Fundamental aspects of coordinated care are applicable across many fields of chronic disease, and health systems that want to develop a coordinated approach for care of psoriatic conditions may benefit by reviewing management strategies employed by other health care sectors. The following interview branches into various sectors to glean insights on how they approach care coordination and the patient and organizational priorities they have established for optimized care. AJMC® interviewed experts from various organizations.
Maria Basso-Lipani, LCSW, Vice President, Care Management for Population Health Mount Sinai Health System
Mount Sinai Health System (MSHS) is an integrated health care system in New York City that includes the Icahn School of Medicine at Mount Sinai, 8 hospital campuses, and regional ambulatory services. Basso-Lipani oversees a clinically integrated care management program at MSHS that targets patients with any diagnosis who are identified as high- or rising-risk through the electronic health record (EHR) system, analytics platform, or provider referrals (eg, those with recent hospital discharges, multiple comorbidities, or psychosocial concerns). The program’s multidisciplinary team approach employs more than 100 direct staff social workers, nurses, and care coordinators to optimize patient care.
Lynne H. Milgram, MD, MBA, Chief Medical Officer, Sharp Community Medical Group
Sharp Community Medical Group (SCMG) is a payer provider group, comprised of a network of more than 800 primary care and specialty physicians representing more than 30 specialties across 350 locations in San Diego County. SCMG is a part of 2 different hospital integrated delivery networks. In the last year (during COVID-19) SCMG was able to get the designation of clinically integrated network and they are now contracting on behalf of all of their providers because they can demonstrate clinical integration with other payers for other types of products. Milgram supervises the medical and associate medical directors and the pharmacy department at SCMG. In her interview, Milgram shared that treating the entire patient, not the disease, is a primary theme in the group. With a recent appointment for clinical integration status, SCMG has established care coordination specifics for psoriasis (PsO) and psoriatic arthritis (PsA) regarding coding opportunities. The Healthcare Common Procedure Coding System (HCPCS) is used in combination with the ICD-10 to ensure PsO and PsA patients are being comprehensively managed, particularly regarding potential comorbid concerns with PsO and PsA.
Emily Barey, MSN, Vice President, and Brian Bentley, Ongoing Services Population Health Specialist, Epic
Epic EHR software is currently used to house more than 250 million patient health records globally. In the program, Epic was used as a support tool for identifying and precisely engaging high- and rising-risk patients. Using a multichannel approach, organizations have the ability to outreach to patients through messaging via the patient portal application MyChart, mail, and telephone outreach, including third party text integration. Epic assesses considerations such as social factors, health care utilization, age, and clinical complexity to identify rising-risk patients as a target for care management. Care managers can then be assigned to patients in a prioritized way and based on need. This also enables early identification of patients who may benefit from a social worker, pharmacist, or community health worker and establishes a team-based approach to care management. Barey is responsible for product management in the areas of nursing, community health and social care.She leads the Epic Nursing Advisory Council, consults with Epic customers on strategies for comprehensive care record adoption and collaborates closely with the Epic research and development teams on enhancements to support interdisciplinary practice across the continuum of care and service. Bentley is a member of Epic’s population health team where he provides expertise in designing and implementing data and analytics tools to improve population health management initiatives.
AJMC®: In an effective care coordination program, how do you identify patients?
Basso-Lipani: We leverage several major pathways into care management. The first is through our analytics. We marry EMR data and claims data and several analytics tools to find a group of people with multiple chronic conditions who are likely to have psychosocial stressors. We also use analytics to help us identify who is currently hospitalized and at risk for a readmission. Lastly, we take referrals from health system colleagues at each of our hospital campuses and from the 5000 primary care doctors and specialists who comprise our clinically integrated network. Whenever possible, we go out alongside our colleagues that help manage those relationships to discuss the practices’ patients whom we think can benefit from our program according to our analytics. However, we don’t stop at analytics. We also want to hear about the patients they are most worried about and who keeps them up at night. The insights from providers on the ground are extremely valuable to us and really help us round out our approach to patient identification.
Our patient identification methodology has evolved considerably in the past 11 years. Previously, our target group was primarily current hospitalized patients who had frequent admissions. With the help of analytics and our relationships with practice leaders, we’re able to do much more around predicting who is likely to need our help and proactively outreaching to them.
AJMC®: How do you identify high-risk patients (ie, cost, quality of life, etc)?
