Q&A with Don Berwick, MD, MPP (Part 3)
Donald M. Berwick, MD, MPP, is President Emeritus and Senior Fellow at the Institute for Healthcare Improvement, which he co-founded and led for approximately two decades. A pediatrician and nationally recognized leader in healthcare quality, safety, and policy, he served as Administrator of the Centers for Medicare & Medicaid Services in the Obama administration and has held faculty appointments at Harvard Medical School and the Harvard T.H. Chan School of Public Health. In this continuing SIR conversation, Berwick considers the movement from healthcare to health and what the health of Americans tells us about the American experiment.
Across three conversations with Dr. Berwick, we widen the lens from healthcare to health to the American Experiment — asking what we choose to improve, what shapes our health, and ultimately, what the health of Americans reveals about the nation we are and aspire to become.
Martin Goldstein: Dr. Berwick, thank you so much for agreeing to our third discussion. The United States devotes substantially more resources to healthcare than other high-income countries. U.S. health spending is approximately 17% of GDP, compared with roughly 9% across the OECD, and per capita spending is more than twice the OECD average. Yet on measures including life expectancy and maternal mortality, the United States compares unfavorably with many other wealthy countries. What factors account for the comparatively poor population health outcomes, despite the resources the United States devotes to healthcare?
Donald Berwick: Well, on the healthcare side, our resources in healthcare tend to be absorbed by the high-tech healthcare that we offer, and it’s as good as any in the world. If you look at the overall healthcare budget, the amount going into primary care and behavioral medicine, the kinds of things that actually affect longer-term health status is relatively under-supported in the U.S., despite the fact we’re spending, as you said, far more than other countries.
In addition, health status is not determined primarily by healthcare. It’s the circumstances of life that people find themselves in, no matter what their wealth or income is, but especially affecting people with lower income. The amount of resources we pour into helping children in their early years, or education systems, or workplace supports, or elder care, or environmental improvements — that’s what’s determining health, and we are quite under-supported in that in the United States in terms of our overall investments.
Goldstein: Applying your principle that every system is designed to get the results it gets, what do these results tell us about what the American healthcare system was designed to do?
Berwick: It’s designed to deliver the best high-tech healthcare we can for most, though not all, people. If you have cancer and need very elaborate chemotherapy, or heart disease, or need an organ transplant, our system is designed for most localities for access and success at the very high end of technology. It is not designed to do for and with people what actually improves their health in the long run.
It’s also designed to make the wealthy wealthier. I have said, and I do believe, that the American healthcare system overall may be the most effective machine in our country for regressive financing, shifting funds from people of lower and middle income to people of higher income.
Goldstein: In our last conversation, you estimated that healthcare accounts for about 10% of the variation in health at the population level. If healthcare’s contribution is about 10%, where should we be directing greater attention and resources?
Berwick: Within healthcare to primary care and behavioral medicine — those are the two most leveraged areas for actually improving health status, I think, and they’re taking a very small percentage of the overall bill.
In society at large, as I’ve said before, I use the work of Sir Michael Marmot as my framework. I’m sure there are other very good frameworks, but he has a kind of five-part framework for where to put money, and the elements are the following.
The well-being of young children. It’s really almost mysterious that the status of kids, young kids, with respect to safe birthing, bonding, early childhood preparation, is a remarkably powerful predictor of what the health status of the entire population will be. So, invest in kids.
Second is the education system overall. Our education system ranks about fifth in the world according to international published comparative data, but it’s much more uneven than many other nations. There are parts of our education system, especially in lower-income urban settings and rural settings, which are way behind in achievement, and so we’re underinvested in education of children and young adults.
The third area that Marmot talks about is workplace protection — minimum wage, for sure (our minimum wage is quite paltry compared to wage guarantees in other nations), but also the conditions of work: protecting workers and allowing them to find meaning in work, to grow in their talent and training. We’re underinvested there.
The fourth area Marmot talks about is support for the aging population. Can you age well in America? The answer is no for many Americans, not at the level we should be at. One-third of elders are in poverty. Loneliness is a scourge among the elder population, and we don’t really have a national aging strategy that intends to preserve social connections, vitality, nutrition, and so on for the aging population.
