ply their pseudo-medicine on an unsuspecting public. It does, however, have to do with expanding the scope of practice of a group of medical professionals, and, unexpectedly and disappointingly, it’s a proposal that’s had considerable resistance from various physicians’ societies in the state. I’m referring to advanced practice nurses (APRNs), often referred to as nurse practitioners (NPs).
Before I go on, it’s necessary for me to point out my conflict of interest. No, I haven’t received funding from the all-powerful American Association of Nurse Practitioners (whose influence, actually, is dwarfed by state medical societies and various physician groups). I do, however, have a very personal relationship with a nurse practitioner, namely my wife. However, I would point out that she hasn’t been an NP that long, and I routinely worked with NPs collaboratively long before the idea of becoming a nurse or even an NP was even a germ of a thought in my wife’s brain. Make of that admission what you will as you read on.
Advanced Practice Nursing
Nurse practitioners are different from physicians in that, first of all, they are nurses. However, they are nurses who have undergone advanced training such that they are qualified to manage common medical problems within their scope of practice. Another way that they are different from physicians is that their scope of practice is generally defined by the state laws that regulate their practice. A physician, once licensed in a state, can practice virtually any kind of medicine legally. It is the professional societies, not state laws, that determine the specialization of physicians. In practice, of course, I as a breast surgeon/surgical oncologist, can’t practice urology or internal medicine, for example. My hospital privileges don’t include them, nor would insurance plans reimburse me for them, nor would malpractice plans insure me to practice these specialties. The law, however, would not stop me. That is how, for example, Stanislaw Burzynski gets away with billing himself as an oncologist, even though he has never completed an oncology fellowship or even an internal medicine residency. (I still can’t figure out how he used to manage to get insurance companies to reimburse him for his services.) I would, however, rapidly run into roadblocks actually practicing.
NPs, on the other hand, go into training with a specific scope of practice. My wife, for instance, is a Pediatric NP. In addition, there are Family NPs, Adult NPs, Geriatric NPs, Women’s Health Care NPs, Neonatal NPs, Acute Care NPs, Occupational Health NPs, Certified Nurse Midwives, and Certified Registered Nurse Anesthetists. As the AANP describes it:
NP students determine their patient populations at the time of entry to an NP program. Population focus from the beginning of educational preparation allows NP education to match the knowledge and skills to the needs of patients and to concentrate the program of academic and clinical education study on the patients for whom the NP will be caring. For example, consider a primary care Pediatric NP. The entire time in didactic and clinical education is dedicated to the issues related to the development and health care needs of the pediatric client. While medical students and residents spend time learning how to manage adult clients and complete surgery rotations, a primary care pediatric nurse practitioner student’s educational time is 100 percent concentrated on the clinical area where the NP clinician will actually be practicing.
Most state laws are specific about NP scope of practice. NPs who practice outside of their scope of practice can rapidly find their licenses in jeopardy. Confusing the issue is patchwork of laws in different states regulating NP scope of practice. These range from regulating APRNs as licensed independent practitioners, as New Hampshire does, to having in essence no scope of practice beyond that of a registered nurse. Let’s compare. In New Hampshire, APRNs have the authority to perform medical evaluation/management; basically, they can diagnose and prescribe treatments within their scope of practice, and there are no requirements for physician collaboration, direction, or supervision. In Michigan:
…there is no statutory or regulatory scope of practice for an NP. According to statute and regulation, NPs have the scope of practice of a registered nurse. Physicians can, by Michigan state law, delegate their authority to perform medical acts. So, because NPs in Michigan cannot perform medical acts without delegation, they cannot be considered licensed independent practitioners under The Joint Commission definition, but they would be licensed independent practitioners under the HRSA definition.
The AANP has a map of the US to show what sorts of practice APRNs have in each state, which range from full practice (like New Hampshire) to restricted practice (like Michigan) to something in between that the AANP calls “reduced practice,” like New Jersey, where I practiced for eight and a half years, in which their ability to prescribe depended on collaboration with an MD, but otherwise the NPs could practice nearly autonomously.
