February 04, 2009

[Harper's Magazine]
Why America won't get the health-care system it needs

The Government Sector

When Congress is in session, Michigan Congressman John Conyers holds a regular public meeting at the Rayburn House Office Building, where, if you happen to be interested in health policy, you are welcome to join like-minded citizens in considering the merits of HR 676, also known as The National Health Insurance Bill. If signed into law, HR 676 would require a single payer (the government) to provide health insurance to every American, which is likely why most Americans have never heard of it. Nearly every other wealthy nation has a single-payer system, but in the United States—or at least in Congress—single payer generally is understood to be too utopian, too extreme, and certainly too socialist for domestic consumption.

I was surprised, therefore, when I went to one of the meetings in July and found a hundred or so people stuffed into a stately conference room. Everyone had a notebook, but no one had the bored look of a political reporter. These were activists, young and mostly black or Hispanic. Conyers, along with several guest speakers, sat behind balusters on a low platform that crossed the width of the room. At the other end, near the door, someone had arranged a banquet table potluck style, with tins of homemade brownies and cupcakes. I pushed my way to one of the few remaining chairs in the back as Conyers, now at the lectern and speaking softly into a microphone, asked whether anyone would like to address the gathering.

The first to speak was a large man in an immaculate green suit. “My name is Kenny Barnes,” he said in a raspy whisper, “and I’ve got an organization called ROOT, Reaching Out to Others Together. It deals with the—my son was murdered, by the way—and it deals with the epidemic of gun violence that’s taking place in the United States of America.” Barnes quickly explained this striking interjection. Children in Washington were being traumatized by a culture of gun violence, and they had little access to mental-health services. A lot of them were being labeled as learning-disabled when in fact what they probably had was post-traumatic stress disorder. They needed help and they weren’t getting it.

Conyers thanked Barnes, and then more people spoke. Each of them told a similarly compelling story. A group of people had been forgotten; they needed help and they weren’t getting it. Some of the groups fit within familiar bounds—minorities with AIDS, for example—but others were parsed to an almost surreal degree of precision. One woman spoke, persuasively, about the special problem of black men who don’t floss. Another addressed the challenge stoplights present to old people who cannot walk across the street in the amount of time it takes for a green light to turn red. Conyers’s aides, watching from seats next to the lectern, would occasionally stand and walk over to someone, whisper in an ear, shake a hand. I wondered what the speakers thought would happen as the result of their varied petitions.

Then two doctors began to put all the divisions and inequities into context. Dr. Walter Tsou, well-fed and graying, first gave a PowerPoint presentation brimming with data about health disparities between various groups in America. We learned that the black infant-mortality rate is still double the white infant-mortality rate, that many doctors are strangely reluctant to recommend cardiac catheterization for elderly black women with chest pain, that Asian Americans had a significantly higher occurrence of hepatitis B than non-Asian Americans until 1993, when doctors began vaccinating all newborns against the disease. Remedying these disparities, Dr. Tsou said, was not a matter of repairing the health-care system. It was a matter of repairing everything. Your health is determined not only by your genes, after all, but also by your environment. And that environment is determined by the rules society itself sets up—rules about who lives in what place, who goes to what school, who gets what job. “Until we actually address the social determinants of health,” Dr. Tsou said, “we will not truly eliminate health disparities.”

The next speaker, Dr. Robert Zarr, continued the line of thought. “The single most important reason why we see these disparities is lack of health insurance,” he said, with staccato confidence. “That is the truth. It’s the truth for those of us who have gone periods of our lives without health insurance. It’s the truth for my patients.” Dr. Zarr explained that he is the director of pediatric medicine for a group of community health centers that process more than 60,000 pediatric visits a year, and that most of the children who come through have a shaky connection at best to any kind of benefits. Without insurance, he said, there was not much he could do for these kids. “What good is it if I write them a prescription for the antibiotic if they don’t have money to go to the pharmacy to get it? What good is it that I diagnose dysplasia of the hip of a baby if I can’t get him in to a specialist to get seen?”

A natural salesman, Dr. Zarr then asked his audience some leading questions. “What if I told you there’s an answer right now, right here and today? There is an answer to getting rid of this single most important barrier. Can anybody tell me what that answer is?” Several people in the audience, anticipating what would become a Dem ocratic campaign mantra, shouted out: “Universal health care!” But Dr. Zarr was indignant. “Not just universal health care—even President Bush talks about universal health care!—single-payer universal health care.” Then he lowered his voice. “Now let me tell you why. HR 676 clearly is going to give lifetime, comprehensive, quality access to care to every single American. Keep it simple. That’s what it is going to do.”

This was a strong claim, of course. Single payer would not end racism. Poor people would still be poor and sick people would still be sick. There was no doubt, though, that in a single-payer system the whole idea of “forgotten groups” simply could be eliminated. Instead of separating the healthy from the sick, the high-risk from the low-risk, the rich from the poor, a single payer would unite all Americans into a single system. There would be no tiers, no ghettos, no red lines, at least not in terms of access to health insurance, because a single payer—the government—would cover everyone.

There was one phrase we had to remember, Dr. Zarr said, and it was this: Everybody in, nobody out. “Say it with me. Everybody in, nobody out.

