Defining the Role and Value of Physicians Who Primarily Practice in Nursing Homes: Perspectives of Nursing Home Physicians

Abstract

Objective:

To identify the perceptions of physicians with expertise in nursing home care on the value of physicians who primarily practice in nursing homes, often referred to as “SNFists,” with the goal of enriching our understanding of specialization in nursing home care.

Design:

Qualitative analysis of semi-structured interviews.

Setting and Participants:

Virtual interviews conducted January 18–29, 2021. Participants included 35 physicians across the United States, who currently or previously served as medical directors or attending physicians in nursing homes.

Methods:

Interviews were conducted virtually on Zoom and professionally transcribed. Outcomes were themes resulting from thematic analysis.

Results:

Participants had a mean 19.5 [SD = 11.3] years of experience working in nursing homes; 17 (48.6%) were female; the most common medical specializations were geriatrics (18 [51.4%]), family medicine (8 [22.9%]), internal medicine (7 [20.0%]), physiatry (1 [2.9%]), and pulmonology (1 [2.9%]). Ten (28.6%) participants were SNFists. We identified six themes emphasized by participants: 1) An unclear definition and loose qualifications for SNFists may affect the quality of care; 2) Specific competencies are needed to be a “good SNFist;” 3) SNFists are distinguished by their unique practice approach and often provide services that are unbillable or under-reimbursed; 4) SNFists achieve better outcomes, but opinions varied on performance measures; 5) SNFists may contribute to discontinuity of care; 6) SNFists remained in NHs during the COVID-19 pandemic.

Conclusions and Implications:

There is a strong consensus among physicians with expertise in nursing home care that SNFists provide higher quality care for residents than other physicians. However, a uniform definition of a SNFist based on competencies in addition to standardized performance measures are needed. Unbillable and under-reimbursed services create disincentives to physicians becoming SNFists. Policymakers may consider modifying Medicare reimbursements to incentivize more physicians to specialize in nursing home care.

Brief summary:

•Experts in nursing home care believe that SNFists provide higher quality care for residents than other physicians, but a uniform definition of a SNFist based on competencies and standardized performance measures are needed.

Introduction

Nursing homes (NHs) provide care to 1.3 million residents¹ who include some of the most vulnerable patients in the nation’s health care system. Physician involvement is essential to the delivery of high-quality care for NH residents.^2–6 Within NHs, physicians are responsible for managing resident care plans, including therapeutic diets, rehabilitation services, medication prescription, and for diagnosing and treating new acute and chronic conditions. Additionally, residents and families expect to engage with the resident’s physician about care plans and prognosis.⁶

Although physicians play an essential role in the care of NH residents, relatively few see patients in NHs.⁷ However, the number of physicians who are “SNFists,” generally defined as physicians and advanced practitioners who primarily practice in NHs, increased 48% between 2007 and 2014.⁸ As of 2017, there were more than six thousand physician SNFists practicing in NHs in the United States.⁹ These clinicians are commonly defined based on the volume of services provided in the NH setting, similar to hospitalists.^8–12 There is not a formal medical specialization in NH medicine, but the term “SNFist” has been used by the Centers for Medicare and Medicaid Services.¹³

Despite the increasing prevalence of SNFists, little is known about them. Previous studies based on administrative data found SNFists to be associated with better outcomes, including fewer hospitalizations and improvement in process measures of care.^11,14,15 No studies have examined how SNFists are perceived by experts in NH care or whether they care for NH residents in ways that differ from other physicians. The objective of this study was to identify the perceptions of physicians with expertise in NH care on the value of SNFists with the goal of enriching our understanding of specialization in NH care. We conducted 35 semi-structured interviews of physicians across the US, including both SNFists and non-SNFists, and elicited their views on SNFists’ approach to providing care to NH residents, qualifications, outcomes, and how the COVID-19 pandemic affected the role of SNFists.

Methods

Setting and Study Population

We interviewed 35 physicians who were currently or previously medical directors or attending physicians in NHs over the period January 18–29, 2021. The sample was drawn from a list of physicians for an ongoing project on NH medical staff organization funded by the Health Resources and Services Administration that was composed of medical directors and attending physicians who are members of the American Medical Directors Association (AMDA).¹⁶ The list included email addresses of 421 physicians across the country. We used purposive sampling to select interviewees based on geographic location. An invitation letter was distributed via e-mail. Interviews were conducted virtually and participants provided verbal informed consent to record them and publish deidentified excerpts. Each participant received a $100 e-gift card. This study was approved by the institutional review board of [“Weill Cornell Medical College”].