Now, you might ask, what does the HRSA have to do with anything? There’s still more confusion, because the federal government’s definition of an independent practitioner is different from that of many states. The US Health Resources and Services Administration (HRSA) also defines “independent contractor,” and, to confuse things still more, its definition conflicts with that of the Joint Commission:
Under The Joint Commission’s definition, in states where they are not required to be supervised or directed by physicians, NPs are licensed independent practitioners, but they are not in states where such supervision or direction is required by law. To HRSA, however, all NPs are licensed independent practitioners. The HRSA definition identifies as licensed independent practitioners NPs “or others permitted by law and the organization to provide services without direction or supervision.”
As the author of the above article drolly notes, if readers are confused, they are not alone. I don’t wish to dwell on this confusion other than to point out that it was Senate Bill 2 in Michigan that got me interested in this topic. The bill is designed to allow NPs to practice independently here (i.e. for Michigan to become like New Hampshire in that respect). More importantly and relevant to SBM, I also have to point out that the opposition to this law from medical professional societies was largely not science-based, and it was because of that opposition that the bill barely squeaked through the Senate only after being amended in ways that arguably watered it down and appears currently stalled in the House. Indeed, that is what disappointed me most of all. I don’t wish to dwell on the politics and health policy aspects of NP practice overmuch. Obviously more than just science determines law and policy needs, although it is not outside the range of SBM to point out that, with the projected shortage of primary care physicians over the next couple of decades, better utilizing providers like NPs to pick of the slack should be on the table as an option.
In which physicians protect their turf
An example of the sort of rhetoric being used against SB2 can be found on the Michigan State Medical Society website, there is a link to a pre-packaged letter to send to Representatives, as well as to a statement by the President of the MSMS Kenneth Elmassian, DO, that reads:
Every discussion in Lansing about health care and health policy should start and end with what is best for Michigan patients. The Michigan Senate today picked special interests over the health and safety of Michigan families when they voted to approve Senate Bill 2, which reduces educational requirements for those who practice medicine, risking patients’ lives and sending the signal to health care providers that medical education simply doesn’t matter.
Senate Bill 2 is bad medicine. Lawmakers in the state House should do what the Senate wouldn’t—reject this dangerous special interest legislation and instead put patients first.
On the MSMS Government Affairs YouTube channel are multiple short videos of MSMS members repeating the same dubious arguments:
The letter the MSMS is trying to get doctors to send to legislators, an effort that the Wayne County Medical Society is supporting by including a link to the form on its website (even going so far as to refer to SB2 as a “dangerous bill”), is no better, repeating the same point about fewer years of education, asking:
Consider this: if nurses were given this broad expansion of scope, what would happen if something went very wrong in the course of treatment–something that a nurse doesn’t have the education or training to handle? Are you willing to put patients in this precarious situation?
This is a transparently weak argument. To illustrate what I mean, let me ask: What happens when a physician encounters something in the course of diagnosis or treatment that goes very wrong and he doesn’t have the training to handle? He calls in other physicians who can handle it! Seriously, by this reasoning, no gastroenterologist should ever be allowed to do colonoscopies because he can’t repair a colon if he perforates one, and no cardiologist should be allowed to do angioplasties because he has to call in a heart surgeon to fix the problem with an emergency bypass if he messes up a coronary artery during a balloon angioplasty, a known risk of the procedure. The key is not being able to handle everything, as every physician specialist knows. The key is to be able to recognize when you’re in over your head and can’t handle a problem and not to be too proud or stubborn to call for help from someone who can handle it. You know who taught me that? Pretty much every surgeon I ever trained under. To quote Harry Callahan, “A man’s got to know his limitations.” This is true whether that person is a physician or an APRN, and APRN training pounds a knowledge of those limitations home.
I’m not likely to win friends among my peers by saying this, given that multiple Michigan medical societies oppose the bill, but, as a member of the MSMS myself, I do not support the stance of the MSMS, and I was particularly disturbed by the faulty reasoning and fear mongering being used to defeat this bill. Indeed, I’m actually rather embarrassed for Dr. Elmassian, who sounds more like a TV pundit or a politician running for office than a physician with that insulting bit about “special interests.” Seriously, I expect the President of my state medical society to make better arguments than that, even if I happen to disagree with his position. After all, I could equally argue that Dr. Elmassian is protecting physicians’ special interests against competition, which I rather suspect the MSMS is. Next, the whole argument about “reducing educational requirements” is disingenuous, particularly the nonsense about “sending the signal to health care providers that medical education simply doesn’t matter.” By that reasoning, I suppose the 17 states in which NPs can practice without physician supervision and the Institute of Medicine, which recommends that NPs be allowed to practice to the full extent of their training don’t care about medical education. The main reason that NPs don’t have as many educational requirements as physicians is because they specialize from the very beginning, unlike physicians, and they deal with a more limited scope of common problems. The AANP actually has a retort to this argument that I fully agree with:
Head-to-head comparison of educational models is not the appropriate measure of clinical success or patient safety. The appropriate measure is patient outcomes. Forty years of patient outcomes and clinical research demonstrates that nurse practitioners consistently provide high-quality and safe care.
