Variability in skilled nursing facility screening and admission processes: implications for value-based purchasing

Abstract

Background:

Hospitalized older adults are increasingly admitted to skilled nursing facilities (SNFs) for post-hospital care. However, little is known about how SNFs screen and evaluate potential new admissions. In an era of increasing emphasis on post-acute care outcomes, these processes may represent an important target for interventions to improve the value of SNF care.

Purpose:

The aim of this study was to understand: (1) How SNF clinicians evaluate hospitalized older adults and make decisions to admit patients to a SNF; and (2) the limitations and benefits of current practices in the context of value-based payment reforms.

Methods:

We used semi-structured interviews to understand the perspective of 18 clinicians at three unique SNFs– including physicians, nurses, therapists, and liaisons. All transcripts were analyzed using a general inductive theme-based approach.

Results:

We found that the screening and admission processes varied by SNF and that variability was influenced by three key external pressures: (1) inconsistent and inadequate transfer of medical documentation; (2) lack of understanding among hospital staff of SNF processes and capabilities; and (3) hospital payment models that encouraged hospitals to discharge patients rapidly. Responses to these pressures varied across SNFs. For example, screening and evaluation processes to respond to these pressures included gaining access to electronic medical records, providing inpatient physician consultations prior to SNF acceptance and turning away more complex patients for those perceived to be more straightforward rehabilitation patients.

Conclusions:

We found facility behavior was driven by internal and external factors with implications for equitable access to care in the era of value-based purchasing.

Practice Implications:

SNFs can most effectively respond to these pressures by increasing their agency within hospital-SNF relationships and prioritizing more careful patient screening to match patient needs and facility capabilities.

Introduction

Approximately one in four Medicare beneficiaries are admitted to a skilled nursing facility (SNF) following an inpatient hospital stay (Mor, Intrator, Feng, & Grabowski, 2010). The high-risk transition between hospitals and SNFs is frequently associated with adverse outcomes (Burke et al., 2016). SNFs that receive Medicare payments have started to participate in value-based purchasing (VBP), which places them at financial risk for patient outcomes such as hospital readmissions and community discharge rates (Burke, Cumbler, Coleman, & Levy, 2017). These efforts are intended to improve the value of post-acute care, including SNF care (Ouslander et al., 2016; Burke et al, 2017,” 2014; Sood, Huckfeldt, Escarce, Grabowski, & Newhouse, 2011).

The increased emphasis on patient outcomes related to VBP places new pressure on the traditional processes SNFs use to evaluate patient appropriateness for transfer to SNF (Jones et al., 2017). SNFs that traditionally had financial incentives to admit every eligible patient for skilled care now face more complex financial incentives tied to patient outcomes. Additionally, hospital-based clinicians are unsure how to maximize the “fit” between a patient’s needs, risk for adverse outcomes, and goals with the resources of specific post-acute care providers (Burke et al., 2017). As the acuity of patients referred to SNF increases (Burke et al., 2015b, 2015a) without commensurate increases in skilled staffing (Rau, 2018), even the best efforts of SNFs to reduce readmissions and promote community discharges may be thwarted in the absence of robust processes for ensuring “fit” between patient needs, risks, and goals, and SNF resources.

It’s surprising given this context that very little is known about SNF clinicians’ perspectives on the process for evaluating and selecting patients for SNF (Baughman et al., 2018; Newman & Toseland, 1985; Toseland & Newman, 1982). While the hospital-based perspective on transitions to SNF (Burke et al., 2017; Jeffs, Lyons, Merkley, & Bell, 2013), and SNF clinician perspective on readmissions to hospitals from SNF have been more thoroughly described (Ouslander et al., 2016), prior research fails to capture the perspectives and roles of SNF clinicians in screening and admitting patients, which is a critical step for safe transitions of care (Burke, Guo, Prochazka, & Misky, 2014; Burke, Kripalani, Vasilevskis, & Schnipper, 2013).

