What the Health Care Debate Still Gets Wrong
A decade ago, Harvard surgeon Atul Gawande helped popularize the idea that U.S. health care spending is high because we use too much medicine. He was wrong: it’s the prices, and who pays them.
October 17, 2019
Oct 17, 2019
25 Min read time
A decade ago, Harvard surgeon Atul Gawande helped popularize the idea that health care spending is high because we use too much medicine. He was wrong: it’s the prices, and who pays them.
Priced Out: The Economic and Ethical Costs of American Health Care
Princeton University Press, $27.95 (cloth)
In the spring of 2009, with the battle over the Affordable Care Act (ACA) in full swing, President Barack Obama called his aides into the oval office for an unusual meeting. As the New York Times reported, the topic of conversation was a recent New Yorker essay titled the “The Cost Conundrum.” It was written by the Harvard surgeon and writer Atul Gawande, now the CEO of Haven—the new Amazon-Berkshire Hathaway-JPMorgan Chase health care venture. His influential story—“required reading in the White House,” the Times called it—described a journey down into the heart of health care darkness: McAllen, Texas, a poor city at the southern tip of the state with some of the highest health care spending in the nation.
What was the root of McAllen’s high costs—and, by extension, of all of ours? Gawande quickly cracks the case. “There is overutilization here,” a general surgeon tells him during the trip, “pure and simple.” Patients went to the doctor too often, had too many operations, spent too much time at the hospital, and received too many days of home care. “The primary cause of McAllen’s extreme costs,” Gawande concludes, “was, very simply, the across-the-board overuse of medicine.”
More important than the question of who paid for health care, Gawande argued, was reforming an entrepreneurial ethos that led to overuse of medicine.
More important than confronting the question of who paid for health care, Gawande argued, was reforming a reckless and inefficient entrepreneurial ethos on the part of medical providers that led to excessive provision of services. His observations squared with decades of research from a group of scholars at Dartmouth. In a slew of influential studies, these investigators demonstrated that health care use (and spending) varies greatly from region to region, and that the people in places that get the most care did not seem to have better health as a consequence. McAllen, Gawande argued, was a microcosm of a whole nation awash in excess health care.
The story didn’t end with overutilizing regions, though. It also included particularly high-utilizing patients. A couple of years after “The Cost Conundrum,” Gawande took a trip to Camden, New Jersey, for another influential essay, describing how some individuals spend a great deal of time in the hospital, owing to a toxic combination of chronic illness and social precarity. In the face of this data, another remedy presented itself: Gawande argued that by “hot spotting”—identifying these so-called “superutilizers” and giving them more intensive outpatient care and social services—we could decrease their trips to the hospital and so “reduce over-all health-care costs” for everyone. The essay was accompanied by an offensive depiction of an obese man, wrapped head to toe in bandages, with a giant “$3,500,000” price tag around his neck—an image that, intentionally or not, suggested that our high spending was the fault of the sick themselves.
The imagery was indeed telling. The conceit at work in both these essays is to imagine our health care cost crisis as fundamentally a technocratic problem of health care overuse driven by a poor alignment in financial incentives. In this, it is largely a specimen of market thinking: pin the tail on the inefficiency caused by individual behavior (in this case, from excessively needy patients and procedure-happy providers), then “realign” market incentives to patch it—but keep the market intact. The policy implications, however, were simpler still: our use of health care had to be reined in.
For more than a decade, nearly all health care cost control strategies were directed at reducing the quantity of medical services we use.
These ideas did not originate with Gawande, but his reporting helped to popularize a broader ethos within the health care community. And those ideas were taken seriously. For more than a decade, nearly all health care cost control strategies were directed at reducing the quantity of medical services we use. Employers, for instance, have raised insurance deductibles year after year in order to deter “excess” care use among their employees. Similarly, the ACA included a “Cadillac Tax” designed to penalize overly generous healthcare plans that, again, ostensibly lead to overutilization.