This brings us to the real issue at hand that science can address and is thus inarguably part of science-based medicine and within the purview of this blog: Do NPs provide quality care? The AANP and Michigan Council of Nurse Practitioners argue that they do to the point that they should be considered independent practitioners. The MSMS and other medical societies argue that defining the scope of NPs would endanger patients. What does the evidence say? (Sorry about that link.)
The existing evidence base
Unfortunately for the MSMS, the evidence isn’t with it. In fact, I find it rather telling that none of the physicians’ groups arguing against laws expanding NP scope of practice seem able to cite any science. The reason, of course, is because they likely know that existing outcomes research looking at the effects of NPs on quality of care does not support their position. Let’s do a quick perusal of the literature, shall we? There are lots of studies; so I have to pick and choose, as well as take a look at a systematic review (non-Cochrane) and a Cochrane review.
One study, a chart review from 2008, compared the family practices in Pennsylvania and New Jersey to examine a single disease: Diabetes. Investigators audited 846 charts of patients with diabetes to compare adherence to American Diabetes Association guidelines for diabetes management between practices that employed NPs, physicians assistants (PAs), or neither. Practices with NPs performed better at providing some types of diabetes care, primarily monitoring tests, than physicians only or physicians with PAs, the latter two of which were statistically indistinguishable from each other. Whether there were confounding factors to account for the differences was not clear. These types of studies do exist in relative abundance. This is not new news, either. Copious evidence for the equivalence of care between NPs and physicians for common conditions that NPs are trained to manage exists dating back at least to the 1970s. For example, the Burlington Randomized Trial of the Nurse Practitioner was published in the New England Journal of Medicine in 1974. This study involved a large family practice in Burlington, Ontario:
…when two family physicians in Burlington, a middle-class suburban town of 85,000 just east of Hamilton, approached the Faculty of Medicine of McMaster University for possible help in introducing this innovation into their practice. For at least two years, their practice had been “saturated” — accepting no new patients or families because of inability to manage an increased case load. The physicians believed that their office nurses, with appropriate additional training, could assume a substantial portion of the responsibilities for primary care.
The NP training was as follows:
Before the study began, the nurses attended a special training program conducted by the schools of nursing and medicine at McMaster University. The emphasis of this program is on decision making and clinical judgment, rather than on procedural skills. The graduating nurse practitioners are qualified to become not physicians’ assistants, but co-practitioners, sharing the family physician’s responsibility for continuing care of patients. The nurse practitioner learns to evaluate each patient’s presenting problems, and to choose from three possible courses of action: providing specific treatment; providing reassurance alone, without specific treatment; or referring the patient to the associated family physician, to another clinician or to an appropriate service agency.
Patients were randomized either to NPs or one of the two family physicians at an allocation of 2:1 to doctors versus NPs, because at the time a case load half of that of a family physician was considered manageable for an NP. The resulting conventional group contained 1058 families (2796 members) equally divided between the two doctors, and the nurse-practitioner group comprised 540 families (1529 members), equally divided between the two nurse practitioners. Over the one year period of the trial, the number of deaths between the two groups was not statistically significantly different, nor was there a difference in physical status in terms of physical impairment, activities of daily living, or disability. The investigators noted a 5% decrease in gross practice revenue, but that was because the physicians were not billing for NP services. It was estimated that if the practice could have been reimbursed for their services, the increased volume of a 22% rise in the number of families under care could have produced a 9% increase in income.
