Redesigning care delivery with patient support personnel: Learning from accountable care organizations

Abstract

Introduction

Accountable care organizations are a value-based payment model in the United States rooted in holding groups of health care providers financially accountable for the quality and total cost of care of their attributed population. To succeed in reaching their quality and efficiency goals, accountable care organizations implement a variety of care delivery changes, including workforce redesign. Patient support personnel—nonphysician staff such as care coordinators, community health workers, and others—are critical to restructuring care delivery. Little is known about how accountable care organizations are redesigning their patient support personnel in terms of responsibilities, location, and evaluation.

Methods

We conducted semi-structured 1-h interviews with 25 executives at 16 distinct accountable care organizations. The interviews were recorded, transcribed, and coded for themes, using a qualitative coding and analysis process.

Results

Accountable care organizations deployed patient support personnel to perform four clusters of responsibilities: care provision, care coordination, logistical help with transportation, and social and emotional support. Accountable care organizations deployed these personnel strategically across settings (primary care, inpatient services, emergency department, home care, and community) depending on their population needs. Most accountable care organizations used personnel with the same level of training across settings. Few accountable care organizations planned to conduct a comprehensive evaluation of their patient support personnel to optimize their value.

Discussion

Accountable care organization strategies in workforce redesign indicate a shift from a physician-centered to a team-based approach. Employing personnel with varying levels of clinical training to perform different tasks can help further optimize care delivery. More robust evaluation of the deployment of patient support personnel and their performance is needed to demonstrate cost-saving benefits of workforce redesign.

Keywords

Patient support personnel accountable care workforce redesign qualitative research health care delivery delivery system reform

Introduction

The Patient Protection and Affordable Care Act,¹ which was enacted in the United States in 2010, expanded health insurance coverage to millions of the previously uninsured. It required insurance companies to provide minimum packages of benefits determined by the federal government and created financial incentives for value-based care to try to reduce health care costs. The Centers for Medicare and Medicaid Services, a federal agency within the U.S. Department of Health & Human Services that administers several key health care programs such as Medicare (the federal health care program for seniors), Medicaid (the federal need-based program), and others, announced that they expect to move 90% of all Medicare fee-for-service payments to alternative payment models by 2018. Alternative payment models, as opposed to fee-for-service that encouraged increased volume of care, emphasize measured quality, patient experience, efficiency of care, harm from care, as well as reductions in overall costs.²

Economists have shown that mixed payment models are superior to fee-for-service and capitation when it comes to patient outcomes.³ A fee-for-service payment model encourages overuse of services and leads to fragmentation of care delivery. Historically, oversight systems in the United States have held providers accountable only for the care within their control.^4,5 Given the typical Medicare beneficiary saw a median of two primary care physicians and five specialists working in four different practices in a single year (for beneficiaries with multiple chronic conditions, these numbers were even greater),⁶ such individual accountability led to poor coordination and failed transitions. Capitation, by contrast, leads to underuse of services because providers’ payments are not tied to the quality or quantity of the care they provide.³ Alternative payment models aim to foster shared accountability by offering financial incentives for well-coordinated longitudinal patient care.

While many countries have implemented various alternative payment models, such as performance-based contracting in family medicine in Turkey,⁷ Practice Incentive Payments in Australia,⁸ pay-for-performance programs in the United Kingdom⁹ and New Zealand,¹⁰ provider payment reform in Vietnam,¹¹ and a shared savings program in the Netherlands,¹² the optimal mix of incentives has been elusive.¹⁰ The alternative payment model introduced by the U.S. health reform was accountable care organizations (ACOs). ACOs are groups of health care providers held financially responsible for the quality and total cost of care of the population they serve. As of April 2016, the largest Medicare ACO program, the Medicare Shared Savings Program, included 433 ACOs that covered nearly 8 million beneficiaries in 49 states and Washington, DC.¹³

All ACOs sign a contract with a federal health care program (Medicare or Medicaid) and/or a commercial payer. The contract includes bundled payments¹⁴ around specific conditions and encourages disparate providers to better coordinate care and provide more appropriate and efficient care because they are at risk together for the same pool of funds. ACO contracts also include a measurement of results in the domains constituting the “Triple Aim” of quality, efficiency, and patient satisfaction.¹⁵

The changes in payment structure require radical new approaches in care delivery. For example, given the financial risk involved, ACOs are pressed more than traditional fee-for-service health care professionals to coordinate services across the care continuum and proactively connect with at-risk patients to avoid adverse results in quality, patient experience, and cost. These new approaches to care delivery require new work practices, but these cannot be accomplished without redefining job responsibilities of the core personnel providing support for patients. For example, to improve care coordination, ACOs need to strategically assess their workforce, hire new staff, and train and redeploy existing personnel in novel settings.¹⁶ Professional roles responsible for care coordination in the past, nonphysician staff (e.g., nurse navigators, care coordinators, case managers, and community health workers), and new roles (e.g., health coaches, scribes, and panel managers) are a prime target for workforce redesign in ACOs. We refer to these staff collectively as “patient support personnel” (PSP): staff who provide additional support to patients and families during their health care experience.¹⁷

Little is known about ACOs’ motivations and tactics for workforce redesign. Through interviews with ACO executive leaders, we sought to learn about the responsibilities of PSP, the settings in which ACOs deploy them, and evaluation strategies of their deployment and performance. As policymakers consider ways to improve new and existing ACO programs and other alternative payment models, our study explores what workforce solutions ACOs are already implementing in response to policy changes. Our findings provide a context for devising strategies to use human resources to improve quality of care while containing or reducing costs.