We sought to understand how SNF clinicians evaluate hospitalized older adults and make decisions to admit patients to a SNF, and to place these decisions in the context of VBP. The goal of this research is to help inform clinicians, leaders, and policymakers about opportunities to improve patient outcomes within this new policy landscape.

Theoretical approach

We used two conceptual frameworks to inform our inquiry: the Ideal Transition of Care (ITC) framework (Burke et al., 2013) and the principles of social constructivism (Berger & Luckmann, 2011; Creswell, 2013). The ITC framework identifies 10 key domains to ensure safe transitions of care, and provides both an opportunity to be comprehensive in assessing care transitions, and to be specific within domains about key factors associated with improved transitions. Inclusion of more domains of the ITC framework is associated with increased likelihood of significantly reducing hospital readmission rates (Burke, Guo, Prochazka, & Misky, 2014), and the ITC framework has been used to inform large prospective cohorts (Meyers et al., 2014), identify preventable readmissions (Auerbach et al., 2016; Herzig et al., 2016) and inform program development (Leonard et al., 2017). This analysis reflects the responses of SNF clinicians across all the domains of the ITC framework, focused on the question, “How do SNF clinicians describe patient selection and the process for caring for patients once in SNF?” The domains of the ITC likely most applicable to this question include symptom monitoring and management, discharge planning, complete communication of information, and coordinating care among team members.

A social constructivist perspective to knowledge generation in qualitative inquiry is one that is ontologically relativist, epistemologically subjectivist, and methodologically hermeneutic and dialectic. People understand their experiences through relationships with others and the context in which they occur. A socially constructed reality is interpreted through language and past experiences. Individuals make sense or meaning from these experiences and express them through culture and language (Berger & Luckmann, 2011; Creswell, 2013). Using a social constructivist perspective, we attended to individual and the participant group’s perspectives as well as meaning within a community. In this study, the SNF is not only a health care location but also the context (or particular situated-ness) in which participants encounter group and system level messaging and language around health status and system capacity. Social constructivism informs how language, culture and interrelationships between individuals, system structures influence clinical practice norms (Burr, 2003). Participant experience shared through language provides a window into potentially modifiable decision factors to reduce readmission risk from SNF and improve safe systems of post-acute care transition. Therefore, we aimed to integrate current descriptions of the practices across a variety of practice contexts (SNFs and roles) with how they might fit with forthcoming reforms.

Methods

We conducted this research as a part of a larger qualitative study exploring how hospital and SNF patients and clinicians make decisions to utilize institutional post-acute care following an inpatient hospital stay. All data was collected prior to implementation of changes in SNF payment related to VBP reforms. The study was approved by the Colorado Multiple Institutional Review Board.

Sample

We used a purposive sampling strategy to recruit a diverse group of SNF clinicians from three unique SNF facilities. These facilities were identified as those that receive the most patients from three diverse hospitals: a VA hospital, a community hospital serving a primarily indigent and immigrant population, and a large teaching hospital. The three participating SNFs included a VA Community Living Center that provided short-term rehabilitation only; a predominantly long-term, Medicaid-funded nursing home with a larger Medicare-certified rehab unit; and a community SNF that only provided short-term rehabilitation under Medicare. All were located in a single metropolitan area but varied in size, ownership, proportion of short and long-stay residents, and referring hospitals (as above). One of the facilities was located within walking distance to a referring hospital. Both community SNFs had an overall nursing home compare rating of five stars (“Medicare.gov Nursing Home Compare”). Participants gave verbal consent to participate, as the requirement for written informed consent was waived. The SNFs are referred to as random numbers in the results to preserve confidentiality.

Interviews

We conducted semi-structured in-depth private, individual interviews with 18 SNF clinicians between February and September of 2016. Clinicians included medical directors, directors of nursing, bedside nurses, physical and occupational therapists, and social workers. Our semi-structured interview guide was developed around key ITC domains to elicit detailed information on SNF clinicians’ experience with care of older adults transitioning to SNF following an inpatient hospital stay. Other areas were probed iteratively across the participant interviews as new issues or examples were raised. Interviews were audio-recorded with the participant’s consent, professionally transcribed verbatim and validated. Participant demographics were collected using a brief questionnaire before the interview.

