Stephen Rosenthal: Developing an accountable care organisation to deliver population health

This content relates to the following topics:

Stephen Rosenthal, Senior Vice President, Population Health Management, Montefiore Health System shares lessons on developing an effective accountable care organisation.

This presentation was filmed at our conference on Sustainability and transformation plans: moving towards implementation on 24 May 2017.


There are so many similarities between the challenges that we are both facing. And at least what I have learned in the last day or so sitting and speaking with many of the individuals who are in this room and on this table, and perhaps give you an opportunity to at least see how 21 years has evolved and give you a sense that this is a journey.  And although we’ve been doing this for 21 years, we’re nowhere near perfect, but yet we’ve had lots of experiences that perhaps in times like this we can share.  So I will just quickly share with you the Bronx which is the home of the New York Yankees, also the Botanical Gardens and the Bronx Zoo.  So there’s a lot going on in the Bronx that has nothing to do with patients or people, but there are lots of activities that kind of the history of the Bronx has always been a place where immigrants have come.  So it’s a very poor community, it’s actually the poorest community in the United States.  The interesting part is that it’s sandwiched in between the two wealthiest communities in the United States, New York City and West Chester County.  There’s as you see a million four, very poor in general, huge chronic disease, and to respond to that, over the many years, Montefiore has built a very broad network that touches some eight counties.  So there are ten hospitals, a nursing home, and a whole variety of community based programmes focusing on primary care, but yet at the same time mental health, substance abuse being a critical component.

As our population is mostly government programmes, about 80% of the patients we see either are medicated or Medicare medicated in the United States is a programme for folks who are essentially under the poverty level, and Medicare is a programme largely for the elderly. But a third of the Medicare beneficiaries are actually under the age of 65, these are individuals who have things like End-stage renal disease who have had transplants and who have significant disabilities falling under that programme as well.  And of course we also maintain a very large specialty network of providers throughout all those communities.  The punch line of our 21 years or so of doing this, to kind of cut to the chase so to speak at the very beginning, is that you must begin with an overarching vision.  You really have to know where you’re going and what the opportunities are.  And there has to be a very clear governance structure.  And I’ve been listening to the challenges that you’re all facing in the various communities with the STPs and all of the challenges of the past coming to play in the future discussions, and so governance becomes a key piece because the whole concept around ACO is accountability, and without governance it is very difficult to have broad accountability.

And then the operations has to be in alignment. You have to look at that network that you’re providing and the population within that network, and begin to think about all the providers that are providing services within that community and begin to then understand and think about ways of creating incentives for them to do the things that make the most sense.  And what you have that provider network you then have an understanding of the population that you’re touching.  But we don’t have the benefit of a National Health programme, but because we’re disproportionally government programmes, it’s almost as though our community of providers and system is disproportionately a government programme or a single payer.

So once you understand that population, the other thing to recognise, as I’m sure many of you know, is that the parietal rule prevails. 20% of the population is generally going to be running the majority of the cost.  And developing the care and programmes around that, we talk a little bit about our centralised hub and spoke programme, I’ll talk a little bit about that, and ongoing management of that operation and how that connects with technology, because technology is going to be our freedom in the future and the lift from all of the burdens that are currently taking place within our practice offices will begin to lift as technology begins to replace some of that noise level.

So our governance structure, I will just talk a moment about that, begins with the Monterfiore IPA structure. And that governance is in equal balance between the providers in the community, the employed providers, as well as the institutions, and that’s the entity that bears the financial risk in all of our models.  And as you can see we have over 4,000 providers that are in that community touching some 400,000 plus individuals.  And then the ability to manage those relationships between the payers, the governments, the providers, and the patient relationships, the community activities, we created the care management organisation or company, which is what I run, that essentially in many ways operates similar to an insurance company, has all of the infrastructure around that, but it’s goal is to really manage the relationship of the patient and the various providers that are in those communities that these patients live in.  And develop the kinds of programmes that will actually ultimately improve their overall care.