Methods

Design

We used a qualitative research design to understand if and how health care providers were using PSP within their ACO. Because little is known on this subject, qualitative interviews provided the best opportunity to learn about both the breadth and depth of ways ACOs were employing, deploying, or using PSP in new ways. We used an interview guide consisting of six broad sections: ACO formation and prior relationships, governance and leadership, clinical and population management priorities, tactics in care delivery transformation, challenges, and future goals. Each section consisted of several open-ended questions and probes (Table 1). The interview guide was formulated over several months of weekly team discussions. Interviews lasted approximately 1 h. The research protocol was reviewed and approved by institutional review board of Dartmouth College. Verbal informed consent was obtained from all participants.

Table 1.

Topics and sample questions from the interview guide.

Topic	Sample questions
Formation and prior relationships	What was the relationship among the participating organizations prior to the ACO initiative? If it has changed, how?
Governance and leadership	Could you briefly describe the leadership structure of your ACO? What kinds of committees has the ACO formed? Have you added any additional committees since our last interview?
Clinical and population priorities	What are your ACO’s priorities? For each area of focus, do you have specific goals or targets in these priority areas?
Care delivery transformation	What changes in clinical operations have taken place as a result of the ACO? Probes: New staff? Standardized treatment guidelines? Disease-management programs? Cross-organizational disease registries? Efforts to identify high-risk patients? Team meetings across settings? Care transition protocols? Integrated EMR or HIE? Of all the programs, personnel, and activities you’ve mentioned, which do you think has been most successful across sites? Why?
Challenges	What has been the biggest barrier or challenge to implementing new tactics? How might/have you overcome these challenges?
Future goals	What one change in the clinic do you think the ACO would most benefit from implementing that hasn’t yet been implemented?

ACO: Accountable care organizations.

ACO participants

Our team (TF, KT, and BH) conducted semi-structured phone interviews with 25 ACO executives at 16 distinct ACOs in July and August 2014. We identified a sample of 22 ACOs using the diverse case selection, a method commonly used in social sciences to ensure that the selected participants represent a variety of manifestations of the phenomenon under study.¹⁸ A subset of sites in this sample had been interviewed by our research team for related work in the prior year (2013). Participating ACOs varied widely by region, number of ACO contracts, type and combination of payers, composition, inclusion of safety net providers, and rurality (Table 2). Our respondents were most often ACO leaders (chief executive officers, chief medical officers, chief financial officers, or chief operating officers) and clinical leaders (directors of primary care).

Table 2.

Characteristics of the interviewed ACOs (N = 16).

	N (%)
Location of ACO
New England	3 (19)
Mid Atlantic	3 (19)
South	2 (13)
Midwest	2 (13)
West	4 (25)
Southwest	2 (13)
Payer of ACO contract/s
Medicare shared savings program (MSSP)	10 (63)
Medicaid	5 (31)
Commercial payer/s	4 (25)
Composition
Physician-led	5 (31)
Hospital-led	11 (69)
Urbanicity
Urban	8 (50)
Rural	2 (13)
Suburban	4 (25)
Mixed	2 (13)
Safety net providers in ACO
Mostly safety net	8 (50)
At least one safety net provider, but mostly not safety net	5 (25)
No safety net providers	3 (19)
Number of ACO contracts
Single contract	11 (69)
Multiple contracts	5 (31)

ACO: Accountable care organizations.

Analysis

To systematically analyze each ACO’s approach to employing new personnel, we used an iterative coding and analysis process. Interviews were audio recorded, transcribed, coded using QSR NVivo software, and analyzed for themes (KG).¹⁹ We first developed a code definition for “patient support personnel” or “patient support staff” (we use these terms interchangeably) through a deliberative process and team discussion (KG, TF, and VL). We defined “patient support staff” as nonphysician personnel who provide support to patients and family during their health care experience.¹⁷ We focused on PSP whose roles, according to our respondents, had been created or changed strategically to reach ACO goals. This excludes, for example, pre-existing care managers whose roles or responsibilities did not change in any noticeable way due to ACO activities.