Data Collection

Three researchers (H.J., M.U., and H.Y.) conducted the virtual interviews on Zoom (Zoom Video Communications) using a semi-structured interview guide. Interviews were confidential, lasted 30 to 60 minutes, and each included two or more interviewers. The interview guide was developed based on previous studies of physician care in NHs,^8,17,18 expert advice from members of the project team, and NH rounds and pilot interviews conducted with NH leaders in New York City.¹⁹ The guide included questions about contractual arrangements between NHs and interviewees, their clinical background, clinician staffing configurations in the NHs where they worked, and their perceptions on the value of SNFists (Table S1).

Data Analysis

Interviews were recorded and professionally transcribed. Data were analyzed using thematic analysis.²⁰ First, three investigators (H.J., H.Y., E.O.) reviewed the transcripts and developed a preliminary codebook. Subsequently, the three investigators reviewed each transcript and updated the preliminary codebook until reaching thematic saturation, which was achieved at the 33^rd interview. The final codebook included 39 unique codes (Table S2). Two investigators (H.Y., E.O.) coded three initial transcripts together to verify basic directions for the final codebook. The two investigators then separately coded four identical transcripts to review interrater reliability. A Krippendorff’s alpha of 0.81 was achieved, indicating a high level of agreement.²¹ The analysis was conducted using ATLAS.ti, version 8 (ATLAS.ti Scientific Software Development GmbH, Lietzenburger Straβe 75, 10719 Berlin, Germany). Investigators who did not engage in coding (H.J., M.U.) reviewed each version of the codebook and coded transcripts to strengthen the objectivity of the analysis and consolidated the codes into larger related groups. Three investigators (H.J., H.Y., E.O.) then discussed the groups to develop unifying themes inductively, iteratively reviewing the transcripts and resolving any disagreements. All investigators further refined the themes and developed the final thematic categories through consensus.

Results

Of the 35 interviewed physicians, 17 (48.6%) were female (Table 1). Medical specialties of participants included geriatrics (18 [51.4%]), family medicine (8 [22.9%]), internal medicine (7 [20.0%]), physiatry (1 [2.9%]), and pulmonology (1 [2.9%]). Thirteen (37.1%) participants had a certified medical directorship issued by AMDA; 26 (74.3%) currently or previously served as both medical directors and attending physicians in NHs. Ten (28.6%) participants self-identified as “SNFists.” Participants had a mean of 19.5 [SD =11.3] years of experience working in NHs and were currently working in a mean of 4.2 [SD = 6.8] NHs and other long-term care facilities. Fourteen (40.0%) participants were or had been employed by private third-party organizations contracting with NHs. Of these, 13 provided clinician staffing in NHs and one provided care management services. Twelve (34.3%) participants belonged to a private solo or group practice while practicing in NHs.

Table 1.

Characteristics of the 35 Study Participants

Physician Characteristics	No (%)
Physician specialty
Geriatrics	18 (51.4)
Family Medicine	8 (22.9)
Internal Medicine	7 (20.0)
Physical Medicine & Rehabilitation	1 (2.9)
Pulmonology	1 (2.9)
Certified Medical Directorship training	13 (37.1)
Past or current position in a NH a
Medical director only	3 (8.6)
Attending physician only	6(17.1)
Both medical director and attending physician	26 (74.3)
Mean years of experience in NH [SD]	19.5 [SD =11.3]
Mean number of facilities where participants currently work [SD] b	4.2 [SD = 6.8]
Employment type c
Private practice	12 (34.3)
Government	4 (11.4)
Academic institution (university)	4 (11.4)
Private organization contracting with a NH to provide care or care management services	14 (40.0)
NH	1 (2.9)
Female d	17 (48.6)
Location (based on state)
Northeast	7 (20.0)
Midwest	6(17.1)
West	11 (31.4)
South	11 (31.4)

Abbreviations: NH, Nursing home.

^a“Past or Current Position in a NH” represents the interviewee’s position when they were/are practicing in NH as a physician.