Data Analysis

We used our well-established team-based general inductive approach to theme analysis of the transcripts focused on the question, “How do SNF clinicians describe patient selection and the process for caring for patients once in SNF?” (Burke, Jones, et al., 2018; Burke et al., 2017; J. Jones et al., 2017) In the analysis process, after identifying relevant SNF practices, we also sought to answer the question, “How might these processes be rewarded or penalized in the context of VBP?” This entailed a detailed reading of transcripts to derive patterns or themes within the data as it relates to the conceptual models, research questions, and the interpretation of data by the research team. We developed the initial code list through group discussion, a priori ITC domains, and additional codes were added after a thorough reading of the transcripts by individual team members. As new codes emerged, we discussed them with the team to reach consensus on code labels and definitions. Data collection and coding occurred simultaneously. Themes related to this question were derived for each SNF site and compared across sites. Our team met weekly for analysis discussions and key decisions were documented as part of our audit trail. Data was managed and analyzed in Atlas.Ti (v7.5.11; Scientific Software Development, Berlin, Germany).

Results

Our analysis revealed two central themes about how clinicians evaluate hospitalized older adults for admission to SNF, and how these processes may be suited to VBP: (1) screening and admission processes varied by SNF and (2) external factors stress SNF screening processes and influence variability. We describe each of these themes in more detail, including how these themes interrelate. Table 1 presents the characteristics of interview participants.

Table 1:

SNF Clinician Demographics & Characteristics

Characteristic	SNF Clinicians N=18 (%)	SNF 1 Clinicians, n	SNF 2 Clinicians, n	SNF 3 Clinicians, n
Race/ethnicity	15 (83)
White/Caucasian	2 (11)
Black/African American
Asian	14 (77)
Native American	1 (5)
Educational level attained, n, (%)
High school/GED	1 (5)
Some college	2 (11)
College graduate	2 (11)
Post-graduate	13 (72)
Clinician Role, n, (%)
MD/Medical Director	3 (16)	1	1	1
Nurse	4 (22)	2	1	1
Physical Therapy	3 (16)	2	-	1
Occupational Therapy	2 (11)	1	-	1
Social Worker	2 (11)	1	1	-
Administration	4 (22)	1	2	1
Average Years’ Experience in Role
MD/Hospitalist   Nurses	7 6
Physical Therapy	9
Occupational Therapy	5
Social Worker	2
Administration	0.5
Women, n, (%)	2 (11)
Veteran, n, (%)	1 (5)

Rounding may result in values >100%, no participants identified as mixed/biracial or Latino/Hispanic

Screening and admission process varied by SNF

While participants at all three SNFs evaluated similar factors when screening patients for admission, the importance of these factors and the processes for evaluating patients varied. We noted variability in this process in three main categories: (1) use of different approaches to screen patients for admission to SNF (2) diverse types of gatekeepers to manage patient admission to SNF; and (3) different approaches to monitoring and managing patients once admitted. Table 2 summarizes each SNF’s process for screening and admitting patients to SNF.

Table 2.