Obama’s signature health law also incentivized “workplace wellness programs,” financial sticks and carrots that employers wield to prod their workers into healthier lifestyles and, so the theory goes, reduce their health care needs (and, consequently, use). Most significantly, there has been the growth of “Accountable Care Organizations” (a coinage of one of the Dartmouth researchers), financial arrangements in which hospitals stand to lose money when their patients use more health care, as in Health Maintenance Organizations (HMOs). Gawande embraced ACOs as a central solution to the many McAllens throughout the land. In doing so, as health care analyst Kip Sullivan has written, Gawande “channeled” the Dartmouth researchers; both influenced Obama’s director of Office of Management and Budget Peter Orszag, and indeed the president himself. ACOs came to have a key place in the ACA.
And yet, it turns out that this entire edifice of reform was built on sand. Quite simply, as a nation, we actually do not use too much health care; if anything, we use fewer services than people in other high-income countries. While “overutilization” may indeed be a major problem in some areas (and who wants an unnecessary slice from a scalpel?), it cannot, simply as a matter of basic accounting, explain our total off-the-charts spending. In particular, it cannot account for the fact that we spend more than $10,000 per capita on health care—approximately double that of Canada—nor for the nearly six-fold rise in inflation-adjusted healthcare spending from 1970 to 2017, according to estimates from the Kaiser Family Foundation.
The real cost problem, all along, has been the other half of the spending equation: not the quantity of medical services rendered, but the prices paid by insurers for each unit of care provided.
So what can? It turns out that the real cost problem, all along, has been the other half of the spending equation: not the quantity of medical services rendered, but the prices paid by insurers for each unit of care provided. This simple but crucial insight is most frequently attributed to the legendary health economist Uwe Reinhardt, who died two years ago. Reinhardt’s final book, Priced Out: The Economic and Ethical Costs of American Health Care, was published in May, and it provides a cogent synthesis of all the reasons why, as he and his colleagues put it in a celebrated paper in Health Affairs from 2003, “it’s the prices, stupid”—not the quantity of care we receive—that drives our high health spending.
Priced Out provides a useful guide to the U.S. health care system as we head into the 2020 presidential election, and the “price hypothesis” that Reinhardt has injected into health policy conversations has one big advantage over the old “quantity assumption”—namely that it is, roughly speaking, true. And yet it, too, may send us down a garden path again if we are not vigilant to keep the whole system in view. For as Reinhardt’s larger body of work makes clear, we cannot separate high prices from the structural failings of our dysfunctional and regressive health care financing system. Indeed, until we transform who pays for health care, cost containment—and more importantly, health care justice—will remain a distant mirage.
• • •
Reinhardt never shied away from issues of ethics or distributive justice in his work, a fact that may owe something to his early life. Born in the German town of Osnabrück around two years before Hitler’s blitzkrieg on Poland, he spent his childhood in a postwar European landscape of “utter misery and desolation,” as the historian Tony Judt once described it. He grew up in an office-sized “tool shed” within a factory that he shared with his mother, grandmother, and four siblings, he recalled in a 1992 interview with the Journal of the American Medical Association. They lacked running water, and stole fuel and food when they needed it. Yet amidst this poverty, there was one thing that his family never went without. “When we needed medical care,” he said in the interview, “we got it at the local hospital, no questions asked. When you were sick, society was there for you.” At that time Germany already had a nearly universal health care system.
Reinhardt’s new book gives a cogent synthesis of all the reasons why “it’s the prices, stupid”—not the quantity of care we receive—that drives our high health spending.
Reinhardt’s next stop in life may also have shaped his outlook about health care. He set off to Canada at the age of nineteen, and, after three tedious years in Montreal running numbers for the steamship division of an aluminum business, he made his way to the province of Saskatchewan to obtain a degree in business administration. He was there at a transformative moment in North American health care history. In 1961, the province’s premier, the Scottish-born socialist Tommy Douglas of the leftist Co-operative Commonwealth Federation, helped pass Saskatchewan’s universal physician care program, which became the model for Canada’s single-payer system. (It also provoked a twenty-three-day doctors’ strike; resistance to reform was typical among physicians in that era, though that seems to be changing today.) Reinhardt’s orientation toward these events at the time is unclear, to me anyway, but he evidently acquired a deep respect for the equity at the heart of the Canadian system. When Taiwan began reforming its health care system in the late 1980s, and Reinhardt was called in as an advisor, his recommendation was clear: adopt a Canadian-style single-payer system. The Taiwanese followed his advice.