After reviewing all coded excerpts, four additional codes were developed: location, content of responsibilities, funding, and target population (KG, TF, and VL).²⁰ Interviews were coded again using these four codes (KG). The lead author (KG) shared an evolving qualitative memo with VL and TF to ensure validity of themes.²¹ VL and TF provided biweekly comments over several months to further hone the analysis. For example, our team discussed inclusion and exclusion criteria for the coding definition; choice of quotes from participants, as well as their meaning; organizational structure of the findings; and implications of the study.

Results

All those we interviewed said their ACOs made changes in the deployment, responsibilities, and structure of PSP’s work. This work included responsibilities across four domains: (a) care provision (needs assessment and coaching, medication management); (b) help with coordinating care (making appointments, facilitating information flow); (c) logistical help with transportation; and (d) social and emotional support. A few organizations were experimenting with home and community settings. In all ACOs, implementation of workforce redesign strategies was based on stratifying patients into subgroups; however, the rationale for stratification varied across organizations. ACOs deployed patient support staff strategically across settings: clinics, hospital units, and central locations.

Responsibilities of PSP

Care provision

In many ACOs, PSP participated in provision of care (Table 3). Their responsibilities mainly consisted of assessing the patient’s needs (fall risk, home safety, nursing services, or equipment issues); coaching (reviewing discharge instructions, educating patients about alternatives to the emergency department (ED)); and medication management.

Table 3.

Care provision.

Needs assessment and coaching

The clinical coordinators … [a]re really trying to connect people to outside resources, our community partners, working with them on trouble shooting, any issues that come up around things like medication or maybe needs at home in terms of nursing services or equipment. We make initial home visits for a new patient assigned to your panel and that gives an excellent opportunity for the care manager … what we're going to call now care coordinators … an opportunity to see people in their homes and assess fall risks, assess safety issues and actually see the medication is not only in the bag that they would bring to the office, but also what's under the sink and in the medication – or in the medicine cabinet and all the things that they don't know about. […] So it's an excellent opportunity to really gain some insight into what this patient is all about. Our care coordinators are not – we’re not – licensed for home care so really what they do are assessments in education and so they’re really a provider’s eyes-on-the-ground if you will in the house so if we’ve got somebody who has multiple weekly visits they’ll be going in there talking with the patient and potentially the caregiver, assessing […] if there’s … some social need for the patient. So we look at patients who wind up in the emergency room and we evaluate whether they're in emergency room for an appropriate or inappropriate reason. [For] anybody who inappropriately utilized the emergency room, [we] will be in contact with them to educate them on how to appropriately utilize the primary care provider so they don't wind up in the emergency room.

Medication management

… there’s always a follow-up call for the patients who go directly home by one of our … outpatient nurse navigators within 24 hours to review the medications. … probably a 30 to 40 percent of the people [our care managers] see on that first [home] visit require some intervention be that correcting – they probably are the fourth or the fifth medication reconciliation […] but frequently there are issues with medicines or they end … up needing to do some sort of an intervention increasing medicine, dosing them, drawing in a lab, etcetera. … we have actually two in-house nurse navigators who follow all the ACO patients and one thing that they do is they personally review the medication list with the patient and family after the physician has reviewed the med rec and written a discharge plan. So the nurse will actually sit down with the patient and this is one of the ACO employee nurse navigators. […] So the patient will understand what their new medication regimen is.

Some personnel traveled to patients’ homes after discharge to assess their needs and provide the necessary education about their disease management; others provided patient education and medication reconciliation by phone. A staff member at one ACO proactively reached out to low-acuity patients in the ED to address their needs. At another ACO, patient support staff reviewed discharge instructions with patients at the hospital, learning about their needs and concerns, and trying to address them before patients went home.

PSP also performed medication management, such as medication reconciliation. Patients often received new prescriptions during a primary care visit, an inpatient hospital stay, or an ED visit; this situation required medication reconciliation to prevent adverse drug events. ACOs reported having multiple medication reconciliations: in the hospital, during the follow-up phone call post-discharge, in primary care clinics, and in patients’ homes. PSP also encouraged medication adherence by discussing each medication with patients and their families.

Help with coordinating care

ACO leaders reported using their PSP to improve care coordination across providers and settings (Table 4). Specific coordination tasks included making appointments for patients and facilitating information flow across the care team, as well as between primary care providers, inpatient services, the ED, skilled nursing facilities, and home.

Table 4.

Care coordination.

Making appointments

We get ‘warm handoffs’ to be sure those patients do get seen as opposed to leaving the patient to call on their own and getting stuck in phone queues and being told there’s no appointments. The case managers make the appointments for them. They have a contact person for the office, kind of the Bat Phone [to] get that patient in. … most of the practices have an assigned person that outreaches to the patients who’ve been discharged and contact them to come in for an appointment with their PCP. Oh the other thing is trying to emphasize getting post-discharge patients back to their PCP within five days. We’re working on that. I mean you can always refer people somewhere to get services done but It’s really nice if the patient is in the emergency room that you can actually go down and start addressing it at that point in time. […] So at this point it’s really wonderful to be able to say, ‘Okay let’s talk about your diabetes but at the same time I’m going to bring in our care coordinator who’s going to help us figure out how to get you into this, this and that.’