– SNF admission processes by site

	SNF 1	SNF 2	SNF 3
Facility characteristics	- Part of large chain - Short-term care only for Medicare and private pay - In-house medical director and physician assistant manage most patients, supplemented by contracted providers	- Small chain - Predominantly long- term care with small skilled unit - Staffed by contracted attending physicians and advance practice providers	- Only accepts in- network patients - Predominantly short-term care with some long-term beds and hospice unit - Staffed by employed in-house physicians only
Screening process and admission decision	-Liaison conducts screening (nursing, business, respiratory therapy backgrounds) - More complex patients referred to Director of Nursing for review - Red/Yellow/Green system	-Liaison (speech pathologist) conducts screening using standardized form - Liaison decides to admit	- Screening by Multidisciplinary admissions committee - Attending physician visits patient prior to hospital discharge to evaluate if uncertainty
Factors Considered When Screening patients	- Primarily payer and perceived length/cost of stay - Patient complexity and stable discharge plan less important	- Payer and cost of stay - Patient complexity important, may turn away complex patients or those without clear discharge plan	- Rehabilitation goals and motivation of patient most important - Medical stability important, will refuse patients until more stable in hospital
Initial Assessment parameters in SNF	-Medical director sees patients within 24 hours of admission - Nurse checks medication list twice and reviews with physician, pharmacist not initially involved	- Attending physician sees patient within 3 days of admission - Approves medication list read by nurse over telephone initially - Pharmacist then reviews medications	- Physician assesses patient in hospital before transfer - Process for automated transfer of hospital medication records to SNF records - Pharmacist reviews medications on admission

Use of different approaches to screen patients for admission to SNF

For participants at SNF 1 and SNF 2, insurance coverage and the “cost” of a patient’s stay was the first and most emphasized step when screening patients for admission. As the admissions director of SNF 1 noted: “… [we] run the insurance information to verify … inpatient skilled nursing benefits first and foremost …After that we…make sure we don’t have any high cost expensive beds…if they have any [complex] behaviors [and if they have a type of] …discharge plan…” Following this initial screening, participants at both SNFs reviewed the patients’ medical documentation via fax or an online portal and sent liaisons to referring hospitals for an inpatient consultation. One liaison described the process: “...[I’ll] cross check what’s on their clinicals [sic]…check their cognition… do a really quick Q&A [question and answer] like ‘oh man, I see that you had a really bad bladder infection, like how are you feeling now’… then I just go through … stuff that I wasn’t able to glean from their clinicals like their date of birth, what kind of food they like.” (Liaison, SNF 2). While assessing rehabilitation goals and motivation to participate in rehabilitation were mentioned, they were mentioned less often or as less important than other factors. Liaisons from both facilities had diverse occupational backgrounds spanning nursing, speech therapy, and business, and reported spending most of their time in hospitals.

While payment was the primary gatekeeper for admission at SNF 1 and 2, appropriate fit was of primary concern for participants in SNF 3. SNF 3 clinicians emphasized evaluating whether a patient’s goals, motivation, and medical acuity were a good match for their resources. For patients referred to SNF 3 from the nearby hospital, screening started with an inpatient consultation form filled out by hospital clinician assessing the patient’s participation in rehabilitation goals and their motivation and willingness to go to SNF. This screening was followed by an inpatient consultation with a SNF 3 physician to evaluate the patient’s ‘ultimate potential’: “I will usually ask them ‘what do you think it will take to allow you to go home, how long do you think that might take [?]…what they think they’re gonna [sic] be able to accomplish’...” (physician, SNF 3). Other factors considered in admission included: “…other alternatives… [patients] have for placement…[and] whether they’re Medicare eligible...” (Physician, SNF 3). When the location of the hospitalized patient made inpatient physician consultation impossible, SNF 3 physicians screened patients based off available medical documentation.

Patient admission to SNF is managed by diverse gatekeepers

The ultimate decision to admit a patient to SNF varied across all three facilities. Participants at SNF 3 described the most team-based approach, while participants at SNF 1 described a triage system, and participants at SNF 2 relied heavily on one liaison’s discretion. At SNF 3, the ultimate decision maker was “usually the whole team” (Nurse, SNF 3). In contrast, participants at SNF 1 described a formal triage system, where liaisons from the parent organization reviewed the patient and decided which specific SNF was best fit for that patient, referring more complex patients to the appropriate director of nursing to assess if the patient was a good match for post-acute care at their facility: “…we have a green, yellow, red system. If its green …the liaison basically feels like they’re safe to come to our building and we can manage them. Yellow means then… [Nursing]… has to review it to make sure that we can meet that resident’s needs.” (Admissions Director, SNF 1). If a patient was classified as red, the patient did not meet the minimum safety criteria for admission to SNF.