Just a quick snapshot of our portfolio. We’re managing a little over 400,000 lives.  Our nomenclature global risk means that we’re getting the majority of the premium for those individuals, and as a result we’re responsible for paying for all of their care, so that’s 221,000 individuals.  We then have a programme called ‘Shared risk’ which is kind of like virtual capitation.  We look back on history for the population that we’re servicing to determine what their average cost was, and on a going forward basis we have to bend the curve.  We have to save money relative to that historic cost, plus a trend factor, and whatever dollars we save, we share with the insurance company, or in the instance of the next generation ACO, with the government.  That’s a shared savings programme with the government.  With the State of New York we developed a programme for care management for high cost Medicade beneficiaries, and of course we keep on trying to expand that population.  We did a programme with the government called the pioneer ACO, that was a demonstration that said for everybody that’s in the Medicare programme, who touches our doctors, will be financially at risk for their total cost of care.  And we did that programme for five years and we saved the government about $70 million over about a $2.2 billion spend over five years, and we got to keep about $34 million of that, a little less than half, or a little more than half, yes about half, how do you like that?

And we’re doing the same thing with the State on the Medicade population, where the State is not necessarily prepaying us as an insurance company, but they’re paying us to change the model of care and come up with different parameters for how we measure success and the ways that we do programmes with the various hospitals and the communities that we’re participating with in this district or state programme. So just quickly to give you an idea of one of the things that I’ve generally been an advocate of is the creation of a standardised approach or a systematic approach for managing your population, because otherwise you don’t have the ability to scale up, or scale down, you don’t have the ability to understand whether you’re successful or not.  And of course you heard earlier at the meeting about measuring your opportunities, you have to be able to measure what you’re actually trying to do.

So we believe the patient is in front of that and is the primary focus here, and we identify and prioritise our patients in a number of different ways. Once we identify who they are, we have to go through a process of telling them why we’ve identified them and whether they actually believe they have the problems that we think they do.  And then they’re much more willing to go through a fairly detailed assessment so we can get a very clear understanding of what their challenges are, and then develop a care plan that’s special for them and then monitor that programme over time.  And we do this in an aggregate basis, but one patient at a time.  So I love this slide, isn’t this cool?  I wish I knew how they did that.

But this is kind of our visual model for the stratification process where the lobe folks, the ones who have very low intensity, they’re us, we’re the as generalised healthy population, we call ourselves the ‘Worried well’, you know? We wake up, we know that we can go to Web MD or call a friend and deal with it.  Then there’s the group of us who actually have chronic illness but we’re managing it really well.  We go to work every day, we take our medication, we touch base.  But then there’s the high utilisers, the ones who truly are problematic, those are the ones that we provide intensive coverage for.  But we’ve learned over time that it’s more than just our clinical issues, that just the big data isn’t enough.  We had to recognise that many of our patients, particularly the ones who are poor, who have challenge, that unstable housing is a measure of social challenge.  In addition to that you’ve got obviously substance abuse, mental illness, and this just the financial distress of the times.  Because none of our governments or our healthcare system being challenged financially, but the individuals who are living in those communities are all being financially challenged.

So we looked at the social determinates of health and we found that in all instances they raised the cost of care. And so focusing on them gives us an opportunity to lower the cost of care.  We know that people who have substance abuse, and have an underlying medical condition, are ten times almost more likely to spend healthcare dollars than someone who doesn’t have that substance abuse challenge, but those same medical challenges.  And we’ve looked at that across many subsets of the population.  But clearly housing is a challenge, and I know many of your communities have housing issues.  Although it looks like a small number in terms of care, so many individuals are challenged by the unstableness of their housing environment.  Understanding that becomes important, because as you develop your care management programmes, we’ve had to develop these wraparound services, and I’ve learned that many of your communities have these programmes.  Where community based organisations are helping you manage the complexities of the populations that you’re servicing.  And some of them have to do with dealing with homelessness, shelter and all of the various other legal challenges or financial challenges that they’re facing.

Mental health is a very big challenge for us, and as a result we’ve developed a number of programmes to focus on that population, recognising that an intervention around that segment of the population, has a multiplier effect in terms of the total cost. So we go out of our way to think through and develop programmes in the community that actually manage and help individuals with mental challenges.  One of the programmes that we developed with some funding from the Federal government in the State of New York, is a programme we call ‘Synergy’ but it was to test whether every area of communication can have an impact on whether individuals are engaged or not when they have mental illness or mental challenges.  So we engage the patient by providing in all of our primary care settings mental health specialists at the point of care, so that when the primary care physician needs assistance, there’s that assistance available, not necessarily for ongoing care, but for the establishment of the next phase of care.  And that individual is often a clinical psychiatric social worker, or a psychologist, is linked up to a psychiatrist for additional support, whether it be medication or not.  But the communication methodologies could be anything, and I know many of you are involved in telehealth, it could be visual, it could be auditory, but giving them multiple options often allows you to find an opportunity to engage them on a regular basis.

