Breaking Down Barriers: Understanding the Impact of Social Determinants of Health (Part 2)

Ryan Bosch, MD, President, Chief Health Information Officer, Co-Founder, Socially Determined

Welcome back to PopHealth Perspectives, a conversation with the Population Health Learning Network where we combine expert commentary and exclusive insight into key issues in population health management and more. 

Bosch: Diseases discriminate. They do. They have for a long time. And so, to recognize that certain diseases, their presence — diabetes, metabolic syndrome, asthma, certain cancers, certain diseases of hypertension, and mental health — these diseases represent a unique social clinical susceptibility. Those clinical conditions are more socially susceptible. And just recognizing and accepting that around how an individual's health care outcomes are, I think is also really important. 

Interviewer: Excellent. So to continue exploring some of the obstacles, do you want to share any more about the challenges and barriers that are preventing health care professionals from better understanding and considering social determinants of health? 

Bosch: Hannah, the barriers that seem to continue to come to play around harnessing the value of social analytics, often come from a lack of specificity of data, a lack of scale to apply that intervention, and then finally a lack of stamina to recognize that the lift here is one that is for the program and the community and often has to be measured in small intermediate lifts, not the ultimate holy grail of better immediate health outcomes or reduced immediate total cost of care. Both of those are achievable, but those are destinational that can often take quarters and years. And the short-term measurements are more intermediate. 

So I'll start with the first one as a challenge for our industry around social analytics, what we do, and delivering to customers that are payers, providers, life science companies, government entities that are at risk. So they have financial and health related risk for their population, their denominator. The obstacles of specificity are one where prior to a social analytic focus and a paradigm that's very scientific, it becomes more of a screening model where you're asking for patient-reported content. Nothing wrong with this, this is current state. You're referring or completing the circle, connecting the patient with a resource, and then you're wiping your hands and hoping that it's done. So we often say, try not to sound snarky, it's screen, refer and hold. And then push the reset button, cause there isn't any cycle back. It's not a feedback loop. It's not natural. 

And so in those endeavors, when there might be resources within a given state, or given insurance plan, or given provider, what I mean by resources, copay coupons for ride-share programs for getting to and from clinic, housing subsidies, utility subsidies, food as medicine, market on the green grocery stores, traditional health-related social needs interventions. Our solutions in the marketplace and again one of the things our company focuses on improving why we exist is because we lack that specificity. We screen, refer, and hope that actually it made a difference. We don't even know how many people went to the food clinic or how many folks went and received that subsidy or how many folks use that supported Medicaid ride service to clinic or to physical therapy or to dialysis. And that model is not sustainable. It's not specific enough. It doesn't allow us to learn from what we missed and it's blunt on so many levels that the lack of specificity is a problem.

And so the first thing about our industry and what we do is we need better specificity. Better cohort creation to precisely match to specific programs and then measure them so that we can increment and improve them in our humble enough to indeed circle back. 

The second obstacle I mentioned was scale. And scale is the specificity and number of times. Let's just make it general. And like you said earlier, it sounds kind of straightforward. Yes, we'd like to see enough scale where when the state of Arizona, for example, and the governor has a large broadband subsidy that has been federally supported and is brought down to the state and a unique program around the 1115 waivers.

We want Arizona, and I'm just using this as a general example, we want Arizona to be able to give more specificity than to just deliver broadband to the Northwest corner of the square state because that area looks like they might need it. It's got to be more specific. And then it has to have more scale where huh, got seven million folks that are living in Arizona. Roughly 2 million on Medicaid. Let's look at just the geographies, not the persons. Let's look at the geographies to the sub-neighborhood level and let's quantify risk for digital access, access to broadband, access to libraries, access to cell phones and Wi-Fi. What is that risk? I didn't mention people at all. Let's just understand the risk 1 through 5. That would inform us. I don't know that anybody could argue with that. That would inform us where that subsidy might be directed. 

Now, if we could double down with our scale and say we're not just going to know the locations that have unique decrements in their capacity to support information and health literacy and health access, but we're also going to do individual risk. Because if we are fortunate to have screened individuals in that northwest corridor of Arizona, we might get to 5%. That's it. That's the current going expected rate of answering a questionnaire like PRAPARE, two A's, a government-supported questionnaire that asks about social determinants of health in a unique way.