A Preference for Markets

The argument for single payer is straightforward. When everybody is in, you don’t have to spend a lot of time and money deciding who to keep out. You also don’t have to worry about what to do with the people you’ve kept out when they get sick anyway. (Uninsured sick people cost insurers nothing, but since they often end up seeking expensive emergency-room treatment, they cost taxpayers a lot.) If you want to quit your job and work someplace else, you can do so without fear of losing your health insurance, which means that labor is more mobile. And employers don’t have to carry the burden of benefits, which means that capital is more mobile. If you get sick, you don’t have to worry about losing your coverage or your house. Your insurance is paid for through taxes. And your taxes don’t go up just because you have a preexisting condition; under single payer, there is no such thing as a “preexisting condition.” Moreover, your provider—the single payer—has an incentive to keep you healthy your entire life, rather than just getting you to age sixty-five and then dumping you into Medicare. And if the experience of most other countries is any indication, the whole thing would cost a lot less than our current bloated mess of a system.

The benefits of single payer were at one time if not a matter of consensus then at least a topic considered worthy of discussion, at least among Democrats. “I happen to be a proponent of a single-payer health-care program,” Barack Obama said in 2003. “As all of us know, we may not get there immediately. Because first we have to take back the White House, we have to take back the Senate, we have to take back the House.” And yet as Democrats began to take all of those things back, Obama began to reconsider. In 2007, he recast the debate in terms that were more reflective than prescriptive. “If you’re starting from scratch,” he told The New Yorker, “then a single-payer system would probably make sense. But we’ve got all these legacy systems in place, and managing the transition, as well as adjusting the culture to a different system, would be difficult to pull off.” And now that Democrats have the White House, the Senate, and the House, it is clear that a single- payer program is not a part of their agenda.

Something is going to happen, though. That much is certain. And it probably will be similar to the approach set forth in a white paper this November by Montana Senator Max Baucus, who is chairman of the Senate Committee on Finance. The plan borrows ideas from (among many others) Hillary Clinton, incoming Secretary of Health and Human Services Tom Daschle, and Obama himself. The details are vague, but the outline is clear. It achieves universal coverage by requiring Americans who do not already receive health benefits to purchase insurance from a private company. (Obama has proposed that such a mandate should cover only children, but Daschle—whom Obama has charged with overseeing the reform process—has called for the mandate to be universal.) In turn, most employers would be required either to provide benefits to all of their employees or to pay into a fund that would be used to subsidize the purchase of private insurance by those who could not afford to pay for it themselves. This approach is designed not only to assuage the concerns of the many Americans who do not want to change their present arrangements but also to keep America’s health-insurance plans—which employ half a million people, and which saw a major decline in profits in 2008—in business.

It would not be unfair to describe the Baucus approach as “market-oriented.” This may, in fact, be why it has emerged as an acceptable locus of reform. In Washington, there is little that is considered wiser or more bipartisan than a preference for markets. And that preference can even be expressed in terms that are surprisingly far from the standard Cato Institute talking points. Jill Quadagno and Brandon McKelvey, researchers at Florida State University, for instance, report a widely held vision of so-called consumer-directed health care, which would inject an almost Naderite devotion to consumer awareness into discussions about health care. The goal, they write, “is to transform patients into informed consumers by making medical care into a commodity that is purchased in the same way as other market goods.”

This preference for markets is common, but it is not wise. The health-care system is not at all like other markets, because health, for obvious reasons, is not at all like other goods. (The demand for not dying, to give just one example, is pretty much unlimited.) And in America, market-based solutions very often end up involving the government anyway, as has been made evident most recently in the aftermath of the failed deregulation of Wall Street. Thus far, as Quadagno and McKelvey note, the consumer-directed health-care vision actually has “been implemented through obscure changes in tax law, technocratic provisions added to bills designed for other purposes, experiments with Medicaid ‘waivers’ and a new option, Medicare Advantage,” which introduces supplementary private insurance into the Medicare system. Which is to say that no matter what happens this year, the dead hand of government will continue to direct the flow of health-care dollars in the United States.

Nonetheless, Democrats clearly do not want to discuss the role of government in terms that could be understood as unfriendly to the market. “We all have to keep an open mind on all this stuff, figure out how to get to yes. Everything is on the table,” Baucus had cautioned. “The only thing that’s not on the table is a single-payer system. That’s going nowhere in this country.”

Which is why I was in Washington. Even the new president, with a near landslide victory and a huge congressional majority, sees an intraversable divide between what he himself has claimed to understand as the best approach and what can actually be done. I wanted to know what defined that divide, and why single payer fell on the far side.


No one doubts that fixing the health-care system is going to require someone to make difficult choices. In 2006, Americans spent $2.1 trillion on health care—at $7,026 per person, more than any other nation—and yet we lag far behind other nations in such measures as infant mortality, life expectancy, and early detection of life-threatening illnesses. This bad bargain is irritating, of course, but, more important, it is also unsustainable. Aging baby boomers will increase their demands on the Medicare system even as the government faces the revenue shortfall that will result from their retirement. Employer-based health care, meanwhile, is increasingly unaffordable, causing many companies financial distress. And even as the cost of the system goes up, a growing number of Americans are being left out of it entirely.

The word for making such choices, so often unsaid in American politics, is “rationing.” All health-care systems, no matter how wealthy, require some form of it, because advances in technology always outpace the ability to pay for them. But there are (at least) two ways to decide who gets what in a health-care system. One of them is to let the market sort it out: those with the most money get the most care. The other is through triage: society seeks to determine, within a given budget, the most effective treatment for the greatest number of people. The difference between these two approaches is significant.