In contrast, participants at SNF 2 described placing more trust in the liaison’s ability to determine if the patient is an appropriate fit for SNF: “…if I think medically that we can manage the patient and if they agree to actually rehab … and/or if they agree to try their best, I don’t typically clear it with the administrator or the nurse manager, they just kind of leave it up [to me] ...” (Liaison, SNF 2). The liaison described occasionally soliciting social work or nursing input on more complex patients, for example: “For patients that are kind of on the cusp, maybe people that are really, really, really high fall risks and if we don’t have rooms that are close to the nurses’ station, people that need 24 hour sitters then typically I review those with the administrator …” (Liaison, SNF 2).

Varied approaches to continuing medical management

Initial contact between a newly admitted patient and the SNF physician ranged from 24 hours to three days after hospital discharge, depending on the SNF facility. At SNF 1 and SNF 3 the first physician consultation visit usually occurred within the first 24 hours of patient SNF arrival. Both SNFs were staffed with in-house physicians. At SNF 2, where attending physicians and other providers were hired from an independent physician group, a physician noted: “So, normally I see people within three days… max, usually one to two.” (Physician, SNF 2)

The process for reconciling medications also varied by facility. Participants at SNF 3 described the most comprehensive process for patients admitted from the co-located hospital: “There is actually a mechanism within the computer ... [to rewrite] … current hospital orders…The pharmacist does an automatic review of everybody …so it serves as a back-up system…” (Physician, SNF 3). In contrast, participants at SNFs 1 and 2 described a manual medication review. Participants at SNF 1 described a check-and-balance process of having two nurses confirm the medication list while SNF 2 participants described having a pharmacist review the medication list. All SNF facilities expressed general trust and satisfaction with the process, however a nurse practitioner from SNF 2 noted: “It’s not ideal…you’re just approving this list that they’re [nurses] reading to you and a lot of times, they’re reading the wrong discharge list … you just have to go by what they’re telling you and basically just approve it or not approve it based on the dosage...”

External factors stress SNF screening processes and influence variability

Some of the variability in SNF selection and admission processes was explained by how the three SNFs respond to three external pressures: (1) inconsistent and inadequate transfer of medical documentation; (2) lack of understanding among hospital staff of SNF protocols and capabilities; and (3) hospital payment models that encouraged hospitals to discharge patients rapidly. In addition to influencing variability in the SNF process, participants described how these stressors made the admission process difficult and could negatively impact patient care. Table 3 summarizes these pressures and each SNFs’ response.

Table 3:

External pressures and SNF responses

	Inconsistent and inadequate transfer of documentation	Lack of understanding among hospital staff of SNF processes and capabilities	Payment models encourage hospitals to discharge quicker and sicker
SNF 1 Responses	Access to select hospital’s electronic medical records allowed them access select patient records, however not for all patients	Nurses educated some hospital nurses regarding how to medicate patient for the trip to SNF and the importance of ordering extra medications in case there is a delay with the SNF pharmacy	Purposefully staffed SNF with individuals with background in acute care setting and brought on an in- house physician
SNF 2 Responses	Access to select hospital’s electronic medical records allowed access to select patient records when needed Inpatient consultation confirms that medical documentation is correct	Nurses educated some hospital nurses regarding how to medicate patient for the trip to SNF and the importance of ordering extra medications in case there is a delay with the SNF pharmacy Physician reported sitting down with hospital staff to fix communication and knowledge gap, however, reported that nothing has improved	When screening patients, sometimes opted for simpler patients (hip replacement) in lieu of more complex patients (homeless, drug use) knowing other SNF facilities were more equipped to care for these patients
SNF 3 Responses	For nearby hospital referrals: Physician inpatient consultation allowed them to solicit information not available in medical documentation No accommodation for patients admitted from non-co-located hospitals	Involved the physician in the screening and admission process allowed SNF 3 to assess medical stability, instead of relying on the hospital’s assessment	Stressor not mentioned by SNF 3 clinicians

