for nurse practitioners and physician assistants, also referred to as advanced practice providers

or mid-levels.

Debates regarding this issue often become echo chambers with both sides repeating what

others have said before them.

But what does the body of scientific evidence actually say about mid-level encroachment?

Let’s find out.

Dr. Jubbal, MedSchoolInsiders.com

Welcome to another episode of Research Explained, where we deep dive into a topic, spend countless

hours scouring the scientific literature, and summarize it so that you don’t have

to.

We’ve covered several other topics on our Research Explained playlist - link in the

description.

Here’s what the scientific literature has to say about independent practice for nurse

practitioners and physician assistants, and how their care compares to that of physicians.

There are several arguments in support of independent practice for midlevel providers.

The first is that PAs and NPs have sufficient training to treat patients independently and

without the need for a supervising physician.

To become an NP, one must hold a bachelor’s degree in nursing, be licensed as a registered

nurse, graduate from a nationally accredited graduate NP program and pass a national NP

board certification exam.

Similarly, physician assistants must complete a bachelor’s degree, complete a nationally

accredited physician assistant program, meet national standards, and pass a board certification

exam.

In addition, many states that currently allow for independent practice require a certain

number of hours or number of years working underneath a supervising physician before

APPs can practice independently.

The American Association of Nurse Practitioners argues that “a head-to-head comparison of

educational models is not the appropriate measure of clinical success or patient safety.”

Nurse Practitioner education is competency-based, not time-based, meaning that NP students don’t

progress or graduate based on number of hours spent in a rotation or by the number of times

they’ve seen a particular ailment, rather they do so when knowledge and skill competency

are achieved.

Some argue that mid-levels also have experience before graduate level training as a PA or

NP which often includes physical assessment skills, interpreting diagnostic test results,

evaluating the appropriateness of medications, and evaluating patients’ response to treatments.

There are also a number of studies comparing outcomes between physicians and midlevel providers

in support of independent practice.

A 2018 study examined the relationship between primary care provider type and diabetes outcomes

among patients and found no clinically significant differences between the three provider types

in terms of diabetes outcomes.

The authors suggest that similar chronic illness outcomes may be achieved by physicians, NPs

and PAs.

In addition, a 2021 meta-analysis of 39 different studies found that the quality of care delivered

by the PA was comparable to a physician’s in 15 studies and exceeded that of a physician

in 18 studies.

Another argument in favor of independent practice is that it improves access to care.

As it stands in the United States, we have a shortage of physicians - specifically in

primary care specialties.

According to the AAMC, we can expect shortages of between approximately 38,000 and 124,000

physicians by the year 2034.

By granting independent practice to PAs and NPs, some argue that we can help offset the

growing demand for physicians, especially in underserved areas.

A 2016 study compared geographic accessibility of primary care clinicians between states

with more restrictive and less restrictive scope-of-practice laws.

They found that access to primary care nurse practitioners was highest in rural areas and

that less-restrictive scope-of-practice states had as much as 40% more primary care nurse

practitioners compared to more restrictive states.

They concluded that removing restrictive scope-of-practice laws may help to expand the overall capacity

of the primary care workforce.

Lastly, proponents of independent practice argue that PAs and NPs can help decrease healthcare

costs.

In the same 2021 meta-analysis, the authors found that in 29 out of 39 studies, the labor

and resource costs were lower when the PA delivered care compared to when the physician

delivered care.

This makes sense as PAs and NPs make substantially less than physicians.

According to the Bureau of Labor Statistics, the average PA makes approximately $122,000

per year and the average NP makes approximately $118,000 per year.

The average physician, by comparison, makes around $208,000 per year according to the

BLS.

As a result, it costs hospitals far less to employ APPs than it does to employ physicians.

In addition, a 2021 study found that the average primary care physician’s cost of care is

34% higher than primary care nurse practitioners in low-risk patients, 28% higher in medium-risk

patients, and 21% higher in high-risk patients.

They conclude that these differences mostly reflect the lower quantity of services provided

by primary care NPs relative to primary care MDs which is reflected most in low-risk populations.

Now let’s talk about the arguments against independent practice.

The primary argument against independent practice is that mid-level providers do not have sufficient

training to see patients without a supervising physician.

The average family medicine physician fresh out of residency will have over 20,000 hours

of graduate-level training including over 15,000 hours of clinical experience.

If we compare this to the average nurse practitioner at the point of certification, they will have

received anywhere from 3,000 to 5,500 hours of graduate-level training and only 500 to

1,500 hours of clinical experience.

If we factor in the 2,000-4,000 hours of practice underneath a supervising physician that many

states require for independent practice, NPs still have less than half of the clinical

hours of a newly-licensed physician.

This is also assuming that clinical practice as an NP under the supervision of a physician

is equal to that of a resident physician being trained to practice independently - which

I would argue it is not.

Physicians are trained with the goal of independent practice from day one.

They go into their training knowing that they’ll be on their own one day and won’t be able

to rely on anyone else for guidance.

They need to be confident in their knowledge and skills in order to make the right decision

as they’ll ultimately have to live with the consequences of those decisions.

In contrast, PAs and NPs are not typically trained to practice independently.

Although they still treat patients and make decisions on a daily basis, they are doing

so with the knowledge that they have a supervising physician to fall back on for guidance.

As such, there’s a different level of responsibility and accountability that is put on doctors

during their training that you don’t get with PAs or NPs who are trained to work in

a more collaborative manner.

There is also a great deal of variability in the training of midlevel providers, especially

among nurse practitioners.

There have been dramatic increases in the number of nurse practitioner programs over

the last decade with many programs promising quick certification and high acceptance rates.