And the fifth category Marmot has is about a collection of community attributes — food security, housing security, a compassionate and restorative criminal justice system, safe water, safe air (now we have put climate change on the list), recreational opportunities — how livable are our communities. And there we lag quite a bit behind other nations, especially in settings with poorer populations.
As we talked about before, Marmot asks why these distinctions exist — why we’re behind, and that leads to a sixth, overarching, category: a zeitgeist of fairness. A society in which we really do understand that we’re responsible for each other, and that helping people of lower wealth and higher disadvantage is our collective responsibility. We need to reach out and help each other. And Marmot believes — I don’t want to put words in his mouth — but I think he believes that when the social contract in a nation emphasizes justice, fairness, equity, you get better investments in these determinants of health. He called this commitment to fairness “the cause of the causes.”
Goldstein: "The cause of the causes" — you've spoken about that. Turning to the federal health infrastructure — HHS, CMS, CDC, NIH, FDA, and the other agencies responsible for healthcare, public health, research, regulation, and prevention — you led CMS and worked closely with these institutions throughout your career. If our aim is a healthier population, what role should they continue to play? Are they structured to meet that aim, or would you change their missions, responsibilities, or how they work together?
Berwick: First, I think we have to distinguish between two contexts. One is what's happened to federal infrastructures under the Trump administration, which has almost systematically devastated the infrastructures you're talking about. But even before Trump, some changes to our federal architecture would have helped. That said, I want to point out that our pre-Trump architecture, in many ways, was the envy of the world.
The Centers for Disease Control and Prevention was the go-to place for the entire planet on modern approaches to epidemiology, prevention, handling outbreaks, and improving data and investments in health. It's on its knees now, thanks to cuts made by the Trump administration, but its prior legacy has been wonderful.
The National Institutes of Health is the go-to place for biomedical science, worldwide. There are other great research institutions, but none more accomplished than NIH. It too has been devastated, with cutbacks in research and research investments that future generations are going to pay the price for.
We have a strong track record of public insurance in healthcare. Although we don't have universal health insurance, we do have a guarantee of it — Medicare for people over 65, and Medicaid for people in poorer segments of society. Trump has hurt both terribly. The Medicare Trust Fund is closer than ever to going bust, and Medicaid is facing murderous cuts.
Our foreign aid programs, too, have been world leading. USAID saved millions of lives in its time. It's now been essentially eliminated, and the cost will be paid in millions of lives lost, and tens of millions harmed.
We had pretty good models that have now been seriously hurt. Could they have been better pre-Trump? Yes. One is the primary care infrastructure and our investment in rural health. The relevant federal agency there is the Health Resources and Services Administration, HRSA. HRSA is great, but investment in primary care has lagged behind countries that have harvested real benefits from strong primary care systems — the UK, Denmark, Sweden — by putting money into the first place people go for care.
Waving a magic wand, even before Trump, I would have strengthened our investment in primary care. We've also lagged in behavioral health care, which is just as important as physical health care. Depression affects almost a quarter of people at some point in their lives. Major mental illness touches many, many families. Substance misuse can be treated, we have good ways to treat it, and in all those cases, behavioral health has played second fiddle to technological care.
Goldstein: Picking up on what Trump has done to those institutions, in our earlier conversations, we examined how the healthcare system came to be structured, and whether improving the system can substantially improve population health. You've emphasized Marmot and the importance of beginning with the aim. Suppose our aim is a healthier population, and we take what Trump has done to the institutions as a blank slate — what would you build, or retain, to achieve that aim?
Berwick: Well, that's an optimistic question, isn't it? I think we have yet to understand how much of the damage Trump has done — to government institutions, but also to private sector supports — is reversible, and in what timeframe. We have to rebuild CDC to what it was, and, sadly, I have very good friends who say that may no longer be possible. The same goes for our biomedical research enterprise — we've driven it offshore. People are leaving their careers, leaving the country, taking that expertise elsewhere. It's not a trivial matter to rebuild, or to use the blank slate, as you put it.
Should we really get the chance? Rebuilding first — we need an NIH and a biomedical enterprise that's fully supported and not politicized the way Trump has done it. We need to develop science the public can trust, protected from politicization. We need to return to the idea that the best healthcare is based on secure science.