Facilitating information flow

Our second year we’ve been working on transitions of care, working on identifying sub-acute facilities that are preferred providers for the ACO, creating hopefully a seamless system of […] moving patients from the hospital to the sub-acute, from the sub-acute to the home with a minimum of change in their expectations, care, keeping the families informed, et cetera, et cetera and also utilizing home healthcare agencies to assist us in that area. We’ve recently … instituted a group of mostly clerical people called panel managers who reach out to patients if they’ve had studies done outside of our practice to fill out records release forms, to obtain records that would show that certain studies have been done. The big push from hiring new staff is around scribes and we're continuing to try to grow that program. … [I]t provides more labor to primary care doctors at the point of care, to help close these gaps in care because one primary care provider can't do all the clicks on the computer. […] [P]hysicians don't need to be transcribing or doing paperwork when they really need to spend their time making medical decisions. So if you have a scribe, usually a medical assistant, that actually comes into the room with you as you see your patient and you examine the patient, you look at them eyeball to eyeball rather than back to your patient, dealing with a computer, you have a better patient experience that way. […] … he can concentrate on thinking about what's wrong with this patient, […] rather than being a part clerk, which is a poor use of his time, his or her time.

Several ACOs used PSP to schedule appointments with primary care and later remind patients about their appointments. Almost all ACOs reported calling patients post-discharge after ED visits and inpatient hospitalizations, but the timeframe for a follow-up phone call varied. One ACO leader reported having a scheduler in the outpatient clinic who handled calls from the inpatient services and the ED to ensure patients could get needed appointments. Two ACOs had initiated “a warm handoff,” a process in which the discharging inpatient team member calls primary care providers and speak to them directly about the needs of the patient. This team member also documents all details in the electronic medical record. In one ACO, a care coordinator connected the patient to necessary medical and social services at the end of a routine diabetes appointment.

A major part of care coordination for PSP is facilitating information flow across the care team and external settings. One ACO hired a group of panel managers, whose job was to ensure that results of all tests obtained outside of the practice were entered in the electronic medical records. The purpose was to eliminate duplicate testing and allow nurses to focus on direct patient care instead of finding documentation. Another ACO used scribes to fill out electronic medical records, which allowed primary care physicians to engage more directly with patients during examinations.

Logistical help with transportation

Several ACO leaders reported that their PSP arranged transportation for safety-net patients to get to the clinic (Table 5). Some patients required repeated transportation appointments. One ACO leader told us about a patient whose handicapped van had broken down; without the van, the woman was not only unable to make her medical appointments, but remained home-bound. Eventually, a staff member arranged for the van to be fixed for free at the local community college.

Table 5.

Logistical help with transportation.

Arranging transportation

[Our centralized care managers] arrange for transportation and they have a whole list of ways to get patients moved around. … the community health workers […] are stationed … in the medical clinic. They’ll do a lot around helping people around – appointment reminders, helping them understand how to set up transportation to get to appointments, calling them after they’ve been in the hospital just to kind of check in to see how they’re doing, make sure they got their medications and then if there’s any issues they … move it on the next level of care. We have a patient who is a 50-some odd year-old solidimide survivor and has been in a wheelchair her entire life. She is morbidly obese. She has multiple wounds because of that, but her biggest issue is transportation. So a care manager has been spending most of their time trying to figure out how to get their handicapped van fixed so they can get out, they can get to the doctors, they can move around the way they’d like to.

Social and emotional support

Many interviewed executives thought that PSP’s value lay in their ability to provide social and emotional support to patients (Table 6). According to our respondents, physicians, pressed for time and hurrying to their next appointment, felt unable to learn or address their patients’ nonmedical concerns; anxious patients needed more time to process the information and ask questions. When an approachable staff member reviews the information with patients and answers their questions, patients learn to trust this person. Later, the patient would remember to call this staff member when she felt unwell. Thus, PSP invest time and energy in getting to know people and building relationships with both patients and caregivers. Several ACO leaders reported receiving positive feedback from patients and families about how these personnel helped them. Our respondents noted how these changes led to some high-risk patients exiting the “revolving door” of the ED and repeated hospitalizations.

Table 6.

Social and emotional support.