Twenty percent of the population. What about the other 95% that didn't answer it, right? And of course, there's some explicit bias in there I have to call out, because if you can get to the internet and you can get to your portal and you can get to fill out the question, yet I'm asking you whether you have health literacy and access challenges, you're already ahead of other vulnerable populations, yet I'm using your answer to extrapolate to what the other 95% of that population are at risk for. And so I think the second exciting challenge and obstacle altogether is how do we bring scale, right? Scale is that repeated specificity in that area so that an intervention around social clinical risk matches that precise cohort.

And then we drive to execution. We don't just refer and hope. Yeah, I hope they went. I really hope they went to the food market because they were at risk. They said they had food insecure. I hope they went. No one knows if they went. We're actually going to and sent them to go. And again, the we becomes the state or becomes the at-risk entity because we want to measure the outcome on health, reduced utilization, reduced inpatient care, etc.

So, those are the things that I think are really important and the last one was stamina and stamina is so hard in health care, not just because of the politics of red and blue states, which I'm not going to touch. But the stamina in health care because we're right in the middle of the unwinding of the COVID pandemic. And so, the incentives around the flow of money in terms of really important instruments in health care like telehealth, like outpatient outreach, like the ability to provide traditionally social solutions, access to your home, housing, transportation, health access, and English and education support that can now be provided under the auspices of some of the Medicaid programs and Medicaid waivers.

The stamina in health care always struggles from the different environment that we're in, and COVID changed a lot of that. Some for the better, I'm not going to blame all of it because telehealth was a win, but we still lack stamina within the social determinants of health because it requires quarter by quarter, year by year, not week by week, day by day interventions. And that's very difficult. It's very difficult for our third-party payer model. It's very difficult for our state budgets. It requires stamina.

So my challenge and my hope when I talk to the feds, when I get a chance to speak to CMS and CMI and others, is they will continue to incent and reward that patient, that whole population approach, not just whole patient, which is a key thing we hear from playbooks and our leaders in government, state, and federal, but whole population. And I think we struggle from a screening myopia right now. We want a screen, screen, screen. And as practice in your internist, I'll never say ever, ever, ever, I don't want to ask my patient and listen to my patient as what's going on. But I will tell you, we put way too much emphasis on a one time, 5% response without context. The questions are reasonable, not throwing shade at the questions, but they require context.
Am I hungry today? Is it before lunch and my food insecure this month about it that have been food insecure for the past 3 years? The context of the questions are still not matching our interventions.

So screening is a vital piece of what we do, but it has to be expanded. So I hope that the obstacles of social analytics, we're going to break out of that myopia of just doing screening and starts to look at the whole population, because with social analytics that my company is endeavoring to provide, and others too, we're looking at the entire population, everyone, and we can know something about everyone in the United States. We're looking at the entire geography, the entire United States broken down into 200-meter hexagons. Each has a sub-neighborhood risk 1 through 5 for unique domain social determinants. So we understand the capacity because it doesn't do any good to make a referral to a real unique cohort, diabetics with food insecurity and ward 9 on Medicaid. If there isn't an intervention in that program, you don't have the capacity. That would be screen refer and wish. And so somehow we have to make sure we're looking at what we can know about the entire population and I think those are some of the biggest challenges for us that I see movements but we're not there yet.

Interviewer: I love how it seems like you're really taking this issue or this instance of “give a patient a fish they'll eat for a day, screen a patient about if they have fish, never check in with them again”. What about all the other people in the community who don't have fish? It really takes that small instance of experiencing how social determinants of health affect the patient and just blowing it up and scaling it to this huge level. That's so amazing.

Bosch: That's a nice way to put it. And I think so, it's just how communities work. I've done a lot of work in health literacy and vaccine hesitation, particularly through COVID. And I always remember my learnings there, I believe all probably seen where the celebrity pop star that looks like the community that's most vulnerable is invited to town to get a shot. And there's a general assumption, presumption, albeit direct myth, that this is going to lift the amount of shots that are received by the community. And in my opinion, my experience and what I've been taught, the community is interested in the lift of the community as a whole. So the stamina of health clinics that are in the community talk about things way beyond just the vaccine.

And quite often when folks are hesitant on intervening to be good to their health, not just with a vaccination, taking their diabetic medicines, going to dialysis, or getting a procedure, it is often for a larger protection of their community, a reason that is beyond what is sometimes thought. And I think just listening and understanding how we can have stamina with social clinical interventions, not just bringing celebrities to town saying we're going to build a food shelter. I think this is where it can be transformative is really understanding that it may be food education, for example, not accessibility, you know, what is it in the community, and sometimes listening is better than acting.