^b“Mean number of current facilities” is based on the number of NHs and other long-term care facilities (e.g. assisted living) that each interviewee is currently working in. Those who were not currently working in a facility were not included in calculating the mean.

^c“Employment type” is either the physician’s current or past employment status when they were interacting with NHs.

^dResponses for sex were subsequently categorized (as shown).

The analysis resulted in six themes (Figure).

Figure.

Key Themes of Participant Perceptions of SNFists and Summaries of Interview Responses

Abbreviations: NH, Nursing home.

Theme 1. An unclear definition and loose qualifications for SNFists may affect the quality of care.

The majority of participants believed that there is no clear definition of a SNFist, but acknowledged the increasing use of SNFists and that the term is widely used. Participants linked the unclear definition to loose qualifications required to be a SNFist and highlighted that a lack of formal requirements and pre-defined standards affect the quality of care.

P24: “We don’t really have good standards for what ... Basically at this point, you’re SNFists if you say you are.”

Ten participants who identified as SNFists had diverse opinions about defining SNFists. All stated that the commonly used definition based on billing for NH visits (e.g. ≥90% of a physician’s evaluation and management claims for NH visits) ^8,10 might not be appropriate. Several interviewees stated that a SNFist should be defined based on competency. They also noted that the definition should not be limited to the NH setting since physicians who frequently provide care in NHs may also provide care in assisted living facilities or continuing care retirement communities.

P31: “Well, SNFist is both, you know, nursing homes and assisted living.”

Theme 2. Specific competencies are needed to be a “good SNFist.”

A few participants suggested that specific components are needed to define competency for a “good SNFist,” including understanding the uniqueness of the NH setting through experience, knowledge of state and federal regulations, training in geriatrics, dedication, and the ability to practice in a much more resource-constrained environment compared to hospitals while now treating patients with higher levels of acuity than in the past.

P11: “ Understanding how the nursing staff works, the various changes - 1 think being on site and understanding that is huge. Also knowing the limitations of what you can, and don’t have available in your nursing home... I think it makes a big difference.”

P5: “I know federal regulations and state regulations.”

There was also a comment describing how the NH population has gradually transitioned to increasingly sicker patients necessitating a specific skillset for SNFists.

P17: “I think that the days of having a building that’s full of really stable long-term patients are done...We have seen the explosion of assisted living, [what] that’s done is really taken stable nursing home patients out of nursing homes...And so [we] now have really significantly complex and not always stable post-acute care patients that need to have a medical provider seeing them almost every day sometimes. And so for that reason, I think that the model of having a dedicated SNFist is at this point a more appropriate model for medical care.”

There were mixed views about hospitalists transitioning to become SNFists. Participants who had only worked as SNFists felt that experience as a hospitalist may not translate into being a good SNFist. Conversely, a few participants who were former hospitalists stated that the experience is beneficial to being a good SNFist.

P23: “In fact, one of the concerns is that hospitals and hospitalists have started, well I would say, encroaching on SNF care, and many of them are not adequately trained and expert in SNF care. And as a result, I feel that they often provide lower quality of care than a true medical provider who has been trained.”

P2: “Because a length of stay in the hospital is getting shorter and shorter, they come to our nursing home. They’re very sick, so much like a hospital patient. I was a hospitalist before. That skill, I think it’s really much needed.

Participants underscored that a physician’s dedication to NH patients distinguishes good SNFists from clinicians who are not devoted to a particular population.

P20: “I know people who are in private practice, community practice, and do nursing homes. Usually nursing home is kind of the last thing on the list during the day, they let their staff answer faxes or phone calls from the nursing homes...I just know that when I was there in community practice, I didn’t, that was not my priority. It was at the end of [the day] after seeing patients in the office or the hospital and then it was on the way home, you stopped at the nursing home...Now [as a SNFist] it’s my focus. It’s the first and only thing I do.”

Five participants noted concerns over for-profit physician-staffing companies that employ SNFists potentially pressuring their physicians to provide care in ways that reduce costs, but could negatively impact quality. However, in general, participants felt that commitment and care quality are not necessarily dependent on one’s employer.

P1: “The thinner you are spread, the less in-depth your services will be. So, my partner who goes to five facilities...his notes are extremely brief...I imagine [the staffing company] is for-profit, but they don’t make a big deal of it.”

Theme 3. SNFists are distinguished by their unique practice approach and often provide services that are unbillable or under-reimbursed.