So the community health worker is a way to build that trusting relationship, to get those conversations going about what really is going on with you, client, and what do you really need and want and what are your goals, not what the doctor’s goals, but what are your goals. So it’s really about relationships and it’s about empowering the member to understand what they want and how they can get what they want and how to work on it, and what supports we can offer. So on a routine basis [the physician wi]ll walk out of a room after seeing the patient and the nurse navigator will walk in and review everything with the patient […] So the nurse will review everything with them and they sort of feel, ‘Oh this is a person who cares about me.’ That translates into, ‘Okay, I’m going to call [Mary] because I don’t feel good,’ and [Mary] who’s the nurse navigator will be able to make recommendations and/or get the patient in that day or discuss it with the doctor and get back to the patient. I’ve got anecdotal stories where we’ve saved unnecessary hospitalizations based on the fact that the patients or the families call the nurse navigators and these are – they’re not only heartwarming but it really has made everyone recognize the value of this person even though they’re not ‘generating income’. They’re invaluable in the process. … a recurring pattern with care management […] is addressing the patient’s obstacle so that health can once again become a priority for them. Because if they’re in a bad social situation or the electricity’s off or they can’t afford their meds, until you address those things, you can’t really move the needle for that person individually. I get a lot of anecdotal stories actually from patients who are just excited to know that they have somebody who’s a direct contact who knows them who can help them navigate. I actually just recently ran into a patient that I was seeing out in one of the shelter clinics who happened to bring up his [ACO] nurse care coordinator and how amazing she’d been. He’d just gotten out of treatment. He was only in the shelter for a week. He’d gotten employment services. He was looking to start his job. He’s getting help getting a first month’s rent. He got new teeth. He was so excited. He’s like, ‘I haven’t smiled so much in so long and it’s all because of [Joyce]. I just can’t say enough about her.’ … we have a couple community health workers that are at the Native American Youth Association, and of course the Indian, Native American … population has tremendous history of being abused by our medical system. So getting that population to trust, those community health workers need a different skill set from the ones that are working in direct health care delivery. […] Some of [the community health workers] are master’s level social workers, and some of them are medical assistants. […] Some of them have no background like that, they’re just people who got the training. A lot of – especially when you’re coming from specific cultures, it’s a wide variety of people, young and old, male and female, all different types.

Risk stratification and workforce redesign

To better address patients’ needs, all interviewed ACOs divided their patient populations into subgroups⁵ based on risk factors, disease severity, and needs (Table 7). Based on our data, the overall rationale for providing additional support to specific patient subgroups appeared to be cost-driven. For example, many executive leaders reported focusing their efforts on the ED, a notoriously high-cost setting, but the scope of these efforts varied across ACOs. Some ACOs flagged all patients who were hospitalized or visited the ED for receiving extra care from PSP. Other organizations focused more narrowly on high utilizers of the ED and patients with low-acuity ED visits that were avoidable. At a few ACOs, PSP worked with high-cost patients, irrespective of their disease profile, while others (notably ACOs serving large safety-net populations) used patient support staff to reach out to vulnerable patient subgroups whose care cost is likely to be high in the future (e.g., children in foster care, pregnant women who are actively using drugs, minority and immigrant adolescents, homeless people, patients from skilled nursing facilities, and patients with chronic conditions and polypharmacy).

Table 7.

Risk stratification and workforce redesign.

Prioritizing subgroups of patients for additional support

So this [24/7 triage line] is manned by two RNs and one mid-level. […] people that man the triage line […] can pull up your record at the time of the call. So they get to know what medicines are on, […] anything that's in your EHR […] which makes a much more informed choice as to what level of care for their particular problem they need. [W]hen a high risk or a high utilizing patient is identified we get them connected with our care managers and try to see what things are driving those particular problem areas and try to hopefully address them. We try to get – bring to the attention of physicians and teams of people the patients who are moving back and forth or having problems and get them focused and hopefully settled if possible or discuss end of life decision kind of hospice or palliative care more formally. We … have instituted IDT. It’s a[n inter]disciplinary team in the clinics and some telephonically. So we’ll take the clinic staff and bring in the medical management, case manager, social worker and round on the high risk patients. We do that in some of our busier clinics once a week and then we do it telephonically for some of the contracted providers. … I think it’s important as we look at our patients and how we [train] these care managers to understand those [patients] that need intensive kind of management for a period of time and what does that look like versus somebody who might need to be checked in on over a period of time and what does that look like. We also have a new intensive case management team who goes – they’re seeking people who have mental health or chemical dependency problems, recent hospitalizations, be hospitalized in out of network hospitals and they’ll actually go out fairly quickly to referral. So our care managers […] initially … were pretty much reviewers of health maintenance and making sure people got the things that they needed to get. And over the last … three or four plus years they’ve transitioned over to more high risk care management so they’re truly care managers at this point. They’re […] RNs. They’re not social workers but there is some case management that they do as well. […] so it’s finding very high risk patients and in a structured way doing what we can to address their needs.

Although executives were interested in data-driven approaches to patient stratification, only some noted well-established communication between the ED and primary care clinics to provide timely information on admitted ACO patients. One ACO leader mentioned using an analytic tool for identifying patients at risk for readmission. At two ACOs, leaders informally surveyed physicians and staff, asking which patients they thought needed extra support.