Barey: Historically, care managers have only focused on the highest risk patients, and now they’ve also started to focus on patients in the rising-risk category. These care managers are increasingly embedded in primary care offices to focus on this population with the PCP [primary care provider] working side-by-side. This model helps the centralized care management team focus on the top 1% to 5% riskiest patients who have the most complex needs or chronic cases. The way that they’re stratifying these patients by risk level is by using predictive models within Epic, or models that they’ve designed themselves, that look at everything from utilization data, primary diagnosis information, how many medications, types of medication, social factors, age, payer—you name it. These models are able to draw from a big data set. I think the most common ones are looking at risk of ED utilization and risk of readmission. Not just the risk of being readmitted to the hospital within a certain timeframe, but across the board looking at “what is this patient’s risk of ED or inpatient utilization?” After that, a number of our groups certainly are looking at care gaps and HCC [hierarchical condition category] scores as another way to prioritize patient outreach, not only as a prevention strategy, but also because there is a financial incentive so the alignment of prevention and payment is a win-win.
Getting the PCPs and an embedded care manager involved much more proactively for the rising-risk population is important because, ultimately, they’re in a better position to make a difference with patients due to their proximity to the patient, often existing relationship with the patient, and their ability to take action on best practice recommendations for prevention and disease management versus a remote, centralized care manager alone. The care manager can make recommendations, but at the end of the day most of the follow-up care for this population still requires a provider’s signature. With the shift to value-based payment, everybody knows they need to be working differently, and they’re starting to work differently, even if they’re not getting paid. Incentives are starting to align, but I think that’s been a really tough spot for some of our groups who recognize the need to work differently, but there isn’t always a way of getting reimbursed to support new models of care.
AJMC®: Among patients that are in your program, is there any stratification to ensure those patients should really be in the program?
Basso-Lipani: We rely primarily on our analytics tools to help us prioritize those within our target group who are at the highest risk and need our help. The primary tool we use, the Johns Hopkins Adjusted Clinical Groups (ACG) System, is very person-focused. [The ACG system is a population/patient case-mix adjustment system that measures health status by grouping diagnoses into clinically cogent groups. The goal of the ACG system is to assign each individual a single, mutually exclusive ACG value, which is a relative measure of the individual’s expected or actual consumption of health services]. It doesn’t look at disease state alone, but at a collection of conditions. This approach helps systems like ours to understand the health of our patients over time, which we think strengthens the predictions around who will have high-cost care tomorrow and who is most likely to have persistent high-cost care. We have embedded this prioritization functionality into Epic, which allows us to sort the target group and assign the patients from within our electronic medical record system.
AJMC®: When your case managers are looking at codes, strictly thinking about psoriasis, what percent of the time are they flagging a biologic because they want to make sure it’s being utilized appropriately?
Milgram: I don’t think they’re doing that quite often at all because we start up front with prior authorization, so we’re making that decision really right up front. Occasionally, I’ll hear something that makes me think, “Why isn’t the patient on a biologic,” but not the opposite because in our state, it’s required to have a prior authorization anyway.
If I really did feel that our utilization was over or under in treatment for plaque psoriasis or psoriatic arthritis, I could mine the data for that. I could extract my patients with psoriasis or psoriatic arthritis, as long as they are coded correctly. That’s what we do with our rheumatoid arthritis metric. The rheumatoid arthritis metric measures the use of a DMARD before a biologic. Even in psoriatic arthritis, there is a step-up way of treating it. The biggest problem is being miscoded or over-coded. You can’t correct the code unless the patient is seen by a rheumatologist and recoded.
AJMC®: Within a multidisciplinary coordinated health care team, what are the roles and responsibilities of each of the members?
Basso-Lipani: Our multidisciplinary team is one of our core strengths. The nurses on our team focus on disease management through education and symptom support. They are there to enhance the clinical care our patients receive in the ambulatory, inpatient, and congregate settings, and to bolster the patient’s confidence in managing their health. The pharmacists work closely with the nurses primarily around medication titration to support better adherence. This often includes addressing symptoms that are troublesome to patients and interfere with functioning. If changes to the dose or to the medication itself can support higher quality of life, we need to do this at every turn. Our social workers are focused on identifying the psychosocial barriers to good health for each patient and then working to ameliorate them. If the patient is not taking the prescribed medications, the social worker will search for the reason why and then, in partnership with the patient, create and execute an action plan to remove those barriers. For example, it could be that the medications are too expensive or the adverse effects too strong. It could be that the patient didn’t trust the provider who prescribed them. By reaching for the patient’s experiences of illness and care as well as understanding their life circumstances, the social workers form a global picture of what is getting in the way of health. A whole host of issues can emerge, from family caregiving to a cultural misalignment with the PCP that impacts connection. Whatever barriers are identified, it is up to the patient to help define where they want to start.
