The historical model of neurology in the hospital

Is the traditional hospital model untenable in the modern era of neurologic care?

The definition of a typical or traditional model for inpatient neurologic care differs widely, but for the most part, there is variation around several themes.

A typical private practice model has usually meant caring for inpatients while working around a densely packed office schedule. For the most part, services are consultative in nature, although a few straightforward primary inpatients are not uncommon. Nurse practitioners or other affiliate providers are increasingly employed. Inpatient service is usually rotated among practice partners and frequently shared between multiple practices to lessen the burden of inpatient responsibilities on any single neurologist. This rotation can be weekly, daily, or even patient-by-patient.

A typical academic model has been driven by house staff. Primary inpatient services are common and often segregated by patient population, such as general neurology, vascular neurology, and epilepsy monitoring. An attending neurologist, one or two neurology residents, and a variety of fellows, medical residents, and medical students usually staff each service. In addition, there may be a similarly staffed consult service that covers the emergency department and inpatients on other services (medicine, surgery, etc.) and a neurology resident located in the hospital to cover nights.

These models have several significant deficiencies with regard to patient care, physician satisfaction, hospital efficiency, and neurology training. First, the inpatient knowledge gap recognized by increasingly specialized, almost exclusively outpatient providers, continues to widen. A corollary is that neurologists removed from their area of expertise while in the hospital lose the opportunity to practice more effectively in their preferred clinical setting. Thus, the quality of both inpatient and outpatient care suffers. Second, responsibilities for inpatient service are widely distributed among providers, with none taking ownership of any problems that may arise. Third, extensive time demands limit the hours available to spend on hospital floors. Fourth, the financial burden associated with caring for hospitalized patients can be overwhelming for both community and academic practices. Finally, all of these deficiencies can negatively impact training future neurologists.

Definition of neurohospitalist

In general, neurohospitalists define themselves as much by where they practice as what they practice. A generally accepted definition of neurohospitalist is “an inpatient neurological subspecialist with a minimum or absence of outpatient responsibilities.” A survey of self-reported neurohospitalists in 2010 indicated that about a quarter spent some time in outpatient clinic and about 20% had some non-clinical responsibilities, but that the vast majority of time was spent in the hospital directly caring for inpatients (16). An updated survey completed in 2018 indicated about 70% of time is spent directly caring for inpatients with the remainder spent on administration or teaching; outpatient care was rare (27). Almost 90% of neurohospitalists are fellowship-trained, with about 40% of that training in vascular neurology, down from over half being vascular trained in 2010. The specialty remains young with over 60% of survey respondents in 2018 being less than 10 years out of training (27).

Changing neurologic care in the hospital

Care models using neurohospitalists are spreading at an extremely rapid pace, and the number of neurohospitalists in the U.S. is growing just as rapidly. The change is similar to that in internal medicine 20 years ago, during which time hospital medicine has been the fastest growing medical subspecialty in history. Most data are somewhat subjective, but the trend is clear. For example, in 2006 there was an average of three job postings for neurohospitalists in the United States, which increased markedly to 12 in 2009. This ballooned to over 400 neurohospitalist positions posted on SimplyHired.com in mid-2015; it has declined a bit but has been steady at over 100 for the past decade.

The change is affecting both community and academic practices. In the 2010 and 2018 surveys, respondents were equally split between academic and private practice settings (16; 27). Two more surveys of academic neurology departments indicated that over 40% have an active neurohospitalist program, 60% will definitely have a neurohospitalist program within the next 5 years, and only 10% definitely do not expect to have a neurohospitalist program in 5 years (26; 34). Community practices are moving equally quickly, with several large exclusively neurohospitalist groups in practice throughout the country and many practices hiring neurohospitalists specifically to manage the inpatient facet of their patient care.

This evolution is also beginning to impact pediatrics as many child neurology practices and programs are starting to employ pediatric neurohospitalists to address increasing needs in pediatric inpatient care (04; 21).