We tend to think about systems in terms of our personal motivations. But systems have their own logic. A market system may be driven by individuals or corporations seeking profits, but the primary function of the market itself is to grow. That is why growth, not profit, is the conventional measure of economic health. Similarly, a health-care system may be driven by individuals seeking to improve their own health, but the primary function of the health-care system itself is (or ought to be) to ensure the overall health of society. Within each realm, these goals can sometimes be in conflict. The success of banks at maximizing profits for a time by extending shaky loans to prospective homebuyers, for instance, has resulted in a recession (i.e., negative growth). Similarly, individuals acting on various beliefs about health care—say, that vaccination leads to autism—can cause the health of society as a whole to decline. More important, though, is that these two realms themselves be understood as independent. Societal health and economic growth are not mutually exclusive, but they are in tension, and when we confuse one with the other, problems occur.

In Overtreated, Shannon Brownlee argues that the major problem of health care in the United States is not that there is too little but that there is too much. “We know that people who don’t get enough care have a higher risk of death,” Brownlee told me. “About 20,000 Americans die prematurely each year from lack of access. But getting unnecessary care isn’t any better for you. In fact, about 30,000 Medicare recipients die each year from overtreatment. This sounds counterintuitive until you think about the fact that practically any medical treatment you can name poses some risk.” For instance, doctors regularly test prostate-specific antigen levels in men to see if they have early signs of prostate cancer. As Maggie Mahar, the author of Money-Driven Medicine, explained it to me, this sounds like due diligence, but in fact the National Cancer Institute does not recommend routine PSA testing, because the majority of older men diagnosed with this slow-growing cancer will die of something else before they experience any overt symptoms, whereas if they are treated for prostate cancer, many will experience such side effects as erectile dysfunction, incontinence, and sometimes even death. “When I was at a conference in Berlin last spring, doctors from other countries were shocked that we still do routine PSA testing,” Mahar said. Why do we do it then? “Urologists stand to gain. The prostate-testing market is worth $200 million to $300 million annually. And no doubt many urologists believe they are saving lives.”

Overtreatment, of course, is another word for growth, and it is the natural consequence of a market-driven system. A triage approach, meanwhile, would save money, both by removing some (though not all) of the incentive to overtreat and by simply eradicating the massive bureaucracy that currently is required just to figure out who is paying for what. Physicians for a National Health Care Program notes that “private insurance bureaucracy and paperwork consume one-third of every health care dollar” and that going to a single-payer system “would save more than $350 billion per year.”

So the mystery remained. Why is single payer off the table? At Conyers’s meeting, it was Dr. Zarr who presented what at first seemed to be the most plausible theory. “You’ve got to get rid of the middleman,” he said, “and that middleman is the private health-insurance industry. And they have got to go.

The Insurance Sector

America’s health-insurance plans are represented in Washington by an organization that is called, plainly enough, America’s Health Insurance Plans. Its headquarters happened to be a very short walk from the Rayburn House Office Building, and I had managed to convince its president and CEO, Karen Ignagni, to spend a few minutes with me. When we sat down in AHIP’s sleek conference room, I mentioned that I had just been at a meeting on the Hill where people were discussing how to enact a single-payer system.

“Oh good,” she said. “That’s great!”

Ignagni, who is brisk and extraordinarily attentive, had worked for a time at Walter Reuther’s Committee for National Health Insurance, a background that might strike some as pretty distant, philosophically speaking, from AHIP. I asked if she was surprised to have ended up representing insurance companies. She said it was not something she had planned on. She seemed, however, to have adapted well to the role.

I explained that I had been thinking a lot about information. The profit in the current system, after all, comes not from acquiring as many customers as possible but rather from creating two classes of possible customers—good risks and bad risks—and avoiding the latter class entirely. As we get better at understanding why people get sick, we will also get better at deciding whether or not to insure them. Ultimately, the entire nation could be reduced to two perfect circles: the people who pay for insurance and don’t need it, and the people who need insurance but can’t pay for it. “I mean, asymptotically,” I said, “you will slowly approach perfect knowledge . . .”

“Which will be a terrific thing for patients, a terrific thing for clinicians,” Ignagni leaped in. “It’ll be a terrific thing in terms of actually improving health.” Which is true. Genetic medicine, everyone agrees, will likely help millions of people enjoy longer, healthier lives. But Ignagni was not addressing my concern about groups. Doctors can use evidence to achieve their ends, I proposed, which presumably are to improve the health of their patients. But insurance companies can also use it to achieve their ends, which presumably include reducing medical losses. “But that’s not true in the health-care arena,” Ignagni said, “because they passed the genetic nondiscrimination bill.”

Ignagni was referring to the Genetic Information Nondiscrimination Act, or GINA, which was made law last May. Under GINA, it is illegal for employers or insurers to deny anyone health coverage on the basis of genetic data. If people feared losing their insurance because their very genes could be understood as a preexisting condition, they might avoid seeking information that could help them stay healthy. Ignagni said that the health plans had thought about the law a great deal. They didn’t want to lose a powerful underwriting tool, but ultimately they decided that the benefit, in terms of long-term health, outweighed the cost, in terms of inefficiency of pricing. My concerns, therefore, were unfounded. “We would have, I think, a very different conversation today if the legislation hadn’t been passed,” Ignagni said, but “this perfect information that you’ve talked about is, I think, a marvelous opportunity to actually reduce health-care costs.”