Interviewer: So true. So shifting gears a little bit, from a policies perspective, how is the relationship between social determinants of health and health care important, and what policies support health care by addressing these social determinants of health? What are the potential consequences of reducing such policymaking? 

Bosch: So when I started pushing within the social determinants of health space, probably a decade ago under different names of what the social determinants of health were called at that time, there was quite a need for evangelism around its impact. And today, fortunately, that's no longer the case. The inflection point has occurred. I say that because any policy statements and policy action in the policy committees that I get a chance to listen to and inform, and I feel like I'm on the pragmatic side, an actor in this, there is an assertion immediately a stipulation that social determinants of health affect 80% of health care outcomes and 80% of health, if not more. It's comforting to see that that has become indelible and has started to motivate policy.

Our policy wins, in my opinion, have been in fits and starts. I'll compare to medical risk adjustment policies from about a decade ago, which particularly changed the fee-for-service paradigm around care to be less about single episode care for a specific invoice or dollar amount and more about capitated care that was prorated and allowance per person within a given group and in that capitated care model, they needed a way to measure how sick that population was. So keeping things real easy, let's say it was a 10 000 person population of Medicaid in an Eastern state.

If each of those individuals had a par level sickness with how many of them may or may not have diabetes, heart disease, asthma, et cetera, by their billing rates and by their bills, then we'll keep it nice and clean. The payer plan would get $1,000 per member a month in prorated, capitated, ahead-of-time payment. Keep it, do whatever you want with it, make people healthy.
If that individual spends more than that 12 000 a year, hey, insurance company, you're on the hook, they spend less than that, hey, you can keep it, right? This is the value-based care model. So I give kudos to the policymakers that created this because this created a need for upstreamism, for proactive population health, not just transactional, maximize the total count of fees for service bills that you can create. 

I take my time to show this and this depict this analogy because to me, that's where policymakers need to be around social clinical adjustment. So not just adjusting a population to its par level for their medical morbidity, but adjusting the population on social clinical susceptibility. And so doing this through analytics like my company provides and other analytics that are out there will enable the proper capitated payment to be free for that individual. So in the medical risk model, if you had a high incidence of diabetics in your hypothetical 10 000 population, then you might have gotten two times par rate. So you got $2,000 a month. And the payers and the at-risk entities really benefited by this because dollars followed risk. 

Hannah, I'd like to see the same thing happen. Social clinical risk adjustment, understanding enough about our population. I mean, all of Medicaid and Medicare is going straight to the government. They have the coffers of the data. Some unique groups can get to the data. We've partnered with some of them to begin matching analytic risk to outcomes. This is the opportunity that the policymakers have a precedent for and need to push. There's a few demonstrations here and there. I think the screening myopia is still getting in the way a bit on that. Everything's about screening. And I think we need to push towards whole patient, create social clinical risk adjustment so that we can start to understand that populations in Louisiana and Mississippi and in other southern states have unique geographic risk. I didn't say person, geographic risks where we know certain diseases and utilization patterns are greater. And we can actually put that into the model. The model, much like medical risk adjustment, could create a social risk adjustment. That's what I'd really like to see lift the needle for our industry. 

Interviewer: So what role do payers, providers, and government organizations play in addressing social determinants of health, if we could dig into that a little bit more? “Why me?” for everyone that is a part of the equation, why is it important and what can they do? 

Bosch: Quite simply, payers, providers, government, entities, all acknowledge that 80% of health and greater than 80% of health care costs are derived through the social determinants of health. And so with that charge, if you will, the alignment among the traditional payers, providers, and government to create programs that will better measure that risk so that the dollars can follow the risk.

It's imperative, there's a dozen or more of our states across our country in the United States that have the largest part of their budget going towards their Medicaid model. And it is ever increasing. And that's inclusive of the federal pull-down for Medicaid. Or exclusive, I should say.
So the challenge in the budgets and the impact, I think creates a unique place for each of those individuals. I'll start with the providers because I am one. Yes, I'm a doctor, but I'm going to channel the provider being the entity, the health care system, the intermountain, the Geisinger, the academic medical center like Johns Hopkins, the very large integrated provider systems.