Inconsistent and inadequate transfer of medical documentation

Participants in all three facilities described inconsistencies between patients’ stated and actual medical acuity and/or disposition, delayed transfer of discharge paperwork, and incomplete hospital records. This delay in medical record transfer made it difficult for SNF clinicians to reconcile medications, execute patient care plans and assess if patient goals were the right fit for SNF. For example, one provider from SNF 2 explained that:” … when I don’t have those records, there’s really nothing I can do for the patient besides monitor their vital signs, their blood sugars… and obtain my own lab work and basically start from scratch to try to help them, so I think they go several days without the care that they need.” The same provider went on to explain: “If I get a discharge summary and I get the appropriate records that I need to care for the patient, I’m surprised and it’s about 10% of the time.” (Nurse Practitioner, SNF 2).

To some extent, access to select hospital’s electronic medical records alleviated inter-facility care coordination for SNF 2 and SNF 3. For example, it was easier to look up medications and notes about the patient’s hospitalization: “the hospital documentation usually always … lacks what we need… if you have a patient whose Coumadin was put on hold, why was the Coumadin put on hold? We don’t know… that was one of the reasons we got me credentialed to [private hospital], for me to be able to access the records.” (Physician, SNF 1). However, it didn’t resolve the reality that documentation was not a substitute for conversation– something the SNF 3 participants highlighted in their approach for screening patients. As noted above, when patients were hospitalized the SNF 3 physician had a face-to-face inpatient consult with the patient when feasible. This served as an opportunity to both double-check the medical documentation and have a conversation with the patient about their goals and how SNF could help them achieve their goals. As one SNF 3 clinician stated “…It’s not always clear from the notes … just what the ultimate potential of the patient is. That’s where having more information about the patient’s living situation, their motivation, talking to the patient directly…. help our sense of whether this is a patient who is going to rehabilitate and return to the community.” SNF 2 also emphasized the importance of in-person screenings, noting: “If I’m super busy or if I’m here by myself I will do a paper review [rather than an in-person review]. Every single time I’ve done that it’s come to back to bite me in the butt, [because]…something’s not in the paperwork.” (Liaison, SNF 2).

Lack of understanding among hospital staff of SNF processes and capabilities

All three SNFs described a lack of understanding among hospital clinicians about SNF processes and capabilities for monitoring and managing patients. As one SNF 3 clinician noted: “…… the expectation of what can be accomplished in a nursing home versus what is not reasonable to expect are differences that could be better addressed when patients are in the hospital.” (Physician, SNF 3). The effect on patient care ranged from delays in medication fills: “… when they [hospitals] do discharge orders… all their other routine medications never have a diagnosis and so that’s frustrating … I don’t know what it is with their…regulatory standards, I know cause we’re both regulated differently.” (Admissions Director, SNF 2) to hospital readmission: “the biggest limitation…to preventing hospitalizations… [is] what I can treat the patient with and how quickly I can get that treatment initiated… it makes working in this environment very challenging… from a hospital perspective…they don’t understand… (Physician, SNF 2)

How the different SNFs accommodated this stressor varied by facility. At SNF 2, a staff physician described trying to improve communication between hospitals and SNFs: “I’ve actually sat down with their chief medical officers …and said look…This is how we need to close this loop of communication and yet the problem still remains....”(Physician, SNF 2), while a nurse at SNF 1 noted having SNF nurses educate hospital nurses during nurse to nurse handoff: “…my nurses are great at telling the nurse on the other side to medicate them for pain … it’s a long process for the patients.” (Admission Director, SNF 1). For SNF 3, they addressed this by having physician inpatient consultations when feasible: “I think if you look at most community nursing homes, the receiving physician is not involved in the decision to admit the patient. That’s often done by a nurse and again done remotely… so I think that in terms of preventing re-hospitalization, we, since we do an assessment of their medical stability and …are skeptical so we want to lay eyes on the patient ourselves, review the records ourselves.” (Physician, SNF 3)