Her reasoning seemed humane and, for the moment, economically sound. Genetic information still makes up only a very small proportion of what underwriters examine to determine health risks. But the trend is clear. What will health- insurance plans do as larger and larger categories of health information are determined to be off limits? As the database grows, so too will the temptation to revise GINA, to use some part of the data, perhaps, to make decisions about who to insure and who not to insure, just for the sake of efficiency. What else are insurance companies for, after all, other than apportioning risk? How could AHIP support a law that logically concludes in the demise of underwriting?

Ignagni’s answer was not what I expected. In fact, she echoed precisely the words I had heard Dr. Zarr chanting just a few hours earlier. “In the new market,” she said, “everybody’s going to be in. And then—and I don’t want to be an irresponsible Pollyanna about it—but if you have everybody in, you have the large numbers working for you.”

In retrospect, I should not have been surprised. The preference for markets is more often claimed than felt; the preference for profit is far more sincere, and the method by which it is achieved—competition, bribery, lobbying—is a secondary concern. After Ignagni and I met, when the Obama transition team had made clear that some form of universal health care was forthcoming, she announced that AHIP would support a law requiring private insurers to provide insurance to all people regardless of their medical condition—a form of insurance known as “guaranteed issue”—if Congress would in turn require all Americans not covered by government insurance programs to buy some form of private insurance. This combination of guaranteed issue and individual mandates would add up to a system wherein the government requires healthy people who do not want insurance to buy it anyway, in order to subsidize unhealthy people who need insurance but can’t afford it—which sounds like what most people would call “socialized medicine.”11. In a notably succinct press release, Rose Ann DeMoro, the executive director of the California Nurses Association/National Nurses Organizing Committee, called the AHIP proposal a “Marshall Plan for the health insurance industry” that “fully privatizes profit while socializing the health-care risk. The public systems could be bankrupted by their responsibilities to care for the sickest while guaranteeing huge new profit streams” to insurance companies, which would continue to avoid selling insurance to anyone who actually needed it. “Rather than subsidizing these industries,” she concluded, “we would be better off either letting them fail, or simply taking them over, as we have been forced to do with other obsolete sectors.”

A Preference for Invention

If the insurance companies themselves were openly endorsing a non-market solution—albeit one that required millions of new customers to buy their products—then what else could be preventing Americans from embracing single payer?

Another possibility is that Americans believe technology itself will be their salvation. If nothing changes in the next decade, at least one fifth of our economy will end up devoted to health care. And most of that money will be spent not on basic care or preventative treatment but on expensive new technologies such as thallium heart scans. One report, from the Center for Studying Health System Change, suggests that new technology may account for as much as two-thirds of spending growth.

So here was another clue. One of the major trends in U.S. health care is “evidence-based medicine,” which calls for making medical choices by comparing empirical evidence about an individual patient’s condition to a larger body of best practices. This may sound like common sense, but medicine for most of history has been imprecise, decentralized, and as much an art as a science. With extremely complex and expensive genetic and proteomic procedures increasingly defining the future of medicine, however, doctors—and insurers—will come to rely on the same industrial practices that previously made it possible to manufacture jet fighters or set up an international retail operation. At least that is the hope.

Americans put a considerable amount of faith in their nation’s industrial capacity. A 2001 study found that 45 percent of Americans “disagree that it is impossible for any government or public or private insurance to pay for all new medical treatments.” That is to say, they believe that the U.S. health-care system has the potential to pay for every single new treatment that someone invents. The only nation with a more positive outlook is France, where 51 percent believe this and where socialized medicine is considered a birthright. (The European average is 36 percent.) This apposition may not be as odd as it seems at first. Many Americans appear actively to desire socialized medicine, even by that name. In one recent survey, a 45 percent plurality of Americans claimed to prefer a system of “socialized medicine.” And another survey found that 59 percent of American physicians now support some form of national health insurance, up from 49 percent in 2002. Here was evidence, contrary to the Washington consensus, that in the American faith, markets were an easily discarded icon—our preference was for something far deeper and stranger.


There is a great deal of literature available on health reform, and most of it is just as colorful as you might expect. Every once in a while, though, you come across something unusual. One recent such exception is Skin in the Game, which was written by John Hammergren, the chairman, president, and CEO of McKesson, a health-services corporation in San Francisco. The title suggests that Hammergren is making another boilerplate argument for consumer-directed health care. He is a fan of evidence-based medicine, believes that health care can be understood as a commodity, and his argument builds from the claim that people don’t currently have enough “skin in the game” because the current structure of health-care delivery hides from them many of the costs of the decisions they make. What is fascinating about Hammergren’s argument, though, is what he proposes as the means by which consumers will be made to understand those hidden costs.

Hammergren shares with his Silicon Valley neighbors an abiding faith in the power of technology to improve the world, and that technology would apply not only to the treatment but to the system itself. A smarter system of health care would first recognize itself as a system, and then it would attempt to perfect itself. This would require a kind of scientific management of the human body, in which health providers analyze every part of the patient’s interaction with the system. The mortal coil would be, as much as possible, shuffled into a controllable digital component. Hammergren’s prescription has the flavor of the early twentieth century—what he calls the “golden age of management and operational efficiency”—when Frederick Taylor and other corporate philosophers began to equate rationality with profitability. A hopeful time.

