Kevin Volpp, MD, PhD, interviews Karen DeSalvo, MD, MPH, Professor in the Departments of Internal Medicine and Population Health at Dell Medical School, and former Acting Assistant Secretary for Health in the U.S. Department of Health and Human Services. Part 1 of a 2-part interview. Part 2 continues here.
Kevin Volpp: Hi, this is Kevin Volpp. I am a professor at the University of Pennsylvania, and I’m joined today by Karen DeSalvo. Karen is a professor at the Dell Medical School at UT Austin and has had a very distinguished public health career. I’m pleased to have Karen join me today, and we’re going to talk a bit about Karen and Karen’s career, and thoughts she has about public health.
Karen DeSalvo: Thanks, Kevin. It’s great to be here.
Volpp: Karen, you’ve had many interesting roles in your career, starting with a distinguished academic career at Tulane, service as the Commissioner of Health for the city of New Orleans, being Assistant Secretary of Health at Health and Human Services, the National Health IT Coordinator. Now you’re at UT Austin, doing interesting work on social determinants of health, technology, and digital health, and you’re also a board member at Humana. Lots of things we could talk about, but I’m going to start by asking you if you can talk a little bit about your career, and how it’s evolved. Are you surprised by the path your career has taken?
DeSalvo: It’s been a wonderful journey. To be honest with you, every role that I’ve had I’ve very much enjoyed, even though in many ways, so many of them have been quite different. I wanted to be a doctor since I was in junior high, and that is still one of the most wonderful opportunities that I’ve had professionally, to be a part of the lives of my patients. I practiced primary care for 25 years, and learned a lot from my patients — not just about medicine, but about how their lives impact them, and the barriers and challenges and ways that we could make the system better.
I think that thematically, one of the reasons that I have been on this journey of having responsibility in medicine, and public health, and technology, is two things. One, I feel inclined, when I see things that are broken, to fix them; I want to get involved in fixes, whether that’s been at the administration level around clinical practices, or in the policy environment in my state, or at the national level. The second area is that I also really love to build bridges between sectors and systems. Across my career, I’ve had the opportunity to build bridges and work between public health and medicine, or technology and public health, as examples.
When I was Health Commissioner, frankly, a lot of what you do is bring various sectors together. I guess if you had asked me when I was 13 and I said I wanted to be a doctor, did I think my experiences would’ve been as enriching and varied as this, I probably would’ve never imagined it. I feel so thankful that I’ve had the chance to be engaged in all the ways that I have.
Volpp: It’s really quite amazing, because you’ve been engaged at a high level in lots of very different, interesting things. I’m curious, given that you’ve had all these experiences in different arenas, what you see as the health sector’s biggest challenges.
DeSalvo: Our challenges are tightly related. I’ll start with what I think [is] the biggest problem that the country’s trying to solve: that we have created an approach to health that is very focused on leveraging the health care system and all of its very extensive parts, and have created an incentive structure for that very expensive system that drives it to want to make more and more money, and not do that on the backs of better and better health. I think the big challenge about the cost is that the model is so misaligned with what we need to do as a country, of having a healthier population and a healthier workforce, more well-being and vitality for people and community, that it’s . . . so many people describe a sickness model, and that sickness model is doing really well.
It’s growing as a part of our economy and building jobs, but we have to find a way to get off of that track and move toward a business model that incentivizes health and find a way that’s just as helpful to our economy and to growth, but that also is helpful for well-being. In all of that, it’s not just that we need better clinical excellence. That’s the necessary beginning, but we have to begin to think about the other parts of health that are less attended to, whether that’s social needs or the built environment, those things that are driving 60% of our health outcomes.
That, for me, is not just a challenge, but a great opportunity for the country right now to think about how we build out, and professionalize, and digitize, and integrate the social determinants as a partner to the health care system.
Volpp: That all makes a lot of sense. One of my least favorite statistics is that the U.S. spends far more than any other country in the world, but we rank something like 26th in life expectancy. So much of it is, as you said, because we see the production of health through the delivery of more health services. Unfortunately, that hasn’t worked out so well for us as a country, in terms of the health of our population. I’m curious, as you think about that, and you think about some of the needs that are out there that aren’t met, and where there’s underinvestment, how do you see us as a country shifting and reallocating resources in ways that might produce bigger returns, in terms of improvements in population health?
DeSalvo: The idea that the sicker people are, the better the health system does sounds so awful, when you describe it that way, because I know, certainly, [that] most doctors and health professionals and health system leaders get up every day wanting to improve health. But the challenges are in a couple of areas. One, the incentive structure, as we all know, drives doing more and not necessarily better. Driving toward value of better health, [there are] great examples of where that’s beginning show us the way, and how that can be the case.