Participants commonly pointed out “teamwork” as uniqueness in the way that SNFists practice.

P24: “SNFists tend to make much better use of the other disciplines that are available for you to work with...the nursing staff and nursing assistants as sources of information.”

Participants reported that SNFists fill a specific niche by demonstrating specialized skills for dementia patients, transitional care, institutional and palliative care, deprescribing, and reducing avoidable hospitalizations.

P23: “Delirium number one, infections in the nursing home, malnutrition, dehydration, physical debility, all of those are common complication of skilled nursing facility care. And I feel that the SNFists are best prepared to prevent those and to identify them...Particularly respectfulness with use of antipsychotic medicines, deprescribing, understanding delirium, incontinence, debility - all that’s our skillset.”

P3: “It is really a very specific body of knowledge transitioning people at the point of their lives...once you need any institutional care, I think you fall under a sense of future medical trajectories that are not familiar to people who like our outpatient or hospital…It’s a very specialized area of knowledge…I want to bring them on a gentle slide to whatever destiny they have, but I think that’s where our expertise comes in handy because we’re looking at all their diagnosis, all their social situations, and what their goals are for themselves, and that’s how we make decisions.”

Participants also pointed out that SNFists often have knowledge of residents’ social needs and the ability to communicate clearly with them and their families. These activities and other unbillable and under-reimbursed services were perceived as a large part of what makes practicing in the NH unique. However, unbillable and under-reimbursed services were cited as a disincentive to physicians specializing in NH care.

P3: “I feel like I do a lot of social work, even though we have wonderful social workers, but we have put in context for the patient and family because we may not be getting the best numbers for their blood pressure...I don’t see my patients as a list of diagnoses.”

P1: “So if you just go by billed claims, you might miss a large part of what the SNFists are doing...I consider the transitional phone call a really important part of the care, even though I can’t bill them.”

Theme 4. SNFists achieve better outcomes, but opinions varied on performance measures.

Participants argued that SNFists get better outcomes for residents compared to non-SNFists. The most consistent assertion was that SNFists often catch changes in a resident’s condition immediately and are able to intervene promptly. Opinions of appropriate outcome measures to examine the effectiveness of SNFists varied (Table 2), but many advocated for resident and family satisfaction with care as a performance measure.

Table 2.

Views of Experts on Appropriate Outcome Measures and their directions for SNFists effect.^a

Resident Population	Outcome	Impact of SNFists b
Short-stay population	30-day rehospitalization	Decrease
	SNF length of stay for predictable admits	Decrease
	Complication rates	Decrease
	Emergency Room visits	Decrease
	Polypharmacy	Decrease
	Spending on labs and X-rays	Decrease
	Care transitions	Improve
	Delays in screening patients upon admission	Decrease (more occur within 24 hours)
	Delay in initiation of medications	Decrease
	Functional improvement	Increase
	Successful discharge	Increase
	Advanced care planning	Increase
Long-stay population	Anti-psychotic use	Decrease
	Pain management	Increase
	New admissions to the hospital	Decrease
	Goals of care	Increase (more concordance with patients’ goals)
	Quality of palliative care (end-of-life care)	Increase
	Advanced care planning	Increase
	Timely enrollment in hospice	Increase
	Satisfaction of patients, families (customized care)	Increase
	Use of sliding scale insulin when possible	Limited impact
	Comorbidities	Decrease
	Pressure ulcers	Decrease
	Urinary tract infections	Decrease
Both populations	Infection rate	Decrease
	Wound occurrence	Decrease
	Falls	Decrease
	Morbidity	Decrease
Administrative outcomes	Staff turnover	Decrease
	Satisfaction of physicians, staff	Increase
	Minimum Data Set quality scores	Increase
	Five-Star rating	Increase
	Expenditures on medications and diagnostic tests	Decrease
	Participation rate of NH physicians in quality assurance meetings	Increase

SNF = skilled nursing facility

^aResponse populations for outcomes can overlap.

^bSNFists’ performance is based on responses of interviewees.

P29: “I know intimately about their health care needs, I know who they are,...I know he doesn’t look well, I know there’s something wrong...because I’m here full-time.”

P28: “I think that’s the most important thing to make sure that we’re connected with the family.”

Theme 5. SNFists may contribute to discontinuity of care.