The average acceptance rate for NP programs is estimated to be around 66%.

However, there are multiple NP programs in the US with 100% acceptance rates.

By comparison, the acceptance rate for the average medical school is only around 6.5%.

These high acceptance rates have brought into question whether the goal of these NP programs

is to produce high-quality providers or to make money.

There is also a great deal of variability in the clinical experience between different

NP schools.

Many schools are 100% online and do not organize the clinical hours required as a part of their

curriculum.

Instead, it is up to the student to arrange for their own clinical experiences.

As a result, it is difficult to ensure consistent, high-quality training among NPs in these programs.

Those that oppose independent practice argue that his disparity in training leads to issues

of patient safety, especially in primary care – the field with the biggest push for independent

practice for mid-level providers.

Contrary to popular belief, primary care is one of the medical specialties that requires

the broadest knowledge.

Whereas in other specialties, your knowledge becomes increasingly specialized as you progress

through training, primary care physicians continue to use and develop the vast information

they learned during medical school.

There are a number of research papers supporting this difference in quality of care between

physicians and midlevel providers as well.

A 2018 study found that “compared to dermatologists, PAs performed more skin biopsies per case

of skin cancer and they diagnosed fewer melanomas in situ, suggesting the diagnostic accuracy

of PAs may be lower than that of dermatologists.”

Studies from 2005 and 2016 also demonstrated that NPs and PAs were more likely to inappropriately

prescribe antibiotics than residents and attending physicians, which can contribute to population

level issues such as antibiotic resistance and the creation of superbugs.

In addition, many of the studies that show that NPs and PAs deliver similar quality of

care as physicians, including the studies from earlier in the video, do not assess PAs

and NPs working independently, but rather those working as part of a healthcare team.

As such, it’s a big stretch to draw conclusions about independent practice when you’re looking

at studies with non-independently practicing mid-levels.

Another argument is that independent practice increases healthcare costs despite APPs lower

cost for services.

Many physicians argue that mid-level providers order more unnecessary tests and have to refer

patients out to other physicians more frequently.

An issue that may have been handled by a primary care physician may now have to be referred

out leading to two separate visits instead of just one.

A 2013 article found that the “quality of referrals to an academic medical center were

higher for physicians than for mid-level providers regarding the clarity of the referral question,

understanding of the pathophysiology, and adequate pre-referral evaluation and documentation.”

Referrals from physicians were also less likely to be evaluated as “unnecessary.”

In terms of cost, a 2015 study also showed that mid-levels are associated with ordering

more imaging services than PCPs for similar patients.

While these increases were modest for individual patients, the authors conclude that these

increases may be problematic for patient care and overall costs at the population level.

A recent study published in 2022 in the Journal of the Mississippi State Medical Association

summarizes all of these points nicely.

The study compared physicians and independently practicing midlevels in terms of healthcare

costs, patient outcomes, and patient satisfaction.

They collected data over 10 years from over 300 physicians, 150 APPs, 200,000 patient

surveys, and 33,000 unique Medicare beneficiaries.

Here’s what they found.

To start, healthcare costs for Medicare patients were $43 higher per month for patients whose

primary care provider was a mid-level instead of a doctor.

They estimated that this would equate to roughly 10.3 million dollars per year in increased

spending if all patients in their clinic were seen by APPs instead of physicians.

When they adjusted these findings for patient complexity, the difference was $119 per patient

or 28.5 million dollars annually.

They found that these additional costs had to do with several factors, including increased

ordering of tests and images, more referrals to specialists, and higher emergency department

utilization compared to patients under the care of a physician.

In terms of patient outcomes, physicians outperformed midlevels on nine out of ten quality metrics

including cancer screenings and management of chronic diseases such as high blood pressure

and diabetes.

Physicians were also found to have higher patient satisfaction scores compared to midlevels.

The authors also tracked outcomes and cost data from patients who were co-managed by

a physician and an APP and found that patients who alternated visits with the physician and

the APP had the best quality and cost outcomes of all.

They concluded that, although APPs are an invaluable part of the healthcare team, they

are best utilized when they are co-managing patients alongside physicians as opposed to

practicing independently.

The Hattiesburg Clinic has since redesigned its care model so that a doctor is the primary

care physician for all patients and no one sees a nonphysician exclusively.

Although this is just one clinic, in one state, this is one of the first studies that has

collected robust data comparing healthcare costs, patient outcomes, and other metrics

between physicians and independently practicing mid-level providers.

Many previous studies, including those mentioned in support of independent practice at the

beginning of the video, have only compared outcomes between physicians and APPs when

functioning in a collaborative role with physicians.

Although NPs and PAs are important members of the healthcare team, the solution to the

physician shortage is not to expand their scope and grant them independent practice.

The bigger issues that need to be addressed are the limitations preventing us from training

more doctors.

There is no shortage of people interested in becoming physicians.

We can see this in the record number of applicants applying to medical schools this past cycle.

The issue is that there is a bottleneck in the number of residency spots due to the lack

of Medicare funding for new residency programs.

According to the NBME, in 2022 there were approximately 43,000 medical students that

applied for first-year resident positions, out of which only 34,000 matched.

That means that approximately 9,000 medical students will either have to apply for the

Supplemental Offer and Acceptance Program and compete for one of the roughly 2,000 residency

spots that went unfilled, or spend a year strengthening their application so they can

reapply next year.

Regardless, roughly 7,000 students who have completed medical school will not be able

to progress to the next step of physician training.

As such, medical schools cannot continue to expand class sizes as there aren’t enough