Now, those providers have started to take risk. So they've started to offer insurance products or partner with traditional payers so that they can take risk. The providers still are the most pragmatic, the least policy driven because they're living in that final mile, that final mile. They're seeing the transformation of less inpatient care. It used to be all about how many beds we can fill in a hospital, right? Not anymore. Not anymore, especially when the better you can care for people, the more of that proactive, capitated payment that you can invest in and put towards your profit line.

So I think the providers are much more final mile driven, much more data information-rich, but yet they're still putting it together. The payers have had a unique experience in the past 10 years, particularly around the pandemic, because they've had a greater flow of money to them directly, and that income of money has been because there's been an awful lot of deferred medical care across the country. So yes, COVID cost an awful lot, but the payers have had an ability to direct some of those funds from here that had been postponed at an individual level to try to support this public health epidemic.

I think the payers are uniquely able, in their process, to make sure that we're delivering solutions that are most justified. And you go right back to, "I'm not allowed to order a head CT for someone with a broken arm." And I know this sounds facetious, but the payer would be like, "I'm sorry, Dr. Bosch, we don't approve of that expense." So again, I'm being facetious, but the moment is that the payers can, in a very similar manner, through information and data, help to add to the interventions that work, not just the interventions that are optics, but push on interventions that work.

They can do that through quality measures, through HEDIS measures, through STARS measures, and there's some movement in that area, but the payers can do more just like the providers. Then lastly, I think the government still has an opportunity to push us to look at whole patient, population care. And I think we are seeing movement in that area. It does take time. Well, in my career of almost 30 years in health care, I've seen meaningful use, accountable care, precision medicine. You know, every 10 years, there's a different name for our movement. I think it's time for social clinical susceptibility. It's time to understand how health equity and social risk impact health care outcomes. And it's time to actually go strong in that area.

Interviewer: Excellent. And so, jumping off of that, looking to the future, what do you anticipate happening in health care related to this over the next maybe like 5 to 10 years? What do you think the future looks like? 

Bosch: Five to 10 years. Well, I have a couple of kids just finishing college and, you know, they kind of look at things in 1 to 3 year long term. So 5 to 10 years, that's a big window. And I would say that in 5 to 10 years, we must have a robust method for real-time cohort analysis within our population. And yes, I went there. This is one population of 280 million individuals, and that content should be shared. It should be known like a utility in the next 5 to 10 years. It doesn't matter if you're a payer or provider or government or not. That information is still individual. I understand. And as an individual, I should have some consent to how that is shared. But much like my silly analogy earlier of a head CT for a broken arm, the ability to intervene with certain programs, many of which the government has pioneered. Chip programs, TANF, food stamps programs around housing subsidies, the 1115 waivers that are now in a dozen states. Five years from now, I'd really like to see a robust social clinical cohort that can be drawn down and shared among interventions, among intervention partners.

So whether you lived in Arizona or Florida or New York, based on your unique social clinical risk, you could be afforded access to certain programs that we know provide lift. I think that would be the holy grail. I think 5 years is enough time to get there. And I think the government would be able to innovate around that from an infrastructure standpoint, meaning the work they're doing with the trust networks and shared networks, particularly around electronic health record privacy and control, yet liberating data that matters. I think it's incredibly important. Patients can still opt out, but there's going to be much more of an obvious incentive to opt in because opt in will come with benefits, will come with an opportunity to really know that patient or that individual and provide the right kind of services that would health care costs, and at the same time, lift health.

I can't help but think about, when you say 5 to 10 years, the one-payer system. I think I'll resist opining on that, other than to say we need whole patient care, and it is just not good on the system to have an extreme amount of redundancy. We see this in almost every state where just when a Medicaid member is starting to get some lists, some care, some relationship, they churn either to another insurance within the MCO model, or they churn to a different insurer. And I think that really is detrimental to the individual, but also to the state.

There's a lot of redundant costs with meeting the patient all over again, establishing a different network, establishing different access. So I think the model of a whole payer system is not the destination, but we need to really look at whole payer analytics. We need to be able to share that information across. So I would see that as just the way we get to better health, because ultimately as we talked at the beginning, social determinants of health have a lot to do with being productive citizens in our society, it's about health. It's not just about health care costs, it's about job productivity and fulfillment and being a member of my community. And I think those things are going to be more and more important and they're going to force us in 5 to 10 years to look much more holistically at our health care model.

Thanks for tuning in to another episode of PopHealth Perspectives. For similar content or to join our mailing list, visit populationhealthnet.com.