Bridging gaps in inpatient neurologic care

As depicted earlier in this article, there are several deficiencies with inpatient neurologic care that have developed over the years, and the main way these deficiencies are being addressed is by a rapid transition to hospital-based neurologic care. The following section details several of the more common and problematic issues and discusses how a neurohospitalist care model may address them.

Knowledge gap and opportunity cost. Similar to hospitalists in internal medicine, pediatrics, and obstetrics-gynecology, a neurohospitalist’s expertise is developed and maintained by continuous exposure to the acute care setting. Honing inpatient skills allows neurohospitalists to immediately meet the needs of patients with acute neurologic disorders and efficiently navigate inpatient resources. This includes rapid and accurate localization and diagnosis, awareness of the effectiveness and cost of diagnostic tests, development and implementation of efficacious treatment protocols, and providing effective leadership of the neurology care team. Data supporting this theoretical benefit are starting to emerge in the form of decreased hospital lengths of stay, more efficient patient flow through the hospital for neurologic cases, and higher hospital satisfaction ratings for neurologic services (06; 05; 28; 11; 09; 24).

During the transition to an outpatient, subspecialty, or research career, most neurologists notice an atrophy of inpatient skills. As they spend more time away from the hospital, they become less comfortable caring for complex inpatients. This reluctance tends to create a vicious cycle that leads to even less time spent in the inpatient arena with a resultant accelerated erosion of inpatient skills.

The flip side to this issue is lost opportunity. Not only might a neurologist have to deal with inpatient issues for which they feel ill prepared, but those duties prevent them from attending to other tasks for which they are more suited. For example, a neuromuscular expert cannot be in clinic caring for patients with Alzheimer disease or in the electrodiagnostic lab performing nerve conduction studies if he is in the emergency room caring for an acute stroke patient. Both efforts are obviously important, but whereas the latter places a “square peg in a round hole,” the former maximizes utilization of expertise, time, equipment, and space. Such practice optimization will be important moving forward as the number of patients with neurologic disease continues to rise.

Time demands and schedule disruption. Splitting time between locations tends to be detrimental to both. For example, rushing to the emergency room to care for an emergent patient causes delays or cancellations for scheduled outpatients. This results in poor patient satisfaction, poor provider satisfaction, and lost revenue. Although in many cases a minor component of practice volume, inpatient consults cause disproportionate disruption to overall workflow due to their urgency and unpredictability.

Practice optimization with regard to time management is equally as important as optimization with regard to skill set, and determining ways to maintain productivity in the face of large fluctuations in workload is paramount for the clinical and financial success of a practice.

Ownership. The term “ownership” is used to convey the importance of having a vested interest in the outcome and function of a clinical service. There are many obvious hurdles to optimizing a service in which every provider has only a small level of responsibility and minor effort commitment. With regard to care inefficiencies, quality variability, poor communication, and personality conflicts, all of which arise daily in any clinical endeavor, human nature is to “just get through it” if one will be off service quickly only to return in the fairly distant future. Practically speaking, a neurologist who spends most of his or her time working in the hospital is motivated to make that time as effective, efficient, and enjoyable as possible. As such, neurohospitalists’ goals are aligned with the hospital and the rest of the care team, which is a powerful motivator for improvement.

Financial and staffing challenges. All of the gaps delineated above raise many challenges with regard to staffing and finances. Switching to a neurohospitalist care model can help address them, although it raises some new challenges of its own.

One benefit from a financial and staffing perspective is that a neurohospitalist care model consolidates inpatient effort under fewer providers. This change in mindset causes a budgetary transformation that is very obvious to administrators. Costs that were distributed across a budget and somewhat hidden are pulled out and consolidated into a line item. For example, in many academic practices, neurologists spend 10% effort in the hospital, but this effort is not accounted for in their clinical or the hospital budget; it is essentially service rendered to the hospital for free. It is obviously only a semantic difference, but the effect on the budget is real. When isolated as a specific expense, it becomes obvious to everyone that inpatient care is associated with real costs and provides an avenue for discussion as to how those costs can be met.