Hammergren first argues that technology will revolutionize specific forms of treatment. Heart surgeons, for instance, will deploy “caterpillar robots” that crawl through a small incision into your heart. No rib cracking, no collapsing of the left lung, no hands inside the chest cavity. It is an expensive technology, Hammergren writes, but one that is only a few years away. Eventually, molecule-sized robots may be able to repair individual cells and even strands of DNA, with the result that people will be able to live two hundred years “without showing any signs of aging.”

But Hammergren’s real concern is much larger. The robot caterpillars will be wielded in service of what he and others call “personalized medicine,” a numbers-driven approach to treatment that tailors every decision to your individual genetic makeup. Personalized medicine “goes beyond the idea of genetic testing to an entirely new level of care, and in turn a higher quality of life,” he writes, and it will require its own entirely new system of data collection—its own aggregation of large numbers:
When my oldest daughter has her first child, I believe that baby will get a genomic profile for roughly $800. The data obtained through that profile will be stored in a central information system, called an integrated delivery network (IDN), to which primary care physicians and specialists will have access throughout the course of my grandchild’s life. Within the IDN database there will be a kind of artificial intelligence search engine—based on the principles of semantic knowledge and driven by complex algorithms—that can support physicians in their decision-making and recommendations.

My grandchildren’s doctors will know from the moment of birth the likelihood that they will develop some form of chronic condition, cancer, or other significant illness. This knowledge will shape and form their health care for the rest of their lives. Compared to today’s 40-year gap in treatment, my grandchildren will receive constant monitoring and prevention. Tapping the database’s artificial intelligence, their doctors will know which clinical interventions will be most effective, which cardiology or cancer drugs they will respond best to, and when care should be delivered.
Hammergren’s culminating vision of an integrated delivery network is simultaneously deeply idealistic—indeed hopeful—and disturbing. With all its tender humanity and seductive hubris, the passage reads as the first chapter of a cautionary tale.

The Technology Sector

McKesson, it turns out, is the eighteenth-largest corporation in the United States, and the largest corporation of any kind that is involved primarily in health care. In 2008, its various businesses—the distribution of drugs and surgical supplies, and the sale of information systems for all aspects of health care—generated nearly $94 billion in revenues. By comparison, revenues for UnitedHealth Group, the country’s largest health- insurance provider, were just over $75 billion. McKesson processes about 80 percent of all the prescriptions written in America, nearly 10 billion transactions a year—more than Amazon and eBay combined.

The company cultivates a low profile. I lived about eight blocks from its world headquarters—a thirty-eight-story black slab on Market Street—at the time of the dot-com bubble, but until I read Hammergren’s book, I had no idea the place existed. Still, when I called the company to see if I could talk to someone about the integrated delivery network, they not only were receptive to the idea but also suggested that I tour the company’s Vision Center.

Which is how, in August, I came to be standing with Tracy Webber, who oversees the Vision Center, and Randy Spratt, who is McKesson’s CIO, in front of a museum case filled with ancient bottles and advertisements. “This is where we honor our 175 years of history,” Spratt said. McKesson had entered the world as a distributor of imported worm seed, effervescent sodium phosphate, soap bark, and Russian oil, all on display. One yellowing flyer proclaimed that McKesson’s imported Russian oil “lubricates and aids excretion without harmful medicinal action.” I asked Webber, who was exceedingly friendly and well informed, if Russian oil had been a cure-all. “I’m sure,” she said.

Webber pointed to several three-foot-thick stacks of dusty Post-It-sized slips of paper that had been pierced with a wire and now hung like sausages from hooks at the top of the case. “That’s one year’s worth of prescriptions from 1910. That’s the very first pharmacy-management system.” She then pointed to what looked like the handheld scanning device that UPS drivers carry to track their deliveries. “This is the Mobile Manager 100,” she explained. McKesson was the first company to use bar-code scanning technology for distribution. With the Mobile Manager 100, McKesson packers could fill orders for drugs with 99.98 percent accuracy. In Skin in the Game, Hammergren had suggested that doctors eventually would use similar devices to keep the integrated delivery network up to date on a person’s condition—they would be the nerves at the tips of the IDN’s fingers.

Webber guided us to the next exhibit, a computer monitor that was displaying what appeared to be a simple email program. She explained that this system would allow me to interact with my physician when I had basic questions that didn’t require an actual visit.

I said that sounded useful.

“Well, physicians would never respond to you,” Webber said, “because they’re not being paid to respond to you.” Spratt explained the solution. “The trick to it was going out to the payers and getting them to reimburse the physician for an electronic visit, which we have done.”

Payers would benefit because it would cost less than an office visit?

“Right,” Spratt said. “The physicians make out because they can do ten of them in the time it took them to do one office visit. And the patients make out because of convenience. And the payers make out because it costs less to treat the patient in total.”

