THE TREATMENT FEBRUARY 18, 2010
Harold Pollack is the Helen Ross Professor of Social Service Administration at the University of Chicago and a Special Correspondent for The Treatment.
Megan McArdle and Ezra Klein (and friends) are duking it out over the potential impact of universal health insurance coverage on mortality. Jonathan Cohn has already weighed in, but I wanted to add my two cents.
It’s useful to divide this controversy into two separate questions. First, does health insurance improve health, and if so, how, and by how much? Second, what else do we need to do?
The first question occasions a very old argument. It’s easy to swing widely between cynicism and naïveté. Although every one of us will eventually die, mortality is a rare and complicated event within almost every medical or social science study. Death occurs from multiple causes that unfold over a long period of time. It is difficult to tease out these inter-related causes.
Two excellent recent studies reached quite different results in exploring this question. The Institute of Medicine planted its flag roughly at the midpoint, implying that roughly 22,000 Americans die every years because they are uncovered. The large accompanying uncertainties in this literature don’t indicate that health insurance isn’t valuable. They indicate that it is hard to know when and why such coverage really matters.
The cynical case is surprisingly easy to make. Uninsured people are not randomly selected from the general population. Their heightened risks for various bad outcomes, including mortality, may reflect aspects of their own behavior and circumstances that are not caused by the lack of insurance coverage. Catapulting insurance cards to every American cannot address many social, economic, and even medical determinants of health. Robert Schoeni, James House, George Kaplan, and I co-edited a rather large tome which explores this theme.
There is also the awkward fact that much medical care is of no proven value. With dismaying frequency, such proof is lacking because the care is of no genuine value at all. Many of us are over-insured, which is one valid reason to tax Cadillac insurance plans.
Even when health care really matters, there are backdoor ways to finance health care for the uninsured. Consider what happens to uninsured breast cancer patients. Many present at later disease stages and require urgent and costly care. Faced with such an emergency, providers scramble, and often find some way to provide this care, often at a county hospital or another safety-net provider. (These same providers may then send a whopping bill, or they may foist the cost of her care onto the hospital or onto others, but that’s another matter.)
Popular impressions of the thirty-year-old RAND Health Insurance Experiment--possibly the most important true policy experiment ever done--reinforces the cynical view. The RAND HIE was far from perfect. Nobody was actually uninsured. Indeed the high-deductible plan was surprisingly comprehensive by year-2010 standards. The experiment was short, three years for most participants, and five years for the others. Still the results were striking and appear supported by many later studies. People assigned to free care received 40 percent more services than those assigned to a high-deductible plan. Yet the typical person enrolled in the free-care group was not tangibly healthier. You can see where McArdle and others might be skeptical given such results.
A close reading of the RAND study and of many others, should dispel the skepticism. The RAND study predicted that low-income patients enrolled in a high-deductible plan would experience a 38 percent higher mortality rate than their otherwise comparable peers enrolled in a free care plan. Almost all of this difference reflected differences in hypertension--a special killer within minority communities throughout the United States.
Health studies based on Medicare tell a similar story. I’ve written in this space before about the impressive body of work produced by McWilliams and colleagues. These authors document marked improvement in many health measures at age 65 when individuals become Medicare-eligible. They also document important reductions in race/ethnic and educational disparities in hypertension and blood glucose control. At age 65, racial disparities in systolic blood pressure decreased by 60 percent. Educational and race/ethnic disparities in blood glucose control decreased by more than 75 percent. Educational disparities in total cholesterol levels became negligible.
(If you are up for it, this rack of graphs is from these authors’ work, published last year in the Annals of Internal Medicine. The graphs are a bit complicated, but they are worth a look. In each of the four frames, the first two data points indicate race.ethnic or educational disparities in systolic blood pressure, hemoglobin A1c*, and total cholesterol before age 65. The second two data points represent the same disparities after age 65. In each figure, the bars that surround each point represent the range of statistical uncertainty.)
For these and other outcomes, important health disparities markedly decline once people join the essentially universal Medicare system. Medicare could do a much better job in addressing hypertension, diabetes, and other chronic diseases. Indeed the House and Senate health reform bills both include important measures to promote preventive care and improve disease management. Yet this is a substantial achievement. It’s too bad that Americans must reach age 65 to receive this benefit. It’s doubly too bad that our over-65 population shows such tenuous commitment to providing other Americans with the same financial and health security.
But these are political rather than epidemiological questions. Does health insurance save lives? Almost certainly it does, though the size and the causal pathways of the protective effect remain uncertain. Should we continue epidemiological research to pin down these relationships and improve other, non-insurance strategies to save lives? Absolutely.
On this first money question, then, would universal coverage make people tangibly healthier? You betcha.
In addressing the second question, McArdle’s skepticism deserves a more sympathetic hearing. Suppose we accept that universal coverage could save 22,000 lives every year. That’s a large number, but there are other ways to save thousands of lives that are much more cost-effective than expanding health insurance coverage. We systematically neglect these other opportunities.
Americans often conflate the challenge of improving population health with the challenge of assuring universal access to health care, which we further conflate with universal access to personal medical services through public or private insurance coverage. Each link of this causal chain is open to challenge, and expanded coverage is not the only or even the most powerful tool to improve population health. More than 400,000 Americans die every year from tobacco use, for example. A stiff increase in cigarette taxes (with the proceeds used to finance other needed tobacco control measures) would probably prevent more deaths than universal health coverage would.
We spend $2.3 trillion on medical care. Yet the public health system remains starved of vital resources for HIV prevention, reproductive health services, substance abuse prevention and treatment, and more. The Senate and House health reform bills both included important new public health funds to redress this imbalance. Whether these provisions will survive remains uncertain.
More generally, our society shows a tenuous commitment to investments outside medical care that profoundly affect population health. In the conclusion to our book, Robert Schoeni, George Kaplan, James House, and I note that America quietly lowers our sights in nearly every arena outside the domain of health. For example, Great Britain reduced child poverty by more than half through policies our own nation could replicate for perhaps $100 billion annually. The health benefits of such investments are hard to judge, but would probably be quite substantial.
Of course, an 11-figure increase in income assistance, child care, and other redistributive programs seems politically impossible. Yet our annual increase in national health expenditures regularly exceeds this figure, seemingly on auto pilot.
Ethicist Daniel Callahan muses that every American city includes gleaming hospitals and crumbling schools. That’s not sustainable or wise, even if our only goal were to promote population health. For this reason and others, controlling the long-term growth of health care spending is essential. Health care spending is already crowding out other investments required to address critical national needs.
Nothing I just said provides a good argument against health reform. The economic and health benefits of near-universal coverage are quite large. No wealthy society should allow people to lose their homes because they get sick. Health reform would be a historic achievement. Still, it is only one step we must take to create a healthier and more decent society.
*Hemoglobin A1c is the key measure of blood glucose control pertinent to diabetes.