Overall, few interviewees were using a clear business model to hire and redeploy PSP. Models of workforce redesign varied across ACOs because they were created to serve specific patient subgroups. Only two ACOs mentioned planning to conduct an evaluation of their PSP to optimize their value. Other evaluation efforts focused primarily on process measures, such as post-discharge follow-up with primary care.

Most ACO executives realized demonstrating financial returns on investment in new staff might take time. However, these leaders believed restructuring their labor force with ACO goals in mind was fundamental to success in a value-based reimbursement world that was here to stay.

Strategic deployment of patients support personnel across settings

ACOs in our sample deployed new and existing patient support staff strategically to address the needs of population they served, and all did so in more than one setting. Their responsibilities varied by location (Table 8). For example, PSP embedded in primary care clinics usually performed responsibilities in all four clusters (care provision, care coordination, logistical help with transport, and social and emotional support). Centrally located staff performed most of their duties by phone, which imposed certain restrictions on what they do. For example, some tasks (like assessing patient needs, and providing patient education and emotional support) require a face-to-face interaction. At the same time, lack of personal interactions could, according to one ACO leader, make these centrally located staff more efficient. At the time of the interview, this ACO was transitioning their registered nurse case managers to a centralized model; in the understaffed clinics where they were located, they were constantly asked to provide direct patient care, which diminished their productivity on ACO-related tasks.

Table 8.

Types of responsibilities by location and role.

Embedded in primary care clinics	Education: role
Care provision (needs assessment and coaching; medication management) Care coordination (making appointments; enhancing information flow) Logistical help with transportation Social and emotional support	RN: care coordinator, care manager, case manager, nurse navigator MA: scribe Unknown: panel manager, CHW
Centrally located
Care provision (medication management) Care coordination (making appointments, enhancing information flow)	RN: care coordinator, case manager, care manager, panel analyst
Logistical help with transportation
Emergency department and inpatient services
Care provision (needs assessment and coaching; medication management) Care coordination (making appointments; enhancing information flow) Logistical help with transportation	RN: care coordinator, clinical coordinator, care manager, nurse navigator Unknown: health partner
Home visits
Care provision (needs assessment and coaching; medication management) Care coordination (enhancing information flow) Logistical help with transportation Social and emotional support	RN: care manager Unknown: CHW
Community settings
Care provision (needs assessment and coaching) Social and emotional support	RN: care coordinator, clinical coordinator, care manager Unknown: health partner

RN: registered nurse; MA: medical assistant; CHW: community health worker.

Many interviewed ACOs spoke about the value of PSP in inpatient services and the ED. As Table 8 shows, our respondents were less likely to mention that PSP in those settings provide social and emotional support, perhaps due to the fast pace of their work or because our interviewees were unaware of those instances. By contrast, home care and community settings encouraged engagement with ACO beneficiaries and providing them with social and emotional support. Two ACOs restructured the health care encounter by deploying new community health workers in settings external to the ACO, such as community organizations (e.g., the Native American Youth Association), charities (e.g., the Lutheran Family Services), civil rights organizations (e.g., the Urban League), correctional facilities, midwifery clinics, skilled nursing facilities, public housing, homeless shelters, and sobering stations. The rationale for deploying PSP in those settings was to (a) build relationships with people who have low trust in the health care system and (b) provide them with preventive services before they need to go to the hospital or the ED.

We also examined the distribution of PSP across settings (e.g., inpatient, primary care, etc.) by education and role (nurses, medical assistants, community health workers, etc.) and by type of responsibilities (e.g. care provision, care coordination, logistical help, social/emotional support) (Table 8). The roles and education backgrounds of PSP were more ACO dependent rather than location or task dependent. For example, an ACO that used nurse care managers was likely to employ them across most settings, to perform almost all kinds of tasks congruent with their training. There were no strong links between the type of PSP and their job responsibilities.

Discussion

All executives we interviewed reported redesigning some aspects of their ACO’s patient support workforce; many had added new personnel. Responsibilities of these employees fit four domains: care provision, care coordination, logistical help with transport, and social and emotional support. All ACOs reported using some type of risk stratification to identify patients most in need of additional support. Approaches to risk stratification varied across ACOs depending on the number of available personnel and the needs of their attributed population. Respondents reported deploying support personnel in primary care clinics, a central location, inpatient services, and EDs. A few organizations also deployed them in community organizations, charities, and public housing. Several ACOs provided home care to high-need, high-risk patients. Our respondents emphasized the social and emotional support PSP provided in these novel locations.

Our findings suggest that health care leaders and policy makers in other countries who are implementing alternative payment models may benefit from assessing their patient population and care delivery models to identify gaps in care. Hospital administrators and other leaders can then plan their workforce restructuring strategically to address problem areas with the help of personnel with appropriate training and experience. For example, some ACOs in our study used community health workers in most settings; other organizations deployed registered nurses to complete similar tasks. Employing personnel with varying levels of clinical training to perform different tasks can help further optimize care delivery.