Finally, our care coordinators help to ensure that the nurses, pharmacists, and social workers are all working to the top of their licenses by helping patients navigate through our system and successfully link to resources. What’s great is that we have wired this interdisciplinary approach into our electronic medical record so that it doesn’t matter which member of the care management team touches the patient first. Patients whose first outreach is a pharmacist will be screened for needs that the social worker can address and vice versa in each direction. We think of it as having multiple front doors to care management, and we assure that whichever door a patient comes through, he or she will end up with the right blend of services. Some patients may have all four of those disciplines involved while others may work mostly with one discipline.
AJMC®: Can you talk a little bit about the problem solving involved in coordinate care? The process from the social worker, from the case manager, then through pharmacy, and on to the provider perspective?
Basso-Lipani: I think that what you’re hitting on, which is definitely a piece of our model, is really sticking with the patient through the journey. Let’s say that the patient with psoriatic arthritis didn’t have other conditions and hadn’t been utilizing the hospital, they may not show up in our analytics. That means that we also wouldn’t see them upon discharge because they’re not discharging. I think one thing that’s good about the way that we approach patient identification is that we have created that pathway from the physician, so from the physician’s office. That’s how we might find that patient—by referral from their physician. One of the major stumbling blocks to coordinating care efficiently is that the right hand often doesn’t know what the left hand is doing. This is where we rely on our electronic medical record, Epic. We’ve invested resources in training various providers so they understand how to identify whether a patient has a care manager, who that person is, what he/she is working on with the patient, how to reach him/her, and also how to refer if care management is not yet engaged but the provider thinks they should be. This helps us avoid duplication in services while also encouraging collaboration. We want to do more problem solving with our health system partners at critical decision points, such as whether or not home care or an inpatient admission is most appropriate. At these times, it might be possible to flex care management resources rather than to default to more expensive types of utilization. Often, the solution is to gather on a call, discussing one patient at a time, to figure that out. We are working our muscles in this area quite a bit right now, as the teams in our EDs and inpatient floors become more aware of care management and as we simultaneously lean on system leadership to set the expectations around shared ownership of outcomes and the value of thinking together.
Let’s say that the patient with psoriatic arthritis didn’t have other conditions and hadn’t been utilizing the hospital, they may not show up in our analytics. That means that we also wouldn’t see them upon discharge because they’re not discharging. I think one thing that’s good about the way that we approach patient identification is that we have created that pathway from the physician, so from the physician’s office. That’s how we might find that patient—by referral from their physician.—Maria Basso-Lipani, LCSW
Milgram: We’ve been very fortunate to have an outstanding care coordination and care management program. We focus across the continuum of care, from outpatient to inpatient to post discharge follow-up and provide pharmacy support as well for chronic care management and medication reconciliation. Our referrals into the case management originate from hospital case managers, patients, caregivers, specialists and PCPs. We stratify our population with a homegrown predictive modeling tool. We mine our data for patients that frequent the emergency room, high-volume or high-cost diagnoses, behavioral health, et cetera. There is robust communication between patient, case manager, physician, and other key external partners as needed. We work with our physicians when we identify a need for physical therapy or DME and assist in ordering. We collaborate with many of our physicians that allow us to order the services patients need. With our current EMR, we can task physicians to communicate updates and needs.
PATIENT AND PROVIDER ENGAGEMENT
AJMC®: How does patient engagement affect quality of care?