Reinhardt left Saskatchewan for Yale in 1964, where he got a PhD in economics, and several years later headed to Princeton, where he taught and lived until his death in 2017. He lived the life of the public intellectual: he served on various public boards (and consulted for various private companies), taught and debated, advised governments, frequently appeared in the media, was constantly on the lecture circuit, and was renowned for his generosity and wit. Throughout his long career he helped topple some major health policy truisms.
Foremost amongst them was the idea that overutilization was the driving force behind America’s exceptionally high spending, the target of his aforementioned 2003 Health Affairs paper. Comparing statistics on health care use among various high-income nations, he and his colleagues found that the United States was no outlier when it came to the quantity of health care we used. “With the exception of a few high-tech procedures,” Reinhardt summarizes in Priced Out, “Americans actually consume fewer real health care services (visits with physicians, hospital admissions and hospital days per admission, medications, and so on) than do Europeans.” The rub is just that in the United States, the prices paid for each of these services are higher: that is the observation Reinhardt is perhaps best known for today.
As Reinhardt’s larger body of work makes clear, we cannot separate high prices from the structural failings of our dysfunctional and regressive health care financing system.
It took some years, however, for the “price hypothesis” to catch on, and even as it gained ground, the overutilization theory remained hegemonic for years to come. After Reinhardt and his colleagues published their work in 2003, a number of journalists and academics, some who have said they were inspired by Reinhardt, broadened the case that overuse isn’t the problem—it’s screwy health care prices. In 2006 Reinhardt himself explored the byzantine world of “chargemasters,” hospitals’ secretive lists of prices for every drug, supply, and service they provide, which bear little resemblance to actual costs or what insurers actually pay them but that can be ruinous for the uninsured. Far more influential, however, was journalist Steven Brill’s 2013 issue-length article for Time magazine, “Bitter Pill: Why Medical Bills Are Killing Us,” which exposed the hell uninsured and underinsured patients are put through when they are struck by giant, bankruptcy-inducing hospital bills built on these chargemasters.
Although Brill’s “Bitter Pill” represented a major shift in focus from Gawande’s “Cost Conundrum”—the former blamed prices, the latter blamed quantity—they still agreed that who was paying for health care might not be the most important question. “When we debate health care policy,” Brill noted, “we seem to jump right to the issue of who should pay the bills, blowing past what should be the first question: Why exactly are the bills so high?” As I’ll get to, this totally misses the point: the bills are high because of who is paying them. The two questions are not orthogonal.
Other journalists have since taken a similar tack. For instance, Sarah Kliff, formerly of Vox and now at the New York Times, has assiduously collected a database of more than 2000 (often outlandish) emergency room bills from patients—along the way, getting more than $100,000 in medical bills cancelled merely by dragging them into public view. Kliff has written about being influenced by Reinhardt’s work, noting that he “shaped what I decide to report on. It is why I tackle projects that try to bring more transparency to American health care pricing, and the reason I think it's important to tell the stories of the medical bills my readers send me.” Along similar lines, Elisabeth Rosenthal wrote a series featuring outrageous medical bills, “Paying Till It Hurts,” while at the New York Times, and today she publishes a regular “Bill of the Month” feature as editor-in-chief of Kaiser Health News.
Although Brill’s “Bitter Pill” represented a major shift in focus from Gawande’s “Cost Conundrum”—the former blamed prices, the latter blamed quantity—they still agreed that who was paying for health care might not be the most important question.
By exposing the financial and even physiological damage inflicted by medical bills, this reporting has done a tremendous public service, sometimes even directly helping the victims to bargain down or even void their bills. At times, though, these stories can mystify rather than clarify, suggesting that our problem is an opaque medical marketplace that even model “consumers” struggle to navigate—rather than the fact that we have a medical marketplace at all.
Consider a recent “Bill of the Month” story featuring a man described as the “perfect health care consumer”—let the phrase sink in—who is nonetheless surprised by a medical bill for an inguinal hernia repair surgery that was 50 percent higher than the “estimate” he got in advance. The article’s “takeaway” is that while it is generally a “good idea to get an estimate in advance for health care,” such estimates are often faulty. The policy recommendation? “Laws requiring some degree of accuracy in medical estimates would help. In a number of other countries, patients are entitled to accurate estimates if they are paying out-of-pocket.” But one might just as well have concluded: “Laws in a number of other countries, like Canada, make hospital services free-at-point-of-use for everyone in the nation.” Beneath the surface, one can sometimes perceive an implicit embrace of the ideology of health care consumerism in such stories.