McKesson, it turned out, also handles about a third of the nation’s health-care claims transactions. Indeed, as we wandered into the Vision Center, it was becoming obvious that the company had at some point traded up from the distribution of cure-alls to the far more profitable business of distributing information, though sometimes it was hard to tell which business was which. Pills certainly are amenable to information metaphors—little bits of data, distributed as efficiently as possible across the country, each with a discrete task and a clear profit margin; a rush of pills through the system, all of them marked, radio-frequency identified, and tracked perpetually like the packets that flow through the Internet. The process of distributing pills, like the process of distributing data, is non-intuitive, does not require the human touch, and is in constant peril of failure due to human error.

I asked Spratt about this blurring of functions, and he nodded. “Some of the most critical processes in health care are the processes around safety,” he said. “And then there’s another set of processes that, essentially, surround the translation of you into information and then the reading of that information and the response to that by a caregiver.”

This sounded familiar. In the 1990s, the process was called “bit-from-it,” and the idea was that solid matter was too difficult to manipulate, whereas bits—information—were much easier to control. Business consultants and futurists made their fortunes by explaining to executives that bits were the future and atoms were the past. I was a little embarrassed to be using decade-old pop-business jargon, so I asked Spratt if the company had a more modern term for bit-from-it. “Oh,” he said. “We would call it adoption.”

McKesson, it was becoming clear, was serious. “Some people, fearing change, say we shouldn’t automate the process of health care because we will lose the human touch,” Hammergren had written. “I believe that the human touch can only be reclaimed by relying on automation.” Such claims seemed ambitious and even somewhat absurd in print. Spratt had been careful to point out, though, that everything at the Vision Center could be purchased and put to use today. The inventions I had seen thus far were just the beginning, a primitive sensory apparatus for an evolving system. I asked whether the information systems at the Vision Center were the prototype for an integrated delivery network.

Spratt nodded again. “We think this is the most effective way to get there,” he said. There had been many experiments with regional databases, but nothing had yet caught on in terms of a large-scale adoption. The plan was to get the doctors and patients involved, such that they built the database from the bottom up with their own hands, by entering millions upon millions of queries, diagnoses, and prescriptions into various McKesson systems. “We don’t need to build giant databases in the sky,” Spratt said. The databases would simply emerge over time.

I looked at the monitor again, and then asked Spratt where the information for the system in front of us resided.

“This resides in servers that we own and operate,” he said.

Interlude: the Measure of Man

We continue to invent devices that “translate us into information,” but we sometimes forget that such measurement itself was an invention, and a revolutionary one at that.

Marcus Vitruvius Pollio, the great Roman architect and engineer, for instance, considered the structure of the human body as a matter of qualities, not quantities; he saw not numbers but rather spectacular symmetries: “If a man be placed flat on his back, with his hands and feet extended, and a pair of compasses centered at his navel, the fingers and toes of his two hands and feet will touch the circumference of a circle described therefrom.” And so from this insight, Da Vinci painted his famous Vitruvian man, a picture of proportionality and health. A perfect specimen.

But it is not the Vitruvian man that is stored in McKesson’s servers. Proportion is not the same as measure. Proportion relates one reality to another. Measure transports reality to a virtual realm. It allows us to translate a material fact into an idea, and ideas, in their Platonic perfection, are considerably more amenable to the strictures of rational analysis than the raw grit of nature. “What else can the human mind hold besides numbers and magnitudes?” asked Johannes Kepler in 1599. “These alone we apprehend correctly, and if piety permits to say so, our comprehension is in this case of the same kind as God’s, at least insofar as we are able to understand it in this mortal life.”

It was by using this godlike ability to translate material qualities into abstract—and therefore controllable—quantities that the physician William Harvey was able to discover the circular nature of the flow of blood in humans, a discovery equivalent in physiology to Galileo’s claim of heliocentrism in cosmology. Caspar Hoffmann of the University of Altdorf reflected the beliefs of the day in a rebuke to Harvey. “Of a truth,” he chided, “you do not use your eyes or command them to be used, but instead rely on reasoning and calculation, reckoning at carefully selected moments how many pounds of blood, how many ounces, how many drachms have to be transferred from the heart into the arteries in the space of one little half hour. Truly, Harvey, you are pursuing a fact which cannot be investigated, a thing which is incalculable, inexplicable, unknowable.”

Hoffmann was a man of the past, though. Even as Harvey introduced new techniques of measurement to medicine, others were already introducing them to commerce, notably with the invention of double-entry bookkeeping, which led shortly thereafter to the invention of the corporation. Indeed, the new truth, as Walter Burley of Merton College was quick to observe, was that “Every saleable item is at the same time a measured item.”

These seem like ancient insights, but our preference for invention has only amplified their relevance. And when the measure of man and the measure of markets combine, the results are positively futuristic. The IBM Institute for Business Value, for instance, envisions a new age of “just-in-time insurance” in which “maximum efficiency in risk pricing” would be achieved by segregating the components of customers’ lives into measurable “spaces” through which they journey in a given day.

Each step of the journey represents a different risk such as car-to-train-station, train-to-city-station, station-to-office, and so on. Each leg of the trip truly represents a varying amount of risk. A “pay-as-you-live” product would trade some location and time-of-day privacy data for lower insurance bills overall. And in the spirit of active risk management, the same network of sensors could also provide convenient information (such as advice on avoiding an overloaded expressway) relayed on the appropriate device such as the car audio system, a phone, and, then, in e-mail or as a phone call in the office.

Pay as you live! Just so.