Basso-Lipani: Patient engagement is by no means the only factor affecting quality of care, but it is certainly an important one. I mentioned our comprehensive psychosocial assessment earlier and it fits here as well. Our team reaches for an understanding of what is getting in the way of health, and the answers almost always reveal a lot about the patient’s experience of care, which can have a tremendous impact on engagement in care. Sometimes it’s a feeling of being alone in a sea of appointments to attend and pills to take. They don’t have the right social supports or maybe they are tired of being sick or had a negative experience with a provider. Any one of these circumstances can lead a patient to stop showing up, to postpone the tests and treatments or to seek care at another system leading to care fragmentation. If that happens, the quality of care they are receiving suffers. Other factors affecting quality of care are structural, such as unacceptably long wait times for critical appointments or the habit of a provider to substitute basic Spanish for a certified interpreter whose services may elongate the visit. Most of the patients we encounter have positive feelings about our health system, but we also encounter those who have been disappointed in some way. Sometimes just listening to that story and authentically acknowledging for a patient where our health system got it wrong can be a very powerful intervention. I’ve seen it restore hope that a different experience is possible moving forward and really ignite engagement in care. As both ambassadors of our health system and natural patient advocates, our care managers are in a unique position to improve the experience of care, which is an essential part of delivering quality.
Barey: We’re seeing this paradigm shift towards precision engagement. For example, and this might fit in very well with the rheumatoid arthritis example, you might have a chronic disease management program today. Frequently it’s in isolation from the care continuum; it’s one channel. It’s a phone call, maybe a letter. And it is implemented in a similar way regardless of disease or the whole person. Kaiser Washington, for example, recognized that their standard care transitions program wasn’t working as well for patients with behavioral health-related admissions. So they remodeled it to have a specific care transitions program for the mental health continuum. That’s been an interesting trend to see; let’s take the standard care management program, but let’s get much more population, disease, person-specific so that we can really understand the needs of that population uniquely and adjust our outreach, care coordination, and intervention approach. If I’m thinking about that arthritis case, I’m sure there are things in there that have nothing to do with what a [congestive heart failure], a diabetic, or a mental health [patient] would need. You can take those programs and really make them much more precise and engagement oriented.
I think the other theme along those same lines is multichannel. No longer do you just make an outreach phone call and hope for the best. Our most successful customers are sending text messages; they’re doing video visits; they’re doing home visits. And they’re developing new evidence of how to titrate outreach efficiently and effectively for different [patients]. There’s really this sense of meeting the person where they are. I think that’s also really shifted again from just a blunt approach of “Oh, I call you once a week for 4 weeks post discharge and hope that you don’t get readmitted.” I think that falls under this larger heading of the whole person approach. Again, thinking about precision engagement, and the sites that are also assessing for social determinants of health are equally the sites that are doing this work. [Many] of our organizations have been using their care management implementations to step back and say, “You know what? We should have a standardized approach to our programs, to how we document and manage the case, and how we measure outcomes.” Despite it being led by clinicians for how many years, you would think there’d be a bit more rigor, but people have been confessing more and more that there’s a lot of variability in how they’re doing it, and frankly missed opportunities for leveraging each other at each point of care. A standard way of assessing and making a handoff and leveraging documentation saved so much time in getting people enrolled in the right program and then getting them the right plan more quickly. I think we’re at an advantage to some extent because of the integrated nature of Epic, so it makes it a little easier potentially than maybe organizations who are using different niche systems. Certainly, that’s what our groups have been looking to do, even within operations and then having the EHR try to help hardwire that. Along those same lines, is working more and more upstream, and this might play well too with your arthritis example which is really having the care managers look at their data and understand the patterns.
AJMC®: What is that initial approach for finding out the patient’s priorities and their goals for treatment, and then how do you build that into the care plan?
Basso-Lipani: Building the care plan begins with engaging the patient. We do a great deal of training with our staff around how to engage a patient in such a way that they feel they have a partner in care, that they feel safe to tell us the truth about what isn’t working, what worries them, what stressors they are juggling that are getting in the way. Our skilled care managers approach patients with a genuine interest in the story of who the patient is and how they’re doing. They know the assessment domains like the back of their hand so the assessment doesn’t need to be linear. It’s not a checklist moment, but an organic conversation that typically concludes with a number of areas in which we can offer support and partnership. From there it is up to the patient where they want to place our collective efforts. We advise, but the decisions around goals are theirs. The care plan defines how we’ll get there and it emerges through more discussion around what they need from us and what they think they can tackle on their own.
AJMC®: How are you really engaging with these providers and educating them on which patients would be good candidates for your program?