Robert F. Kennedy Jr., the Secretary of HHS, has undercut the understanding that science matters and produces better results, and we're going to have to do some real rebuilding there. That's a matter of institutional rebuilding, like NIH and CDC, but also rebuilding public trust, because when political leadership goes as wild as this administration has, the public gets understandably confused and mistrustful. We've got a hard job ahead, building back mature dialogue with a trusting public about the value of science.
In healthcare delivery, there are disagreements about exactly how to improve as much as we should. I think we need to move back closer to healthcare as a human right — not leaving tens of millions of people out of coverage, or out of reach if they're in rural or isolated areas. That's all been seriously injured now. We have to rebuild those infrastructures.
I think this is a values question, though. If I really take your blank slate seriously, I would have a national dialogue about whether healthcare is a human right in this country, and, if so, then to hold our political leaders accountable for seeing that everybody has it and giving them the resources to do that. The country has spoken that language off and on — health care as a right — but it has never delivered. This is the time. That's why I favor national health insurance — not because it's a perfect solution; it isn't, but because we can finally use it to guarantee that everybody gets healthcare. Period.
There's a bigger question, a little risky to get into: historically, we built about half our healthcare coverage on an employer-based system, and we got used to it. It's a little odd, not every country uses employers as health insurers. Is that something we might want to consider changing? Not taking anything away from people, they get frightened when we talk about that, because many do like their employer coverage. But is there a smarter, better way to get coverage to everybody than through employers? Maybe, and that's a conversation we ought to have.
Goldstein: There were three things in that response: trust, values, and a change to the employer-based system. In the context of the political turmoil we're currently in, how would you go about addressing those three elements?
Berwick: Trust begins with trustworthiness. You must have leaders who aren't behaving like the current administration — lining their own pockets, supporting their friends, shifting money to the wealthy, away from the poor. We have to say what we believe in and act that way, and that starts with leadership. That's why elections matter more than ever right now, and we need some real dialogue about this. People have become mistrustful, and we need to be honest about the nature of science that it's always evolving knowledge, which means what you think today may not be what you learn tomorrow. We have to return to that understanding. That comes back to education. We need to invest in schooling that produces a scientifically literate public, one that understands how knowledge is built over time.
We need to protect people better. Right now, insurance companies are playing prior authorization games, coding games, leaving people to find out at the pharmacy counter that the drug they thought they were covered for is not covered. We have to stop that. If we're going to have private insurance, we must hold it accountable for delivering what it ought to deliver — and if we move toward public insurance, national health insurance — it has to serve the needs of people. People are smart enough to know when they're being tricked, and to some extent today, they are being tricked.
That was trust. The second is values. Here, we need leaders who speak out directly about healthcare as a human right and recognize when we're falling short of it. The remedies vary. I believe government's role is essential, that it has to step in as guarantor, at least one of last resort. Others disagree — they think the government is not a very good insurer and prefer the private sector. I'm willing to have that conversation, but it should center on one test: does everybody get the care they need?
The employer question is more technical, and I wouldn't fall on my sword over it. If that's how we want to do it for now, we can work with that — but it has to be mediated through insurers that actually put the patient first. I've said publicly that I think the best insurer, in the end, may well be national health insurance.
Why should a car company, a tech company, or the CEO of a grocery chain have to focus their energies on providing healthcare to their employees? If that's how we want it, we need to hold them accountable for doing it properly. But is that really what they should be doing, compared to a compassionate, accountable government? I don't know — we need to talk about it.
Goldstein: Taking a turn to measurement — building on your experience as VP of quality of care measurement at an HMO. You said we had a lot of metrics, but nothing changed. In our second conversation, you called it a myth — that if you measure, some magic will happen and things will change. You've just described at length how the system should change, and I think of investigators like Starr, who described a medical system organized around treating illness, and Wennberg, who showed that more medical care doesn't necessarily produce better outcomes. How do those observations relate to what you've just laid out — how you'd address poor population health, and unequal access to effective healthcare?
Berwick: We're a bit off-base in this country on measurement — in some ways too little measurement, in many ways too much. First point: measurement isn't magic. There's this meme — “If you can't measure it, you can't manage it.” That's not true. We manage many things we don't measure quantitatively. A better idea is that “If you can't see it, you can't improve it.” That is true — but the seeing can be narrative, and more forgiving and flexible than strict measurement.