Some approaches ACOs used to redesign their workforce are consistent with prior research on opportunities to reduce costs and improve quality of care; other approaches were more novel. By employing new staff, such as panel managers and scribes, and rearranging the scope of responsibilities of existing personnel so that more people could practice “at the top of their license,”²² ACOs sought to optimize the productivity of their workforce.¹⁶ By connecting at-risk patients with primary care and addressing their needs in low-cost settings such as primary care clinics and home, PSP were helping shift care away from the more expensive ED or inpatient care. Prior research suggests that focusing on “high-utilizer” patients with many hospitalizations and ED visits and a score of unmet medical and social needs can help reduce costs.^23–25

A unique approach used in only a few ACOs was embedding PSP with culturally appropriate training in the community to connect immigrant, minority, and other safety-net populations to primary care. Minorities face language barriers, a lack of trust in the health care system, financial burdens, low health literacy, and other cultural or religious barriers.^26,27 Low access to primary care may lead these people to develop conditions and seek health care at advanced stages of the disease, putting additional financial burden on the patients and the health care system. More robust evaluation approaches are necessary to understand the impact of these strategies.

Our respondents indicated that ACOs have put PSP in charge of care coordination and logistical help in various settings.²⁸ While the model in which multiple providers take care of the same patient is a problem that plagues health care in many countries,²⁹ PSP could connect disparate parts of the system while focusing on patients’ needs. However, reaping the benefits of improved coordinated care may require physicians to relinquish some authority to PSP, and it is unclear whether physicians will embrace such change.

Our analysis has several limitations. First, although the ACO executive leaders we interviewed were knowledgeable about the workforce changes within their organization, they have a high-level view of the process. Furthermore, because organizational leaders are more likely to emphasize stories of success, how they portrayed workforce redesign in their ACOs may appear more problem-free than it actually was. Interviews with practice leaders, physicians, PSP, and patients, and observational and ethnographic studies of primary care clinics’ workflows, could help determine if and to what extent team-based approaches are truly integrated within practices. For example, it would be interesting to know how much independence PSP have in making their decisions or if they still have to consult the physician regarding many of their activities. Second, interviewing the same ACO leaders again at a later time could reveal which strategies remained and which did not. Finally, only some ACOs mentioned providing home visits and embedding PSP in community organizations; future studies could investigate whether these approaches become more widespread.

In conclusion, our data suggest that workforce redesign is an area of concerted efforts for ACOs. PSP perform responsibilities from four major domains: care provision, care coordination, logistical help with transport, and social and emotional support. Overall, the deployment and expansion of PSP in ACOs indicates a broader shift from physician-led to team-based approaches in health care. With new communication technologies, new opportunities to redesign health care workforce may emerge. For example, face-to-face time with physicians may decrease as phone or online communication provides opportunities for remote consultations. As a result, interactions between patients and support personnel may proportionally increase, affecting the traditional roles, responsibilities, and professional identities of physicians and nurses³⁰ and opening up opportunities to renegotiate boundaries between professions.

Footnotes

Acknowledgments

We thank Ms. Katie Tierney and Mr. Benjamin Harris for their contribution to data collection and Ms. Karen Klein for her editorial assistance. We are also grateful to participants of the Professional Development Seminar at The Dartmouth Institute for constructive feedback on earlier versions of the manuscript and anonymous reviewers for their helpful and insightful comments. Finally, we thank participating ACOs and each of our interviewees who graciously shared their experiences with us.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by a grant from the Commonwealth Fund (award number: 20150495).

References

Patient Protection and Affordable Care Act, 42 U.S.C. § 18001 (2010).

Francis

Clancy

. Implementing performance pay in health care: Do we know enough to do it well? J Gen Intern Med 2016; 31: S6–S7.

Brosig-Koch

Hennig-Schmidt

Kairies-Schwarz

. The effects of introducing mixed payment systems for physicians: experimental evidence. Health Econ. Epub ahead of print 28 December 2015. DOI: 10.1002/hec.3292.

Fisher

Staiger

Bynum

. Creating accountable care organizations: the extended hospital medical staff. Health Aff (Millwood) 2007; 26: w44–w57.

Porter

Pabo

Lee

. Redesigning primary care: a strategic vision to improve value by organizing around patients’ needs. Health Aff (Millwood) 2013; 32: 516–525.

Pham

Schrag

O'Malley

. Care patterns in medicare and their implications for pay for performance. N Engl J Med 2007; 356: 1130–1139.

World Bank. Turkey - Performance based contracting scheme in family medicine: design and achievements. Washington, DC: World Bank, http://documents.worldbank.org/curated/en/880861468338993763/Turkey-Performance-based-contracting-scheme-in-family-medicine-design-and-achievements (2013, accessed August 12, 2016).