Basso-Lipani: This is a big area of focus right now for us on 2 fronts—the inpatient hospital teams who are in a position to refer currently hospitalized patients who need more longitudinal support following discharge, and the PCPs and specialists in our clinically integrated network that can help us spot patients that our algorithms missed. With the inpatient teams, we are really making a big push with the leadership, some of whom are new. We want them to really understand our services and to get into the habit of thinking far beyond the discharge as they are busy planning it. What is this patient likely to need next week or next month and is that something care management can provide? If the answer is yes, then we want that referral before discharge so we can be that bridge for the patient. With respect to the providers in our clinically integrated network, it’s equally important that they understand our service offerings but also which patients we are targeting on their panels and why. We want them to know that even if they never make a referral, we are still delivering value to them. They can increase that value by getting into the habit of sending us the subgroup of patients they encounter who really need added support to sustain.
AJMC®: Are there any initiatives or technology that would target the patient to let them know about services that they could engage with their provider about?
Basso-Lipani: Yes; [there are] a couple of ways we do that, which are new this year. Our analytics uses our ACG tool to recommend about 1000 patients on a monthly basis who may be right for care management. We are able to use Epic, our EMR, to bulk assign those patients to our care managers. Simultaneously, we ensure that any patient on that list that has a cell phone and can receive texts, gets a text message from us that alerts them that they are eligible for care management and gives them a description of what we can do for them. We’ve leveraged a lot of the technology that came to the forefront during COVID-19, including a texting interface that allows a patient to access support in near real time. Through this technology, we can ask patients, “What is most challenging for you? What do you need help with?” And based on their answers, we can then say, “Would you like us to connect you to resources for that right now?” (and they’ll self-serve) or “Do you want to speak to a care manager?” We are planning the same kind of communication to our target patients via Epic MyChart. We will soon be able to embed some assessment questions that will help us understand their degree of risk, and their answers will actually display directly within Epic’s Social Care wheel.
OUTCOMES AND ALGORITHMS
AJMC®: What outcomes have you been looking at specifically within the patients that are enrolled in your program?
Basso-Lipani: We regularly look at a group of key performance indicators. These include enrollment volume, conversion rates (outreached to enrolled), the number of rapid follow-up appointments with ambulatory-based providers we can schedule following an admission, the show rate at those appointments, and 30-day readmission rates among patients enrolled in care management as compared with those who were eligible but didn’t enroll and those who were not eligible. Additionally, our analytics team leverages various tools to help us risk-stratify our enrolled patients and produce an expected spend, which we then plan to compare to what is actually observed. Through each of these efforts we are endeavoring to understand the impact of our work on contract performance. This will be the most sophisticated impact analysis we’ve ever had in care management and aligns with the work of many in our analytics group around forecasting.
Barey: Particularly with high utilizer groups, everyone is trying to get quality care but also shift to value based payment. They’re really focused on determining the right model to manage risk, and I think that maybe that would be a good transition. We’ve seen a lot of really interesting models related to the care team in a much bigger interdisciplinary team focus, but something really unique that Queens is doing is they now triage each case and decide whether it should be a nurse or social worker as the primary case owner. From a culture perspective, one of the biggest things that has changed is around physician engagement, particularly with using social determinants of health and being accountable for readmissions or outcomes long past them leaving your office or hospital. Culturally, it is a big shift to be thinking about an entire continuum, when we just historically have been so transactionally focused and we’ve kind of left it to the case managers to pull it all together. I think there’s definitely a cultural shift of we’re all on case management on some level. That’s been kind of exciting to see. I think most… who I’ve worked with have focused on the obvious heavy hitters in readmissions and the obvious high-utilizer groups. Although interestingly enough, rather than having a disease-centric approach, more and more taking this whole-person approach, and saying, it doesn’t really matter what the constellation of risk factors is or what the constellation of comorbidities is. As soon as you hit this threshold for utilization or as soon as you hit this threshold for risk, we’re going to assign someone and make an outreach. Then that person is now again more whole-person-centric as opposed to just [doing] your diabetes part or I just do your rheumatoid arthritis part. It’s interesting to think in that context about using some of those codes, which we certainly could as another kind of contributing factor for ruling in how you might look for high risk.
Bentley: The cutting-edge groups are the ones that are very deep in the data, and they’re looking for trends in outcomes. Then they’re tweaking the upstream workflows to either change the approach and/or update algorithms that prioritize populations with a chance to influence and really make improvements in outcomes. In many cases, we have some standard algorithms for that, but in some cases, customers are ahead of the curve. They’re using their own data analysis and Epic to drive
AJMC®: Do you have off-the-shelf programs built for case management?