I'd like to break our obsession with measurement as a key to improvement. Ideally, measurement is a servant, not a master. Proper measurement answers a question that comes after values, after goals, it tells us whether we're making progress on what we want to make progress on. It can't substitute for the goals themselves.
If we say every person in the United States should have access to the care they want and need, when and how they want and need it — the first step is, do we believe that? Do we want it? Then we ask how we'd know. The idea that we can drive values with metrics is a costly mistake.
That said, once we have goals — I just had a hip replacement, two months ago. Pain I'd lived with for years is completely gone. I've got rehab ahead to get my strength back, but it's really something. I didn't get an infection, didn't bleed excessively, didn't die from anesthesia mistakes. My outcome was great. We want that for absolutely everybody.
If that's what we want — you get a hip if you need one, and not if you don't, that's the restraint side — and you don't get infected, and you can walk better in two months, maybe better still in six, we can measure that. There may be variation: hospital 32 has the best outcomes, hospital 16 isn't up to snuff. That should be visible, and hospital 32 should help hospital 16. We need a learning system built on that.
Measurement has a role, but it needs to be subordinate to values. Don't let it come down to making metrics the goal, a metric is not the truth, at best it is a mere shadow of the truth.
Goldstein: You've spoken about Marmot's five social determinants of health, and the sixth, the cause of the causes — he emphasizes the importance of people's control over their lives, which I take to be that cause of the causes. What does control mean in practical terms?
Berwick: I think it starts with a sense of agency. It's very interesting. Marmot has some stories in his book from research in Canada that when people, communities, individuals, families feel agency, a sense that they can control their destiny, they're actually healthier. Mentally and physically. Healthier by a lot. There's an important connection between not feeling helpless or on the positive side, feeling agency and well-being.
To me, that's the shape of the answer. People should feel like they can actually affect things that affect their lives. In healthcare, it shows up at the micro level in the dialogue between a doctor and a patient, or a nurse and a patient. I'll go back to my own hip replacement — I was in quite a bit of pain postoperatively, but the nurses put me in charge. "How's your pain? Would you like pain medicine now? Is there anything you need that you're not getting?" Those were the questions they asked me, and that helped me get better, because I was a partner with them.
That's the micro-clinical encounter — the doctor who says, "We could do this one of two ways. Let's talk about the advantages and disadvantages of each, and how you feel about it. What matters to you?" At the more general level, in communities and societies — does the community get to affect the care they get? Speaking up and saying, "That's not going the way we need it, and this would be better," and hearing back, "Is this what you meant? Is that better now?" That kind of dialogue is needed.
The exact opposite is the insurance company that says, "You're not covered for that." "Why not?" "You're just not covered." I'm a pediatrician, but also a father, and I remember an episode with my daughter — she needed a test done, no question, medically. And the insurance company said, "We refuse to approve it." "Why?" "Explain it." "It's just turned down." I had no agency at all. We have to stop that.
Goldstein: That dialogue is critical, and so is building trust. How do we change it — how do we enable people to exercise the kind of meaningful control you had over the conditions that affect their health?
Berwick: I'm going to be boring about this. We first have to decide that we want to change it, and there needs to be a social dialogue — maybe the term is helplessness in healthcare, and how to end it. I'm starting an organization with colleagues called Power to Patients, a new nonprofit designed to answer exactly that question: how to shift the balance of power.
Part of it is regulatory — we need government to step in with insurers, or others, and say, "You can't do that, it doesn't respect the rights and prerogatives of patients and communities." Regulation does have a role.
I'm less sure about this next part, but I think we have to bring it down to smaller aggregates — municipalities, even neighborhoods — because the conversation we need has a lot of fine structure to it, and I don't think a big national conversation is what I have in mind. It means the community I'm part of has a locus, a place to talk about who's controlling my health and my care.
In hospitals, I've become more and more a believer that we need to change the structure of boards of trustees. They're well intended, but largely made up of donors, real estate agents, and hedge fund managers, especially at big hospitals. What about patients? Maybe we need to put patients and families into governance — that's one way to have the dialogue you and I are talking about.