Of course, these two studies are not examples of studies comparing physicians with NPs practicing independently. One of the earlier such studies I became aware of was a randomized study published in 2000 that randomly assigned 1,316 patients to nurse practitioners or physicians for primary care follow-up and ongoing care after an emergency department or urgent care visit. (Ironically, the study was published in that mouthpiece of the American Medical Association, JAMA, and is available in full text for everyone, no pay wall.) The outcomes compared included patient satisfaction after initial appointment (based on 15-item questionnaire); health status (Medical Outcomes Study Short-Form 36), satisfaction, and physiologic test results 6 months later; and service utilization (obtained from computer records) for 1 year after initial appointment, compared by type of provider. There was no difference in any of the parameters between physicians and NPs, other than that patients with hypertension were found to have a barely statistically significantly lower diastolic pressure at six months when treated by NPs and that patients rated provider attribute scores slightly higher for physicians. The study had limitations, such as being primarily a Medicaid population and thus not necessarily generalizable to an overall patient population and only following patients for a year, but overall it strongly suggested equivalent short-term outcomes.
There are multiple other studies. For instance, a randomized study from 2004 similar to this one found no differences between the groups in health status, disease-specific physiologic measures, satisfaction or use of specialist, emergency room or inpatient service. However, given that there are enough of these studies out there to result in systematic reviews, let’s check out the systematic reviews. For example, by 2005 there were enough studies for a Cochrane systematic review on the topic. Cochrane concluded:
The findings suggest that appropriately trained nurses can produce as high quality care as primary care doctors and achieve as good health outcomes for patients. However, this conclusion should be viewed with caution given that only one study was powered to assess equivalence of care, many studies had methodological limitations, and patient follow-up was generally 12 months or less.
A more recent systematic review from 2011 by Newhouse et al. examined the published literature between 1990 and 2008. Authors included randomized controlled trials or observational studies of at least two groups of providers (e.g., APRN working alone or in a team compared to other individual providers working alone or in teams without an APRN), conducted in the United States between 1990 and 2008, and reported quantitative data on patient outcomes. The review started at 1990 because practice and interventions have changed since before then, and studies in which health outcome effects could not be isolated to the involvement of NPs in patient care were also excluded. They found 69 studies (20 RCTs and 49 observational studies) that met these criteria, of which 37 were about NPs.
Among the conclusions:
- This systematic review of published literature between 1990 and 2008 on care provided by APRNs indicates patient outcomes of care provided by nurse practitioners and certified nurse midwives in collaboration with physicians are similar to and in some ways better than care provided by physicians alone for the populations and in the settings included.
- Use of clinical nurse specialists in acute care settings can reduce length of stay and cost of care for hospitalized patients.
- The results indicate APRNs provide effective and high-quality patient care, have an important role in improving the quality of patient care in the United States, and could help to address concerns about whether care provided by APRNs can safely augment the physician supply to support reform efforts aimed at expanding access to care.
Although there were limitations in this study, including heterogeneity of studies, limited number of randomized designs, often inadequate descriptions of NP versus physician roles for purposes of the studies examined, and the difficulty in attributing to the NP specific outcomes, at the very least we can say that this review of the literature does not support the contention that expanding the scope of NP practice is likely to result in decreased quality of care. Taken in its totality, the medical literature on the subject does not support the fear mongering about SB2 in which the MSMS has been engaging. It’s shameful. There might be political or economic reasons to oppose the specifics of the bill (although, if there are, I haven’t yet been convinced of any of them), but there are no scientific reasons to oppose it on the basis of patient safety and quality of care. The MSMS and the rest of the Michigan medical societies who make this argument are, quite simply, wrong. The scientific literature does not support them, and I rather suspect that they know it. If they had any outcomes data to support their fear mongering, they would have cited it. They don’t, because there isn’t any. Even the Institute of Medicine says so, and I bet any of my colleagues who oppose SB2 can’t prove me wrong.
I support increasing the scope of practice of APRNs/NPs commensurate with their education and training. Existing science and my own personal experience that began when I first started working with NPs in 1999 lead me to that conclusion. If there were strong arguments against this from a patient safety standpoint, believe me, I would have grave doubts. (After all, I am a physician, and I recognize that my inherent bias would almost certainly be that physicians provide better care, making me more inclined to take such arguments seriously if they were evidence based.) There aren’t, at least none that are scientifically supported by outcomes data, which is why the reaction of my fellow physicians to such measures, which occurs in every state where such bills are introduced, saddens me. It’s pure turf protection, nothing more. My recommendation to my state medical societies would be to spend less time trying to shut out APRNs and more time trying to prevent naturopaths from being licensed in Michigan. That would prevent far more harm to patients than the worst fears the MSMS can conjure up about expanding the scope of practice of APRNs.