Part of this newly recognized cost is counterbalanced by more efficient inpatient revenue production by providers more experienced in the hospital setting. Rather than billing for a series of low-level follow-ups, neurohospitalists are more likely to bill appropriately for services rendered. Given the alignment of goals discussed earlier, neurohospitalists are also inclined to grow their service. At the author’s institution, inpatient revenue increased approximately 2.5-fold over four years during the transition to a neurohospitalist model as a result of a combination of increased patient encounters and increased complexity for each individual encounter (11).

Increased revenue is helpful, but in most cases, it is not sufficient to fund the financial gap associated with neurohospitalists. Fortunately, the system-wide benefits to effective neurologic care are increasingly being recognized by hospitals. Financially most important is support for higher-revenue care, such as neurologic and cardiac surgeries, but increased radiology usage, faster throughput through emergency rooms and intensive care units, and decreased overall length of stay are all contributors. This has led many hospitals to subsidize neurohospitalists by a mechanism similar to that employed for medicine hospitalists, and overall, it appears that this is essentially the only way to make inpatient neurologic care feasible in the long run no matter what care model is ultimately used.

The financial relationship between hospitals and neurohospitalists continues to evolve. Much like the prior problem of the costs for inpatient neurologic care being “hidden” in the budget, the benefit to the hospital is “hidden” within hospital metrics such as length of stay and utilization rate. This makes it very difficult to demonstrate benefit to hospital administrators and can lead to negative interactions if not managed proactively. Anecdotally, many neurohospitalists have only discovered their actual worth to a hospital when they have considered decreasing or eliminating services in the face of financial difficulties. One report indicates there are substantial financial advantages for both hospitals and neurology practices that employ a neurohospitalist model of care (11).

Overlap with other subspecialties

There is significant overlap between neurohospitalists and other neurologic subspecialties. Most obvious are stroke and neurocritical care, but epilepsy and essentially every other subspecialty overlap to some extent. As mentioned earlier, about half of neurohospitalists are vascular trained. Neurointensivist care could logically be considered a subset of neurohospitalists, but for clinical, practical, and historical purposes are a separate group of providers.

As discussed earlier, a neurohospitalist is defined primarily by where they practice rather than what they practice. One potential benefit to the more overarching term “neurohospitalist” is to raise awareness for hospital neurology needs and consolidate the voice with which hospital-based neurologists speak to their hospitals and the healthcare community at large.

Impact on neurology or neurohospitalist training

The neurohospitalist model is causing a fundamental change in inpatient neurology education. By virtue of focusing purely or primarily on inpatient care, hospitalists are more intimately familiar with inpatient care systems, as well as teaching challenges and opportunities in the inpatient setting. The educational benefits of neurohospitalists include greater expertise, greater availability, and more emphasis on quality and cost-effective care.

Several surveys indicate that the majority of academic neurology departments have an active neurohospitalist program in one form or another or will have one within the next few years (26; 27; 34). There are many theoretical advantages of this model for residency training, but whether or not they will actually be realized is not yet clear. Certainly, training programs must now demonstrate competency in their residents and not merely show available opportunities and exposure, and neurohospitalists are ideally situated to assist in that task by directly supervising and evaluating residents’ performance (23; 03; 30).

The internal medicine hospitalist literature has provided some evidence that house staff learned more effectively when on teams lead by hospitalists. In a study performed at UCSF, internal medicine hospitalists were rated more highly for teaching effectiveness, knowledge of relevant subject matter, discussion of pathophysiology, emphasis on cost-effectiveness, and provision of appropriate and effective feedback (12; 13). These findings were confirmed by a systematic review of available data on the effect of internal medicine hospitalist attending physicians on trainees’ educational experiences performed more than a decade after the introduction of this model (22). Although these data are from the internal medicine literature, these findings will likely hold true for education in neurology as well as the neurohospitalist model is implemented in different academic institutions.

Currently, there are only two studies that have evaluated the impact of neurohospitalist models on resident education. Both addressed night coverage, and both were very positive. The first transformed a poorly evaluated night float rotation into the most valuable rotation in the program (10). The second showed a marked improvement in nighttime resident education with the addition of neurohospitalist advanced practice providers (35).