The Technology Sector

As we continued our exploration of the Vision Center, it was becoming clear that McKesson had also considered the problem of transforming data back into action. At some point, the prescription would become a pill, and the pill would become a cure. A real person would really be healed or not. But the translation from the information realm to the physical realm is often highly flawed. One major breakdown is in the simple process of selecting a pill. A large hospital delivers thousands of doses a day, and each of those deliveries presents a chance for error. A 1994 study in the Journal of the American Medical Association found that patients in intensive care underwent an average of 178 “activities” per day, and that 1.7 of those activities were in error—a 1 percent failure rate, which is quite high compared to what engineers would find acceptable in assembling airplanes or nuclear power plants.

The McKesson solution, as Spratt was now explaining to me, was to remove people from the process. We were standing in front of a six-sided device that looked like the game-show cage into which contestants are locked with a vortex of swirling loot and invited to grab all they can. This was a far more precise system, however. At the center was a robot arm. And on each of the six interior walls were two dozen spindles, each hung with UPC-labeled plastic bags full of candy, which stood in for pills. The entire contraption was “an automated inventory-management tool,” Webber said—basically, an extremely reliable vending machine.

Webber pushed a button and the machine leaped to life with a series of Star Wars –type servo noises. “You see these little packets?” she asked. “We dump unit bar-coded inventory into the robot. It stocks it on the right spindle, and then it intercepts orders from the pharmacy, and picks those orders, plops them in a name board, labels it with the patient’s name, and that’s your medication for today.”

This seemed like a fairly elaborate approach to selecting an item from a bin. I asked if it was popular. “To my knowledge,” Spratt said, “we have never had one uninstalled.”

Just as a computer can be made to understand the contents of a roomful of pills, it also can be made to understand the contents of an entire hospital. Spratt pointed to another monitor, which displayed a floor plan. “This is a map that links into the systems that say, for example, ‘This guy’s meds are due,’ ‘This guy needs to be transported to get an X-ray,’ ‘This guy is due in the ICU in nine minutes,’ and so on.” It was another feeler for the future IDN. “We can take patients, we can take high-value assets like pumps or parts on wheels, and we can put tags on them that would tell you where they are in the hospital, so if I need a pump right away, I can find the closest one.”

I asked if that meant hospitals would have to tag everything with some kind of radio-frequency identification device.

“Everything of significant value,” Spratt said.

What about at home? Was there a way that people at home could be similarly virtualized for more efficient analysis and treatment?

Spratt nodded again, and Webber explained, “So the Health Buddy is our home monitoring device that would be placed in the homes of patients who are suffering from chronic ongoing diseases.” The Health Buddy looked like a clock radio. It had a very simple display—large letters and icons to make it easy for old people to comprehend—and, like a video-game console, it was capable of accepting various input devices, including a blood-pressure cuff, a scale, and a blood-glucose meter.

“That’s all transmitting back to a caseworker or care coordinator who is trying to decide how often to send a nurse into the home,” Webber explained. What sort of data does it track? “I really think something interesting might be the fact that you got out of bed,” Webber said. “You have a sensor that shows that the patient is being active and has picked the phone up.”

I asked if the Health Buddy would eventually be capable of actually treating people, for instance by taking genetic or proteomic data and simply manufacturing the appropriate prescription on the spot. “That would be very long range,” Spratt said. “But it would be, for example, very easy—we’re not there yet—to have it, with a caregiver intervening, electronically prescribe a new drug and have that delivered to you overnight.”

In the meantime, there was our next stop: the automatic pharmacy machine. It looked like a regular ATM, only with a door flap like on a candy machine instead of a bill dispenser. Webber pushed a button and a robotic voice said, “Please enter your data first.”

“One of the biggest problems is finding new pharmacists who are willing to count pills,” Spratt said. “If you have a prescription and a retail automation pharmacy system, this device here can automatically count the pills from a pretty good sized inventory of potential molecules.”

The robot voice said, “Please enter your PIN.”

Spratt continued, “So it will say, okay, you get thirty—in my case, maybe Xanax—and it’s going to dump thirty Xanax into here.”

The robot voice said, “Please touch the boxes below to select your prescriptions.” Then it said, “Please sign your name to acknowledge that you are receiving medication that has been billed to insurance.”

A Preference for Ideas

I had been, and remain, skeptical of Hammergren’s vision of data-driven salvation, but as I left the McKesson office, I did find myself unexpectedly drawn to the idea that our fallen world could be reborn within a system of our own creation. And as I considered my own strange compulsion, I began to suspect that the preference for markets and the preference for invention were just subsets of another, deeper preference—so deep perhaps that we were not even quite aware of it.

Anyone who has spent any time fighting for the health of the disembodied entity known as a corporation knows that disembodiment is itself a primary advantage. The human body is frail, after all, and subject not only to the physical laws of life—somatic decay and other factors that cause life to require death—but also to the physical laws of all material, living or not. Newton and Yeats agree that everything must fall apart eventually. Corporations, on the other hand, are inventions of the mind. They exist as agreements, brands, and statements, none of which answer to any of the laws of thermodynamics. Corporations can live forever, because they are really just ideas.

Of course, only the most outrageous West Coast transhumanists actually believe they can cast off the mortal realm of meatspace for the immortal realm of the virtual. Those of us who are not making plans to enter cryogenic storage, however, still are part of a culture predisposed to virtuality, a culture where in every realm of endeavor, from industry to politics to art, the word trumps the deed and the immaterial emotion trumps the material fact.