Payment models that encourage hospitals to discharge patients quicker and sicker

Clinicians from two facilities felt that patients left the hospital too soon, and that some patients were not ready for SNF, thus increasing the risk for readmission and making patient care difficult. Of the participants in two SNFs that described this pressure, one noticed it more acutely than the other. Both facilities attributed this to economic pressures put on hospitals to turn over patient beds. As one clinician noted: “… Medicare…has its restrictions … [to] save money… and yet what it ends up doing is potentially compromising the care of the patients because you’re putting pressure on the hospitals to get them out of the hospital sooner, when they’re not medically stable enough to come to an environment like mine where I don’t have the resources to be able to take care of them …” (Physician, SNF 2). When asked why this trend was happening, the participant warned that if the trend continues: “…you’d have to bump up your staffing, you’d have to hire respiratory therapists, you’d have to have an ability to get lab work done on site…you’d have to have an onsite pharmacy… parent corporations that own these facilities aren’t gonna [sic] be able to do that. They’d never survive.” (Physician, SNF 2).

Reactions to this pressure varied by SNF. Participants at SNF 1 described shifting their care model to include an on-staff physician and maintain high numbers of nursing staff to better care for these “sicker” patients. In contrast, participants at SNF 2 described dealing with this tension through sometimes turning away complex patients and opting for more straightforward patients: “…there have been times where … [a referral] uses heroin, he’s homeless and we don’t have a discharge plan…as opposed to the lady that is either gonna [sic] move up into long-term care or the little guy that has really good insurance and we can get home health … and we don’t actually admit that [homeless] person cause … the thing that’s hard about [SNF 2] sometimes is it’s a smaller company, so we don’t rival with the [private SNF company] …” (Liaison, SNF 2). One clinician from SNF 1 summarized this tension as: “it depends upon what kind of patients you are accepting and what kind of patients you have in your building. [Other private SNF] will have minimal re-hospitalizations. They only accept safer, stable patients, not too many medical complexities…so they’ll be just there for rehab and they’re just, they’re a short stay… so when you pick and choose, your readmissions can be curtailed. In our case, we don’t pick and choose.” (Physician, SNF 1). The clinician went on to explain that they are able to accept more complex patients because they have an on-site physician and: “…we have a pool of people who work for us who have worked in acute settings… so we have those, both the mindset as well as the skillset where we can take complex patients and be comfortable with it. We also live in an area…where there are patients who may not have access to health care as much…so we cater to that kind of patient population…” (Physician, SNF 1).

Discussion

Interviews with SNF clinicians revealed that SNFs have different screening and evaluation processes, as well as different priorities, when making decisions to admit patients. However, each is responding to similar influences: (1) inconsistent and inadequate transfer of medical documentation; (2) lack of understanding among hospital staff of SNF protocols and capabilities; and (3) hospital payment models that encouraged hospitals to discharge patients rapidly. Identifying best practices for patient screening and admissions at SNFs is critical as more than 15,000 facilities across the country face similar internal and external pressures (Kaiser Family Foundation, 2016) and a growing number of U.S. patients require post-acute or long-term care at skilled facilities (Kahn, Benson, Appleby, Carson, & Iwashyna, 2010; “Long-term Care and Patient Safety | AHRQ Patient Safety Network.”). Our results provide an important starting point for categorizing SNF screening and admission processes. Such categorization may be useful in future work for two major reasons: first, to be able to evaluate the connection between these processes and patient outcomes in a larger sample, and second, to provide a common language for disseminating best practices. Indeed, one of the reasons for development of the ITC framework was to provide a common “language” for describing care transitions interventions (Burke, Kripalani, Vasilevskis, & Schnipper, 2013).