Discontinuity of care was most frequently reported as participants’ biggest concern about SNFists. SNFists may replace residents’ primary care physicians, creating fragmentation in care. Additionally, isolation of SNFists was mentioned. For example, SNFists practice almost exclusively in NHs, may not have familiarity with specialists for referrals, and may not see patients before NH admission or after discharge, e.g. for post-acute patients.

P5: “I’m just another doctor who is now in this string of doctors that this patient is seeing. They have their primary doctor, now they have their hospitalist, multiple specialists, and now they have a SNFist.”

However, diverse ways to improve care coordination were reported.

P5: “So we spend a lot of time talking to the hospitalist for the transition of care. And then we communicate with their primary care physician to handover the care back to them.”

Theme 6. SNFists remained in NHs during the COVID-19 pandemic.

Participants who identified as SNFists noted that they took responsibility for NH residents during the COVID-19 pandemic because many physicians and hospitals refused to treat them.

P26: “SNFists’ve gotten hammered in this pandemic, because the hospitals don’t want our patients. They don’t want people with COVID-19.”

Due to NH visitor bans, SNFists were often the only physicians onsite and became decisionmakers as well as educators by taking on tasks such as providing guidance on quarantine procedures and reviews of policy and regulations for infection control.

P10: “And then now I’m running COVID units...I think people don’t realize that SNFists are frontline providers.”

Discussion

This is the first study to explore perceptions of SNFists among physicians with expertise in NH care. Fundamental issues regarding a lack of a standard definition and competencies for SNFists were raised by interviewees with the belief that they affect the quality of care. Participants generally agreed that the current volume-based definition of SNFists used in the literature should be adjusted to focus on competencies. Additionally, the manner in which SNFists practice was considered unique and thought to result in better outcomes than for non-SNFists, such as better infection control, deprescribing antipsychotics, and fewer hospitalizations. However, unbillable and under-reimbursed services were believed to create disincentives to physicians becoming SNFists. A consistently noted attribute of SNFists was their commitment to NH residents, as demonstrated during the COVID-19 pandemic.

Specialization is an important topic in medical care, and there is a body of literature on inpatient specialization, i.e. hospitalists.^22–24 NH care is a similar type of specialization. Practice specialization allows physicians to focus on areas of comparative advantage by developing clinical expertise and experience in a particular area of medical practice. Despite not being a formal medical specialty, SNFists’ “specialization” in NH care may increase their effectiveness through extensive knowledge of NH regulations and guidelines in addition to in-depth knowledge of care provided to clinically complex residents.^8,25 NH specialization may also enhance communication with direct care staff and administration and could help build relationships with residents and families. The benefits of NHs using SNFists may warrant a formal specialty based on care for NH patients, in the same manner that specialization as a hospitalist is based on the site of care. The need to link physician care and quality in NHs has been recognized by AMDA, which has developed a set of competencies specifically for attending physicians in NHs,²⁶ in addition to quality indicators.²⁷ Similarly defined competencies and performance measures are needed as a first step to more clearly define the link between SNFist care and outcomes.

Services that are unbillable or under-reimbursed were considered disincentives to physicians providing the best care possible for NH residents and may lead to fewer physicians becoming SNFists. This indicates the need to evaluate both the level of reimbursements and the types of services that are reimbursed by Medicare.

Our study has two limitations. First, our results are based on expert opinion and have not been validated with quantitative analyses. Second, interviews were limited to physicians and did not include other types of NH clinicians, such as nurse practitioners, or NH administrators. However, all of the participants had substantial expertise in NH care as a medical director or attending physician in a NH, or both, with an average of 19.5 years of experience in this setting, and included both SNFists and non-SNFists.

Conclusion and implication

Our findings suggest a strong consensus in favor of increased use of SNFists among physicians with expertise in NH care based on the belief that SNFists provide higher quality care for residents than other physicians. However, there is a need for a uniform definition of a SNFist based on standardized competencies and agreement on performance metrics. Additional research is needed to assess different models of care that incorporate SNFists to identify those that lead to higher quality NH care. Moreover, policymakers may consider reimbursing for services currently not covered by Medicare that are frequently performed by SNFists and increase the level of payments for services that are reimbursed to incentivize more physicians to specialize in NH care.

Supplementary Material

1

Supplement Table Sl. Semi-structured Interview Guide for Physicians

Supplement Table S2. Codebook