At the same time, following in Reinhardt’s footsteps, new lines of academic research have confirmed the “price problem” exposed by this new wave of price-hunting narrative journalism. Past studies, including much of the recent Dartmouth work (and Gawande’s shoe-leather reporting in Texas), focused on variations in use by the Medicare population, mainly because this data is easy to obtain. But when Yale health economist Zack Cooper and three colleagues recently analyzed a new giant data set from three of the five largest private insurance companies, they found something very different.
Asking why the bills are high without asking who pays them totally misses the point. The bills are high because of who is paying them.
The key to this new research is the ability to see through a methodological blind spot in some of the earlier work. Since Medicare rates are administratively set by the government, most of the variability in Medicare spending reflects differences in utilization rates rather than in prices—but this says nothing about spending by private insurers, where prices are determined by the market. In a paper published this year in the Quarterly Journal of Economics, Cooper and colleagues reported that service prices paid by private insurance companies to hospitals varied greatly by region (and even within hospitals), and that hospital industry consolidation was an important driver of higher prices. It is worth noting that in absolute terms, rates paid by private insurers are substantially higher than those paid by Medicare (at least nowadays), and that while Medicare spending accounts for about a fifth of total spending, private insurers account for about a third. As a result, decreasing arbitrary variations in private insurers’ payments for care, rather than reducing the quantity of services Americans use, might be the more effective path to reducing health care spending. Like Reinhardt’s “It’s the Prices, Stupid,” this paper turned the conventional wisdom on its head. The authors acknowledged in a footnote that they “drew inspiration” from Reinhardt and dedicated the paper to his memory.
And yet, Cooper still seems relatively unconcerned with the question of who is paying for health care. In a recent Health Affairs blog, he calls for three policies to reduce high prices paid by private insurers to hospitals: antitrust action to reduce the leverage hospitals have when they negotiate with insurers, incentivizing physicians to refer their patients to lower price hospitals, and regulation of hospital payments in the one-in-five hospital markets considered “highly concentrated.” From this point of view the problem is not private financing, in other words, but merely the balance of power between health care providers and private insurers. We don’t need a universal public insurance system: on the contrary, our private insurers just need more market power. The who, again, isn’t the issue.
• • •
It is a good thing that the new price hypothesis has displaced the old quantity assumption. This change has been propelled not just by the new price-hunting health journalism and research such as Cooper’s, but also by the failure of policy after policy that was engineered to control spending by reducing use. High-deductible plans, for their part, may indeed deter the use of needed health care (for instance, they keep women from obtaining breast cancer treatment), but their proliferation has done little to stem rising private health insurance premiums. The large reported savings from “hot-spotting” “superutilizers” that Gawande described in Camden have not been replicated in larger studies. Workplace wellness programs are not merely despised by workers, but, increasingly, appear to be a total sham when it comes to lowering costs. ACOs, meanwhile—the holy grail of cost containment, according to some—have been shown to produce little to no cost savings. None of this is surprising, of course, once we acknowledge that the very premise upon which these policies were based—that overuse is the problem—was dead wrong.
It is a good thing that the new price hypothesis has displaced the old quantity assumption. But it is constantly at risk of being oversimplified, ignoring the costs attributable to privatization itself.
Still, as Rosenthal’s and Cooper’s prescriptions illustrate, the new price consensus has failed to jumpstart thoroughgoing change, and it may now threaten to cloud the reform debate rather than clarify it. For all the lucidity Reinhardt’s price hypothesis has brought to our understanding of the political economy of U.S. health care, it is constantly at risk of being oversimplified—distorted into an internal problem of markets and divorced from the concern for equity at the heart of Reinhardt’s work. In this increasingly common interpretation of the price hypothesis, privatized payers are simply taken for granted, so the costs attributable to privatization itself—ethical as well as economic, as the subtitle of Priced Out puts it—are simply rendered invisible. Reinhardt, by contrast, recognized the harms of a fragmented and privatized financing system; no doubt that partly explains why he recommended a single-payer system to the Taiwanese (even if he was pessimistic—for political reasons—about the prospects of such reform at home).