Barey: Yes, we do. I would say the library right now is focused mostly on care transitions and mostly on chronic [conditions]. I think those are probably the highest-volume programs that are implemented, and we have what’s called a starter set in our Foundation System. We have the case and the targets and the tasks, an initial set of assessments, and a couple examples of care plans to start with.
AJMC®: How do you evaluate a need for changes to the program?
Basso-Lipani: We use a combination of data and direct staff feedback to identify changes that are needed within the program. On the data front, we’ve made huge investments of time and resources in 2021 to deepen our understanding of staff alignment to patient opportunity, as well as overall staff performance. These investments are beginning to bear fruit as we now have actionable information that supports our staffing decisions and even helps us understand when we have a performance issue that requires training and/or enhanced supervision. At the same time, staff feedback is very important. With a large team like mine, I rely on communication with various levels of staff to ensure I’ve got a pulse on how the team is doing. I have regular meetings with directors who have a bird’s eye view and also with managers who are closer to the direct care. To connect more directly with frontline staff, I launched two initiatives a few years ago. The first is a monthly meeting with a self-nominated group of staff who reflect our various disciplines. Each month, this cohort surfaces concerns brought forward by their peers and we brainstorm solutions in real time that are then disseminated or worked on further with the managers. The other initiative leverages technology and is essentially an “open ticket” that enables any staff member to raise something they’ve noticed that day that we could do better. These tickets are collected and summarized daily for me so that we can act on them.
AJMC®: What would you say to an organization that wanted to start a program like this? Where do they start?
Basso-Lipani: The good news is that you can start anywhere. For me, the Triple Aim is still truly the North Star. Reducing cost and improving health outcomes take time and require data. However, evaluating the patient’s experience of care and taking steps to make it better can start tomorrow. So, what I would say to an organization just starting out is that they should build their population health strategy around the patient experience and watch the ripple effect of that action on the other two aims. Hire people who want to understand the patient’s perspective on care and train them to spot what is in the way of optimizing it. Then give them the latitude to address those problems and support them to do that work at the top of their licenses. We are in a very crowded landscape in New York City. There are a lot of health systems. Population health management works best when people stay under your tent, but we have to give people reasons to stay. Our genuine interest in the unique challenges our patients face and our commitment to improving their circumstances are among those reasons.
Barey: You’ve got to typically start with the data. We have a self-service data tool called Slicer Dicer which allows you to start exploring your hunches or exploring patterns. More and more we’re seeing our organizations give those analytics tools to clinicians so that they don’t have to request reports from the IT department. I’m really curious about “What are the zip codes for my readmits?” Or “How many of these patients have had a certain specific medication like in this case?” Exploring the data is the first step to try to find out what a population has in common. From there, you can determine the right interventions, or different interventions you could try with different groups. I have aggregate claims data. I have EHR data. I have data from other EHRs via interoperability. It is no small feat to even just get the data aggregated in a way that is valid.
AJMC®: What are still some existing challenges that you’re trying to address?
Basso-Lipani: I think with 500,000 lives in attribution, you’re really hoping that you’re not just doing well with the patients you’re serving but that you’re also serving the patients who need you most. Time will tell as our impact analysis becomes more sophisticated, and we’ve set up processes for course correction based on what we learn. Another ongoing challenge is awareness of our involvement with a patient and what we’re working on. The visibility of Epic’s Social Care wheel will help considerably with this but requires providers to behave differently than they do now in order to access that information. Ongoing education on how to do this and understanding the value of these critical tools will hopefully gain traction over time.
Barey: There are a lot of proxies out there that people are using, but I would say that another request I get quite a bit is, “Can I put an acuity or a working weight on this case to better leverage the team I have?” That’s definitely an area I think of future work for us. A lot of our groups have again done proxies on their own but not anything that we’ve developed a specific model for yet. I suspect we will get there. The only other one was the CBO [community-based organization] leveraging. I think all of our leaders are not only assessing social determinants, but they do have the community partnerships and they are trying to leverage community-based organizations. They will all tell you that they’ve overwhelmed them. That there’s just too much unmet need and there isn’t a payment mechanism to augment the funding from the CBOs. They’re anxious to see how that might change in a value-based world as well from a Medicare perspective in particular.