Neurohospitalist fellowship training is offered at several institutions across the country, and there has been extensive early debate about whether or not fellowship accreditation ought to be pursued. On the one hand is the argument for standardization and legitimacy that accredited training would provide to a nascent subspecialty. On the other is the argument that there is no specific clinical need for additional training to become a neurohospitalist because most residency programs still provide intense inpatient experience. In addition, other established fellowships can provide training to fill in expertise or marketability gaps (vascular, electrophysiology, neurocritical care, etc.).

At this point, a neurohospitalist fellowship is usually pursued by neurologists seeking either an academic inpatient experience or additional supervision during their transition to independent inpatient neurologist, but this is beginning to change. The Neurohospitalist Society has started certifying fellowship programs across the country to ensure certain education and clinical standards are met for neurohospitalist fellows. This is a major step forward in advancing neurohospitalist training, especially as more residency graduates seek fellowship training to hone their inpatient skills as typical residency programs devote less and less time to inpatient training.

Potential downsides

There are several potential downsides to this type of care model. Some are practical, such as coverage concerns or loss of flexibility, and some are clinical, such as loss of understanding about the continuum and course of neurologic disease.

To address the clinical concern first, this care model introduces an inherent discontinuity of care impacting care of the individual patient and group of patients. From an individual standpoint, it is perhaps better to have a single physician care for them throughout the course of their illness, so as to minimize confusion, repetition, and missteps. Of course, this is a practical impossibility. Instead, one could argue that there has always been a level of discontinuity in modern medical care and that addressing the issue specifically with targeted communication between the inpatient and outpatient settings is a more effective way of overcoming the hurdle. As all specialties of medicine have evolved, this discontinuity has become more obvious and has spawned an entire area of study in effective transitions of care (32; 33). Several articles suggest this hurdle can be overcome by targeted discharge strategies implemented by neurohospitalists, including institution of specific discharge clinics (29; 31).

Answering the question from the standpoint of individual providers becomes more difficult. There is something to be said for understanding the entire course of the disease process in order to be able to provide the most competent and effective care to patients. Focus only on the inpatient care of neurologic disease will by definition decrease neurohospitalists’ understanding of both the patients and their disease. This is a problem that frequently arises in other specialties, for instance, an emergency room physician wanting to admit a patient with Parkinson disease who is delirious or falling frequently, whereas their outpatient neurologist recognizes these problems as usually best managed in the home setting. Much as the issue from the patient’s point of view, the “cat is out of the bag” insofar as the transition to neurohospitalist care is already underway. The only solution would seem to be developing effective communication strategies between inpatient and outpatient neurologists.

Staffing issues are also a potential problem. What happens if the neurohospitalist gets sick? Unlike the situation in an outpatient clinic, patients cannot be rescheduled. In addition, if there is no redundancy with inpatient providers, the discontinuity between inpatient and outpatient care may mean that there is no one with sufficient expertise to fill in the gap. This loss of flexibility creates a new staffing challenge because providers are no longer interchangeable. In this vein, it is helpful for practices to have some providers comfortable in each setting. In the author’s academic program, there are several younger faculty members who are not so far removed from residency as to feel a wide competency gap in inpatient neurology. These neurologists are encouraged to perform extra duty, such as telemedicine or weekend call, with some regularity so as to maintain a pool of providers who could be called on to provide inpatient coverage in an emergency. This was extremely important during 2020 and the COVID pandemic. This plan is feasible in a large academic practice, but collaboration between groups may be required in a smaller community setting. Regardless, plans for emergency coverage should be developed prior to the emergency arising.

The converse situation can also occur in that having a provider available at a moment’s notice for hospital coverage results, by default, in unproductive times if patient flow is slow. In fact, the balance between too much staffing and not enough is a difficult one to strike. This is a complicated issue somewhat beyond the scope of this article, but effective use of advanced practice providers and teleneurology are important strategies that are becoming more common (27). There have been several reports of the effective use of teleneurology by neurohospitalists for both initial and follow-up care for both stroke and non-stroke patients, particularly at centers with one main hospital and several satellites (20; 01; 19; 25; 36).