Cashin

Chi

. Major developments in results-based financing (RBF) in OECD countries: Country summaries and mapping of RBF programs Australia: The practice incentives program (PIP), Washington, DC: The World Bank, 2011.

Campbell

Reeves

Kontopantelis

. Quality of primary care in England with the introduction of pay for performance. N Engl J Med 2007; 357: 181–190. http://www.nejm.org/doi/abs/10.1056/NEJMsr065990 (accessed August 12, 2016).

10.

Buetow

. Pay-for-Performance in New Zealand primary health care. J Health Organ Manag 2008; 22: 36–47. http://www.ncbi.nlm.nih.gov/pubmed/18488518 (accessed August 12, 2016).

11.

Cashin

Phuong

Shain

. A simple simulation model as a tool to assess alternative health care provider payment reform options in Vietnam. Global Publ Health 2015; 10: S104–S119. https://www-tandfonline-com.web.bisu.edu.cn/doi/abs/10.1080/17441692.2014.986156 (accessed August 12, 2016).

12.

Hayen

van den Berg

Meijboom

. Incorporating shared savings programs into primary care: from theory to practice. BMC Health Serv Res 2015; 15: 580. http://www.biomedcentral.com/1472-6963/15/580 (accessed August 12, 2016).

13.

Anon. 2016. Medicare Shared Savings Program ACO Assigned Beneficiary Population by ACO by County. Fast Facts. Centers for Medicare and Medicaid Services, https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/sharedsavingsprogram/Downloads/All-Starts-MSSP-ACO.pdf (accessed August 19, 2016).

14.

Guterman

Davis

Schoenbaum

. Using medicare payment policy to transform the health system: a framework for improving performance. Health Aff 2009; 28: w238–w250. http://content.healthaffairs.org/cgi/doi/10.1377/hlthaff.28.2.w238 (accessed August 19, 2016).

15.

Berwick

Nolan

Whittington

. The triple aim: care, health, and cost. Health Aff 2008; 27: 759–769. http://content.healthaffairs.org/cgi/doi/10.1377/hlthaff.27.3.759 (accessed August 19, 2016).

16.

Ricketts

Fraher

. Reconfiguring health workforce policy so that education, training, and actual delivery of care are closely connected. Health Aff 2013; 32: 1874–1880.

17.

Patel K, Nadel J and West M. Redesigning the care team: the critical role of frontline workers and models for success. Washington, Dc: Brookings Institution, March 2014, https://www.brookings.edu/wp-content/uploads/2016/06/FINAL-Hitachi-Toolkit-32014-1.pdf (accessed August 19, 2016).

18.

Seawright

Gerring

. Case selection techniques in case study research: a menu of qualitative and quantitative options. Polit Res Q 2008; 61: 294–308.

19.

QSR International Pty Ltd. NVivo Qualitative Data Analysis Software. Australia: QSR International Pty Ltd, 2014.

20.

Crabtree

Miller

A template approach to text analysis: developing and using codebooks. In: Crabtree

Miller

(eds). Doing Qualitative Research, Newbury Park, CA: Sage, 1992, pp. 93–109.

21.

Birks

Chapman

Francis

. Memoing in qualitative research. J Res Nurs 2008; 13: 68–75.

22.

Is your staffing ratio appropriate for your case management model? Hosp Case Manag 2015; 23: 25–28.

23.

Lynch

Wajnberg

Jervis

. Implementation science workshop: a novel multidisciplinary primary care program to improve care and outcomes for super-utilizers. J Gen Intern Med. Epub ahead of print 28 March 2016.

24.

Xing

Goehring

Mancuso

. Care coordination program for Washington State Medicaid enrollees reduced inpatient hospital costs. Health Aff 2015; 34: 653–661.

25.

Roberts

Crigler

Ramirez

. Working with socially and medically complex patients: when care transitions are circular, overlapping, and continual rather than linear and finite. J Healthc Qual 2015; 37: 245–265.

26.

Shommu

Ahmed

Rumana

. What is the scope of improving immigrant and ethnic minority healthcare using community navigators: a systematic scoping review. Int J Equity Health 2016; 15: 6.

27.

Genoff

Zaballa

Gany

. Navigating language barriers: a systematic review of patient navigators' impact on cancer screening for limited English proficient patients. J Gen Intern Med 2016; 31: 426–434.

28.

Bodenheimer

. Coordinating care—a perilous journey through the health care system. N Engl J Med 2008; 358: 1064–1071.

29.

Knai

Nolte

Conklin

. The underlying challenges of coordination of chronic care across Europe. Int J Care Coord 2014; 17: 83–92. http://icp.sagepub.com/lookup/doi/10.1177/2053434514556686 (accessed 21 October 2016).

30.

Berwick

. Health services research, medicare, and medicaid: a deep bow and a rechartered agenda. Milbank Q 2015; 93: 659–662.