Unfortunately, there is scant recent prior peer-reviewed literature describing SNF admission processes, and most is more than thirty years old and focused on specific roles (Newman & Toseland, 1985; Toseland & Newman, 1982). However, case reports describe individual SNFs revising their admission procedures to make them more standardized and streamlined (Baughman et al., 2018). This is striking when juxtaposed with more robust literature describing potentially adverse outcomes related to transitional care processes between hospital and SNF, such as readmissions (Britton et al., 2017; Burke et al., 2016; Mor, Intrator, Feng, & Grabowski, 2010; Ouslander et al., 2016). Potential contributors to these outcomes within other ITC domains are also beginning to be characterized, such as difficulties with medication management (K. Boockvar et al., 2004; King et al., 2013), missing information (K. S. Boockvar, Fridman, & Marturano, 2005; C. D. Jones et al., 2017), and inadequate discharge planning (Horney, Capp, Boxer, & Burke, 2017).

Strengths of our analysis include the diversity of SNFs included and roles interviewed. This allowed us to triangulate different perspectives within each SNF to accurately identify screening and admission processes. However, our data should be interpreted in the context from which it was derived. For example, while our goal was to describe the variability in practice across SNFs with different underlying patient populations, size, and ownership, identifying the screening and admission practices at only high-performing SNFs could provide additional useful insights into best practices. All of our SNFs were located in a single metropolitan area; rural SNFs or those in other regions of the country may follow different processes. While interviewing providers with different roles allowed us to triangulate these interviews to describe the process as accurately as we could, we did not observe the process in action and thus cannot assess whether actual processes might vary from those described.

Practice Implications

Value-based purchasing both increases incentives to discharge patients to SNF “quicker and sicker” but also provides counterbalancing forces (such as bundled payments or penalties for readmissions from SNF) that create incentives for SNF staff to shape how they assess, admit, and monitor new patients. Our results highlight innovations SNFs are already making to improve these processes that may be useful to other facilities responding to similar pressures.

To improve screening and admission processes, one of our SNFs had the attending physician visit the patient in the hospital to contribute to the screening and acceptance decision, often resulting in a “no for now” decision and clearer explanation of what must be accomplished in the hospital before SNF discharge. While this may seem impractical when the institutions are not co-located, telemedicine and video-conferencing approaches may provide a similar solution for SNFs. As hospitals create “narrow networks” of preferred SNFs, SNFs may find they have enough volume from their referring hospitals to make the investment of time worthwhile even for facilities that are not co-located.

Alternatively, ensuring hospital liaisons have appropriate clinical backgrounds and standardized training for adequate assessment of hospitalized patients may improve screening and admission decisions. The liaisons used by SNFs in our study had strikingly varied backgrounds, including liaisons without a clinical background.

Value-based purchasing may also inspire different SNF staffing models. SNF 1, for example, mentioned purposefully staffing more employees with acute care backgrounds and hiring an on-site physician to better handle their changing patient population. SNF 3 had multiple on-site physicians and involved them in every phase of the screening, admitting, and initial care process. These changes make it more likely a patient can be seen within 24 hours of admission, a critical perceived part of the monitoring process of both SNFs. Without scaling up staffing, SNFs can change screening criteria to accept lower-risk patients if staffing does not permit adequate monitoring; prognostic tools may be helpful in determining patient risk for adverse outcomes (Burke, Hess, Barón, Levy, & Donzé, 2018). However, this approach runs the risk of appearing to “game the system” to improve outcome rates (which may be uncovered when risk-adjusted outcomes are used for readmission penalties) and leading to fewer and fewer hospital referrals as the acuity of patients referred to SNF continues to increase. SNFs can adapt to the increased responsibility for patient outcomes and increasingly sick patients by prioritizing staffing clinicians who are better equipped to treat more complex patients and involving those staff members at multiple stages of patient admission and care, and the financial incentives under VBP may align the costs of increased staffing with payment.

Under new performance and payment measures, SNFs must choose how to respond. This study of SNF-clinicians provided several models that facilities can learn from or adapt. Hopefully, SNFs will embrace their increased agency within hospital-SNF relationships and choose to engage in adaptive practices that prioritize more careful patient screening, better patient care, improved health outcomes, and more equitable access to care for all.