The conversation’s neglect of the financing system itself leaves three major questions unanswered. First, what exactly are we talking about when we talk about prices? It is a trickier problem than it sounds. Second, what are those prices actually paying for? And third—the most important—how do we lower them?
The goal should not be to rationalize point-of-service prices; it should be to abolish them.
The answer to the first question might seem obvious, but popular discussions often conflate two very different numbers. When most people speak of health care “prices,” they often have in mind point-of-service prices one pays when picking up a prescription, being hospitalized, having a baby, or seeing a doctor. If the patient is insured and in-network, this price is typically a copay or deductible; if the patient is uninsured or out-of-network, or if a claim is denied by an insurer, it might be an arbitrary and often ruinous number pulled off a chargemaster. Either way, there is one obvious solution to these point-of-service prices: just get rid of them. Out-of-pocket payments, after all, are possible only because people are either uninsured or inadequately insured. But lack of insurance is deadly, while out-of-pocket payments made by those with insurance are associated with negative health outcomes. Study after study has demonstrated that copays, deductibles, and the like deter patients from needed medical care, including sufferers of cancer, diabetes, heart disease, emphysema, multiple sclerosis, and other illnesses. In Canada, the United Kingdom, and Germany, by contrast, point-of-service prices, for the most part, either do not exist or are nominal—and hence entirely divorced from the cost of production. Doctor visits are free in all three nations (at least for the 86 percent of Germans with public insurance); Wales and Scotland have gone a step further and universally eliminated prescription fees, too. The goal should not be to rationalize “prices” of this sort; it should be to abolish them.
However, when economists refer to “health care prices,” they mean the overall payments for a service—not just what the patient pays to the provider in the form of a copay or deductible, but what the insurer pays to the provider on behalf of the patient. Defined this way, of course, prices cannot be eliminated, because goods and services cost money to produce, regardless of who is paying for them. But the distinction between these two ways of thinking about prices leads me to the second problem with the emerging price consensus: the failure to consider what is baked into the payments that payers (whether public or private) make to providers. For one thing, as health economists Katherine Baicker and Amitabh Chandra have written, prices reflect technological innovations, and relatedly, I would add, hospital capital expansion, some of which may be useful, and some of which may be profit-driven and wasteful (and hence controlled). But as Reinhardt makes clear in Priced Out and other work, our private financing system also produces even more blatant forms of waste. Who pays does matter.
The special sauce of cost containment, common to basically all high-income nations, is simple: universalism in conjunction with single source funding.
The year he died, for instance, Reinhardt wrote about how, even putting aside profits, insurance companies spend some eighteen cents for every dollar they collect in premiums on administration costs: “marketing, determining eligibility, utilization controls (e.g., prior authorization of particular procedures), claims processing, and negotiating fees with each and every physician, hospital, and other health care workers and facilities.” Much of this administrative of the waste would simply be eliminated by a universal system. (A common response to this point is that administrative spending in Medicare is too low, opening the door to enormous amounts of healthcare fraud. However, a recent investigation by ProPublica turned this argument on its head. Private insurers, it turns out, do far less about fraud than Medicare—they simply pass the costs down to consumers.) Our freedom of choice of insurer, Reinhardt argued, not only comes at the expense of freedom of choice of doctor (the opposite choice made by those in other nations), but also at a great economic cost.
While it is true that those insurance administration costs aren’t typically considered part of the “price” (i.e. they are not payments to providers), other costs driven by the financing system are. For instance, as Reinhardt describes in Priced Out, hospitals and other providers have met insurers’ bloat through profound administrative distention of their own. Duke University’s health care system, he observes, has some 1,600 billing clerks for 957 beds. The costs of these giant revenue-maximizing insurance and provider bureaucracies are packaged into the prices we pay for health care, adding up to hundreds of billions of dollars in waste a year nationwide (to say nothing of the psychological drain and time suck imposed on patients buried beneath these bills). High prices are not incidental to our reliance on private insurance: they flow, in no small part, from it.