I share this series of slides again with you to kind of summarise the development cycle that I talked about in terms of the care management process where in the end we have an interdisciplinary team that’s focused on managing that subset of the population. But I did that in large part to kind of illustrate in the next slide that in each of the areas that there are some skilled individuals that are needed to help manage those subsets or that continuum of services.  The analyst obviously looking at data is a clear understanding in terms of the role and type of individual for that.  But when we talk about enrolment, we use often just community participants, people who live in the community, who understand the challenges of those individuals, so that they can actually engage the patient and make them feel comfortable that this is an opportunity for them to get help and there’s no reason why they shouldn’t engage, and we have a 97% success rate with engaging those individuals and process.

Then in terms of actually developing the assessment, we often use a skilled and/or trained motivational interviewed person, it could be a nurse, but it doesn’t have to be, it could be an LPN or a social worker. And then for those who actually develop the plan, those tend to be clinical individuals like a nurse, or a highly clinical skilled social worker, to actually work on and develop the plan with the patient, their family and the caregiver at the same time.  And then of course in the end it’s a multidisciplinary approach that involves not only the clinical individuals caring for the patient, but all of those other support mechanisms, the Pharm D, there are pharmacists I know in the room who didn’t bring any medication with them today although I had asked them to.  And also experts in dealing with end of life issues, which is often a real challenge in certain ethnic communities as well because of just the translation of having to prepare for the end of someone’s life.

The systematic approach so that one could scale up in terms of staffing and developing it, we created a pod system that allows us to grow and/or decline based on the need of the population that we’re servicing. And then we’ve recognised that it all begins in the emergency room and we want to make sure that when they transition from one side of care to another that we’re managing that transition, and for those that are truly challenged we want to provide an intensive case management.  And it’s this triad that we’ve developed as a model that allows us to be replicable and at the same time have a very high interrelated reliability, so that when we move patients and/or providers from one setting to the next, we know exactly what to expect in terms of our outcome.

Technology becomes a critical piece to this. And as you know EMRs and analytics, patient portals, where patients will have much more opportunity to exchange information with their providers and support health, telehealth, and of course you need to know what’s going on and so we’ve created a number of different dashboards to begin to look at opportunities to understand whether you’re being successful in managing the population.  Because again a lot of our dollars are based on incentive earnings, so that the providers benefit as well as...we call providers individual doctors, I know you guys call providers more broadly, but hospitals and systems all participate in the financial advantages.  And these are all demonstrated, and you saw in the brochure that’s out there, some of the data associated with the populations that we service, of course we want to change some of these measures, lower them, which translates into financial savings, particularly in populations where you know you have chronic disease and where you’re looking at efficiencies like emergency room use, preventable admissions, as well as preventable readmissions.

And of course because we’ve been doing this for 20 something years, we’ve become very important in our different communities around us to help support those organisations that have begun to see more and more government programme patients within their delivery system. And I will end there.


Dr Umesh Prabhu

Medical Director for 15 years,
Comment date
13 June 2017
Well done. Today in our NHS there is no clear accountability for leaders, managers and also for some very senior consultants and GPs in our NHS and also to some senior nurses. Good news is 80% of NHS staff are simply wonderful but sadly NHS culture of bullying, harassment, victimisation, blaming, shaming and naming and club culture means there is different accountability for different people. Whistleblowers and BMEs are severely punished and sadly many bullying doctors, nurses, managers and in some cases Board members escape and keep on moving from one Trust to another or stay where they are and create corrosive energy and poor team working and this puts patients, staff and NHS at risk.

Even CQC Inspection has failed to expose some of the Bullying culture by the Board members! It is simply shocking.

STP must appoint right leaders and create right team and there must be clear accountability for all and most important good governance at every Trust, every CCG and every NHS Institution and most important the commissioning must be for safety, quality, outcomes, staff and patient feedback.

Good to see this event at last. It is accountability and good governance missing from our NHS and in a culture of bullying patients, staff and NHS suffers.

Add your comment