In our culture, though, this will to dematerialize remains, perhaps somewhat ironically, unarticulated and inchoate. This may explain our strange, indeed neurotic, sense that the needs of the “system”—whether it is the system of governance or the system of markets or the system of technological innovation—are now more important than the needs of the individuals such systems once were assumed to serve. We are a nation of closeted Pythagoreans, ashamed of and enthralled by our secret desire to assume the eternal nature of numbers, and made powerless by that shame.


Jonathan Simon, a sociologist who has studied the use of statistics in law enforcement, made a similar observation in a 1988 paper for Law and Society Review. “Over the last century there has been a significant growth within our society of practices that distribute costs and benefits to individuals based on statistical knowledge about the population,” he wrote. And these practices “are successful largely because they allow power to be exercised more effectively and at lower political cost.”

The market price of maintaining or slightly modifying the current system is indeed quite low; the health-care lobby—which is to say, all of the people who benefit from the current system—gave just under $150 million to Congress and the presidential candidates last year. That is a terrific bargain for them, but it does not explain why the rest of us are willing to sell our health so cheaply.

In the United States today, we can use our belief in numbers, which borders on the religious, to rationalize any amount of inequity. We can tell ourselves that we can’t have a system that guarantees health care to every American because such a system would be “inefficient.” We can tell ourselves that we must accept a world in which children suffering from post-traumatic stress don’t get any help because the numbers don’t support it. We can tell ourselves that we must trust our health to insurance companies, that markets are wiser than doctors, that we can afford every technology. We tell ourselves stories and we rationalize our prejudices, and, unless we are willing to more specifically address the “social determinants of health,” as Dr. Tsou said, we will continue to drift toward a change far more substantive than anything currently under consideration in Congress, a change suited to the few who care to exert their will. It may be a revolutionary system of corporate medical control, or a catastrophic financial collapse resulting from hubristic overtreatment, or a medical crisis stemming from some seemingly minor flaw in the heuristics of the integrated delivery network, or just a further increase in the massive inequities that already disgrace our current system. Whatever that change is, though, it will in the end be defined by the passivity of a people that has sacrificed its own, democratic power of large numbers on the altar of strange and unstated beliefs.

The Government Sector

On the third morning of the Democratic National Convention, a few hundred delegates and a dozen reporters gathered around cloth-draped dining tables in a conference room at the Denver Performing Arts Center to listen to the leaders of the Democratic Party describe how they would fix the health-care system. Andy Stern, the stubby, exuberant head of the Service Employees International Union, outlined the aim of the conclave with a convention-style chant.

Stern: What do we want?

Delegates: Health care!

Stern: First hundred days?

Delegates: Yes!

Stern: First hundred days?

Delegates: Yes!

Stern: In the first hundred days of the Obama Administration we’re gonna solve this health-care problem!

The speakers were notably specific, and they demonstrated genuine passion and intelligence. Michigan Congressman John Dingell, the longest-serving member of the House, made his way to the podium on crutches and explained that health care reform was needed because “a man ought not die like a dog in a ditch.” He added that a man has a right to an attorney when he commits a crime but no right to a doctor when he gets cancer. His anger—and his emphasis on “a man” as the relevant object of social consideration—were unusual. And yet, he concluded, “This is no longer a matter of social concern, or of humanity, but of economic salvation,” and thereby seemed, at least to my mind, to have upended the order of priority that once had informed the conscience of all thinking people.

The room grew more and more crowded. Every speaker was receiving a standing ovation. And a theme was developing. The Democratic governors and congressmen, the labor leaders and community activists, all of the public figures on the very farthest outer-left fringes of the respectable health-care debate in the United States, knew this one thing: the U.S. was never going to get health-care reform until the big money decided that heath care reform was in its interest. And the good news—the amazing news—was they have decided it is in their interest.

Up next was Kathleen Sebelius, the governor of Kansas. Until just a few days before the convention, Sebelius had been a leading contender to be vice president of the United States. She was tall and leathery and practiced. Her father had been governor, too, and she had made her own start in electoral politics as the state insurance commissioner. In Kansas, she said, the insurance regulator could take contributions from the insurance company he or she regulated, and she had long seen this as a preposterous conflict of interest. She had tried and failed to change the law, so she ran for insurance commissioner herself, on a platform of refusing contributions from insurance companies. She won, and within a few years was governor herself. She said that health-care reform was on the way, to believe it, that this was real. “We are finally at a tipping point in this country,” she said, “when we have a lot of voices in private industry calling for change!”

Then Ed Rendell, the Pennsylvania governor, with his bright orange tie and waxy block of a head, said, “If we control costs, the rest is easy.” Then Daschle, adding that “we’ve got a huge, huge cost problem.” And finally Hillary Clinton, whose husband asked America in 1993 to ask itself “whether the cost of staying on this same course isn’t greater than the cost of change.” Another standing ovation, longer and louder than all the rest. We cannot wait, she said. We have a plan—this current plan, this plan that represents the very best we can hope for—and it “is not only the moral approach, it is the economically sensible approach.” The crowd applauded again and again.

“This will be the kind of transformational change,” Clinton finally promised, “that comes once in a generation.” ■

No comments:

Post a Comment