Which brings me to the third and final inadequacy with the contemporary price discourse: it lacks a workable theory for how we could lower them. For instance, some point to greater competition among hospitals as the answer; after all, Cooper and colleagues found that hospital monopolies can charge prices 12 percent higher than those in competitive markets with multiple rivals. But this means that even a vast hospital trust-busting operation—the likes of which, to my knowledge, the world has never seen—would still produce relatively modest savings. It also ignores the fact that many communities may only need one or two hospitals, and that competition has never been the way nations have controlled health care costs. We don’t need to reinvent the wheel here: health care costs have been reasonably well-controlled in almost every high-income nation apart from our own, with the exception of Switzerland, which has the next most expensive system. What do these nations have in common? As the great Canadian health economist Robert Evans, writing with colleagues, described in 1991, the special sauce of cost containment, common to basically all of these nations, is simple: universalism in conjunction with “single source funding.”
High prices should be seen less as the underlying disease than a symptom of the true malady, our uniquely privatized and fragmented financing system.
For the fundamental problem in health care financing—and this is a point made both by Evans and Reinhardt, who calls it a “cosmic law”—is that every dollar in expenditures is somebody’s income. All that income, in turn, creates powerful vested interests. Consequently, the process of cost control is, as Evans and colleagues put it, “fundamentally a political problem, not a technical one.” It doesn’t require complex new financial arrangements to change the behavioral psychology of profit-seeking doctors, as Gawande contended after traveling to McAllen. And it won’t come from a rationalized and more competitive medical marketplace, or from price-savvy medical consumers shopping for better bargains, as some imagine. Instead, as Evans and colleagues noted, to control costs one must build “a payment system in which all expenditures flow through one budget, and then one places that budget in the hands of an agency with the political authority and motivation to limit its growth.”
That is what Canada did in the late 1960s and early 1970s when it built its single-payer system along the lines of the system set up in Saskatchewan when Reinhardt was living there, and it explains why that was the precise moment when its health care cost curve first began to diverge from that of the United States. It also explains why the rate of growth of spending in Taiwan actually slowed after implementing its single-payer system, even when the economic theory of “moral hazard”—that insurance increases use—suggests it might have exploded. What these universal, tax-financed systems have in common is that they have both the incentive and power to control spending.
• • •
Reinhardt helped reorient the health care reform discussion from quantity to price, which was a step in the right direction. Yet high prices should be seen less as the underlying disease than a symptom of the true malady, our uniquely privatized and fragmented financing system.
Until we transform who pays for health care, cost containment—and more importantly, health care justice—will remain a distant mirage.
That system leaves millions uninsured and underinsured—by current counts upwards of 87 million are inadequately insured. It is premised on the notion that private insurers can control costs by forming restrictive provider networks—increasing their market leverage but reducing patients’ choice of providers—but this scheme invariably results in out-of-network bills of the ruinous sort Kliff and others describe.
By the same token, it is our financing system that has accommodated, and indeed rewarded, hospitals that transform into capitalistic, consolidating, revenue-maximizing behemoths—because those institutions can then extract higher prices from payers through greater leverage of their own. It is the way we pay for health care in the United States that has led to an arms race of administrative bloat, as insurers and providers fight over payments with legions of bureaucrats and billers. And it is our financing system that has allowed some hospital systems to flourish and expand facilities of ever-increasing technological prowess and splendor, but that forces others—the unprofitable ones—to wither, and sometimes die.
And in the end, it is not just empirical questions that are at stake, but ethical ones. “Unfortunately,” Reinhardt notes in the prologue of his book, “we are too shy in this country to debate forthrightly the ethical precepts we would like to see imposed on our health care system.”
But debate them forthrightly we must. For above all, it is our financing system that is increasingly giving way, as Reinhardt recognized, to the rationing of care according to economic class, as policymakers seek to control costs by passing them through to patients instead of doing what high-performing universal systems across the globe have long done: control that spending at its source. The way we pay for health care has produced a curious but deadly mix of deprivation and excess. There is no great mystery behind it. It’s the financing system, stupid.
While we have you...
...we need your help. You might have noticed the absence of paywalls at Boston Review. We are committed to staying free for all our readers. Now we are going one step further to become completely ad-free. This means you will always be able to read us without roadblocks or barriers to entry. It also means that we count on you, our readers, for support. If you like what you read here, help us keep it free for everyone by making a donation. No amount is too small. You will be helping us cultivate a public sphere that honors pluralism of thought for a diverse and discerning public.
October 17, 2019
25 Min read time