The second is that we don’t have a lot of visibility yet or understanding about how to quantify or even qualify the other inputs to health. Meaning, what are the life circumstances, the context of people’s lives that are driving their health issues? And then, that system, even if we find that maybe there are some social care needs, it’s not really modern and ready to partner. It’s incredibly underinvested in. To give you an example, if you think about asthma and the fact that very often, when we want to tackle asthma using a medical model, we think about better drugs, better drug regimen, care management, case managers, being able to track and monitor people using run charts — all the ways that we try to think about their clinical experience.
But if you step back and try to understand their home contact, and understand that maybe they have mold exposure or insect-part exposure in their house, and are able to work with the housing system and/or the legal system to make change in their home environment, we know that that has great impact on reducing kids’ presentations for asthma exacerbations, and that means they get to go to school and learn.
There’s some even bigger work that people are starting to do. I point people to what Louisville, Kentucky, recently published, that they not only looked at this housing environment, but they began to step back and look at things like the canopy, air quality, and other input to driving asthma exacerbation in their community. They used a digital tool to track when and where people were using their inhalers, and that created heat maps that public health and the city government could use to change not just the house environment, but the environment around that.
That kind of model is great, and those will only be exciting projects until, to your question about investment, we really start to understand how to make investment that, seeing that we’re able to move upstream and give everybody quality housing in the first place, and make sure that we have transparent and public ways of monitoring air quality, and have resources sufficient to address keeping clean air and having great space for all the reasons that matter to the public health.
We have not done a good job at solving for that in the U.S., but there are some ways that we can either learn from other places, or that we’re experimenting here to do work, like Pay for Success or social impact bond models, where we’re saving money on the health care side, and not just having those gains stay in health care, but get reinvested in social services, or the public health infrastructure, so that, again, they can be better partners to health care.
We’re at the beginning of that journey, but we have to get our best minds around that from a policy and an economic standpoint, to make sure that we’re really being as innovative, but also that we’re remembering that there’s more to health than the health care system. We have to be investing in housing and food and transportation, just as we are in things like air quality.
Volpp: Let’s imagine for a moment that you were chief medical officer of a large health plan, or a large health system, and you control a fair amount of resources, but those resources were limited, and as part of those resources, you had to also provide clinical services to those populations. How would you go about trying to think through how much to allocate toward reducing social determinants of health? How would you decide which social determinants to go after? Do you feel like there is solid evidence on how to approach specific areas in a cost-effective way?
DeSalvo: The evidence is, for interventions, still emerging; it’s an exciting era of evidence-generation. But in that case, what I would do is I would go to the data that I have in hand, and I would look for the individuals in my population, whether they’re my members of a health plan or of my attributed patient population, and using stratification, find out who seems to be needing the health care system more than others, who are the high-cost, high-need patients? Those patients or members typically also have a lot of complex social needs as well, and I would target them by asking questions.
I would call them or go into their home as part of my risk assessment, or as part of my regular intake, and I would ask them, using structured tools that are already available out in the world, whether they’re going to bed hungry, if they have stable housing, if they feel safe in their home. We do know that there are some domains, where there’s some evidence that if we ask a question, there are resources that we can link people to, and that that will improve outcomes. For most people, most organizations are finding that where they’re able to get some advancement in addressing the social determinants, especially for those high-cost, high-need populations, is in addressing challenges around transportation and food insecurity.
Sometimes it’s, for example, in food insecurity, as simple as looking at the data that you have about that person and knowing that they’re already eligible for SNAP, but not enrolled, but to get them additional food assistance, or get them enrolled with Meals on Wheels, or know where there are local resources that might be available to assist them.
That step one is, what’s already available that we’re just not taking advantage of for the people in front of us? I’d say, also, though, [that] I would build additional data. There’s a lot of good data to be had about who’s at risk in your population and what reinforcements there are in a community, and where there are needs.
There’s an exciting array of tools and approaches that are emerging to do some social risk stratifications, and also to use digital tools to help link people between the health and social services sector. I would be an early adopter of some of those tools, because I think solving this for our patients, whether this is the cost you’re worried about, or the health outcomes, or their social needs, we’re going to have to be able to measure and track it. And that’s a part of not just evaluation, but also of the evidence-generation. Making sure that on the individual level somebody’s referred to Meals on Wheels, that they showed up, and what was the quality of that, but also, is it working? Is it lowering cost?
So, I would start with a defined population, but I would be very deliberate and intentional about asking questions, using structured questionnaires, and beginning to build an evidence base within our own organization, and use the new emerging digital tools to advance it.
I also would hope that I was in a state that was really advanced in the area. Take a place like Minnesota, where their Medicaid program is already working on shared risk models around addressing social determinants. They’re pretty forward-leaning in their understanding that the health care system needs to not just refer to, but be a part of, resourcing up the infrastructure to support social needs, and is really wanting them to place an incentive structure, so that the health care system really understands and acts on it more aggressively.
Read or listen to part 2 of this interview.