Goldstein: Over the course of American history, the social contract has expanded through constitutional and statutory protections, and through programs like Medicare, Medicaid, community health centers, public health programs, and the Affordable Care Act. Yet formal inclusion doesn't necessarily mean, as you've described, that the conditions and opportunities that contribute to health are equally available. We say people's health counts, yet where Americans live, their income, education, race, and social circumstances remain strongly tied to how long and how well they live. What does that gap between our stated values, and our actual outcomes tell us about the American social contract? I framed this around the debate that's come up around America's 250th anniversary.
Berwick: Good time for that debate. I think this connects to our earlier conversation about trust — if you say one thing and do another, that erodes trust. Are we going to say healthcare is a human right in America, and then leave 30 million people without health insurance? That's the jig being up. You said one thing, and you're doing another.
We say we believe in equality, and yet your zip code determines your lifespan, for reasons we know and could change. We can't keep saying one thing and doing another — it isn't the foundation for trust. And that's what the social contract is, it seems to me. What do we promise each other as a nation? How do we know if we're fulfilling that promise? And where we fall short, what are we doing to close the gap? There has to be a story to tell there.
The Affordable Care Act was an example — heavily negotiated, not a perfect law, but we had a president who said, "You ought to be able to get healthcare, and we're going to pass this law so you can." Then we counted how many people got more healthcare, and we saw we'd made progress, but not enough. That's the nature of a social contract, a promise, some form of monitoring, and action.
Goldstein: Over the course of these three interviews, it's been a privilege, and it's really changed my thinking. The debate is always about healthcare. It's never about health. Those are two very different things — you've spoken about them, and Marmot speaks about them, as do others. Looking at the differences in health and health outcomes, how would you describe how well American institutions are functioning?
Berwick: American institutions — first, I've tried hard to resist zero-sum thinking about your question. Maybe I'm naive, but I think a wealthy nation can have it all: healthcare that meets needs, and health. I don't think we have to choose one or the other. I really don't.
So how are we doing? On the healthcare side, at the very high technical end, pretty darn well. I don't think it's true that if you need an organ transplant, cardiac surgery, or a new hip, you're better off in the U.S. than in every other nation — that's not true. But we are equal to the best. Our results speak for themselves, and, by comparison to other wealthy nations, they're quite good.
Two gaps in that answer, though. First, I don't think we've really grappled as a nation with patient safety, things go wrong at that high-tech end that could be prevented, interrupted, or eliminated, and we don't have an industry focused on safety the way most high-risk industries are.
Second is equity. I'm lucky, if I need a new hip, I can easily find my way to a doctor and hospital. But we know the data: if I were Black and poor, I'd wait, and I'd have a much rougher time. So, yes, our high-tech care is good for most people, but not all.
With that exception on safety, high-tech, fine. More general healthcare, at the primary care and consultant level, not so hot. The robustness of our primary care system is eroding. Physicians are leaving the field — it's become an undoable job in some important ways, and we generally haven't solved that. Other countries are doing a better job at primary care than we are. We need to tighten our belts and get back to that.
On health, look at the results. Rank the wealthy OECD nations on lifespan, health span — how long you have a healthy life before you get chronically ill — mental health outcomes, substance misuse outcomes. We're generally at the bottom, or close to it, on any list of wealthy nations. Something's not right, and it has to do with investing in the infrastructures that actually produce health.
I'd give us an A-minus on high-tech care, maybe a B-minus or C-plus on the robustness of primary care and our referral systems for specialty care. And on actually producing health for our population — a D, maybe a D-minus.
Goldstein: Not even barely passing.
Berwick: It's not great. I hope I'm not being too critical here. People have pointed out that we still do miracles, and we do. My new hip is as close to a miracle as I'll experience in my life.
Goldstein: There we go, that's a good thing. The way I've begun to think about this is as a lens, starting with healthcare, then widening it to the factors that contribute to health. The widest lens, which you've been describing, takes us to the social contract: whether the conditions and opportunities that contribute to health and flourishing are broadly and fairly shared. That brings us to a distinction we haven't explored yet — health and flourishing are related, but they're not the same. What is the relationship between the two?
Berwick: What a great question. Let me think for a minute. Health — being able to walk where you want to walk, play the sports you want to play, learn the things you want to learn — must be a precondition to flourishing. It's not that people who are ill can't have full and satisfying lives, but overall, in a society, the conditions that make your body stronger, and your mind less troubled, are preconditions to flourishing.
But I think you're on to something here: if we want to expand to flourishing, we have to look at what makes life meaningful — art, companionship, music, avoiding loneliness, enjoying one's community, spirituality as one chooses to practice it. These are the conditions of a happy life, a flourishing life, a life of joy, to the extent joy can be had. I don't know if that's part of a social contract — but life, liberty, and the pursuit of happiness is in there. We've got "pursuit of happiness" as an explicit goal for our nation since its founding.
Goldstein: You just anticipated my next question, which was: the Declaration of Independence identifies the pursuit of happiness as an unalienable right. At America's 250th anniversary, how should we think about that promise in relation to the opportunity to live a healthy and flourishing life? Have we passed?
Berwick: You’re taking me far afield from my expertise, but no. Hopefully people in their localities and neighborhoods are able to feel happiness, the joy of love and being with others. But too many, I think, have big obstacles in their way.
I've become very interested in Buddhist philosophy, and I think the Buddhist belief is that joy doesn't lie in transactions, or accumulation, or attachment. It lies in being in the moment, being present with oneself, whatever one's condition is.
And I think we need to be careful to say that even people living with a great deal of burden can, in the Buddhist tradition, be happy by being in touch with reality. I do believe that. On the other hand, that's easy for me to say — here in a nice house, with the right temperature and plenty of good food, where I'm not at a disadvantage.
I might have to get real first and make sure I've got access to the essentials, and that goes back to the social contract that assures I will. It isn't just about excusing my own circumstances, it's about owning the circumstances of others. That's the Saint-Exupéry quote I gave you earlier: to be a man, to be a grown-up, means to feel responsible for miseries you did not cause. That, to me, is the social contract we must have in our country.
Goldstein: A common assessment of the American experiment is economic prosperity and technological innovation, military strength, democratic endurance, and individual liberty. But our conversation suggests another lens — the health of the American population. Looking at America at 250 through that lens: what gives you reason for optimism, and what concerns you most? I'll frame it by asking you to look forward to America at 300. Will America be a healthier nation by then? And what are the most important choices we need to make now to get there?
Berwick: There's no reason we can't be. I'd say the positive side is abundance — look at what we have: our natural resources, our human resources, each other. The demonstrated creativity and ability of this country to do amazing things, including under crisis — winning World War II, responding to the pandemic better than we have. We have so much abundance, and yet we continue to struggle with a sense of scarcity. Why would we think about scarcity, given the kind of nation we are?
But to reach the goals we're talking about, we have to readjust our scales. It's not okay for wealth to be continually concentrated the way it is, leaving so many people at a disadvantage. It's not okay for wealth to control our political structures, as Citizens United lets it. It's not okay for the size of one's bank account to determine one's power. It's not okay for people who have less to not feel agency. We've got to decide whether that's the kind of nation we want to become.
Can we get there? No question. I think the critical path includes leadership — we need leaders who will call us to our better angels, as the phrase goes, and who can articulate a vision people can buy into, one where no one is left out. Right now, we're in a leadership crisis, and that needs to change. Can we get there? No question.
What worries me most is that forces of division, accusation, exclusion, and fear are now dominating our airwaves. We have a president who's a criminal, and policies that blame people who don't deserve the blame at all. We'd better change that and change it fast.
Goldstein: Leadership is important. And trust — building trust, and the values you've discussed. This has been a privilege for me over the last three conversations, and I can't thank you enough.
Berwick: Thank you, Martin. You've taught me a lot in these conversations — I've got a lot more to think about, thanks to you. I really appreciate that.
Goldstein: You've been very gracious. Thank you.
About the Author:
Martin H. Goldstein is a 2020 Harvard ALI Fellow, a 2021 Senior Fellow, and Managing Editor for the Social Impact Review. Martin has thirty plus years in biopharma, initially at Hoffmann-La Roche, followed by Genentech. Subsequently, as the founder and CEO of ViroLogic, Inc., a clinical laboratory guiding therapy of HIV-infected patients. More recently, he has been working with venture to build biotech companies pursuing cutting-edge science to develop therapeutics to treat disease.
This Q&A has been edited for length and clarity.
Cover Photo Credit: AI-generated editorial illustration, America at 250: Through the Lens of Health