FOUR

Health Care

The United States spends over a trillion dollars a year more than it should by peer standards on health care, and our health outcomes are no better for it. It is far easier to identify where we spend more money than peer countries do than it is to prevent the overspending. A full measure of savings can be gained only by moving away from competition among providers and toward what is generally called a single-payer system—although the single-overseer aspect of such a system is more important than the unification of the payment mechanism.

 

WHAT COUNTRY SPENDS the highest percentage of its GNP on health care in the developed world? I would be surprised if you don’t already know the answer. It’s our country, of course, and by a wide margin. It is a safe wager, in fact, that anyone who has shown the interest to read this far already knows the central story about our American system of health care. The depressing truth is that we spend substantially more than other prosperous nations on health care, yet our health outcomes are less favorable than most. The supersize spending comes right from your pocketbooks, through a combination of out-of-pocket costs, health insurance premiums, and taxes. The excess over international peer norms is, by any standard, thousands of dollars per year per person, and it is money you should think hard about whether you really want to be spending. Here is a refresher on the facts, followed by a discussion of why it doesn’t have to be this way.

A report from the Institute of Medicine sums up the outcome data this way: “When compared with the average of peer countries, Americans as a group fare worse in at least nine health areas: infant mortality and low birth weight; injuries and homicides; adolescent pregnancy and sexually transmitted infections; HIV and AIDS; drug-related deaths; obesity and diabetes; heart disease; chronic lung disease; [and] disability.”¹ The World Health Organization makes the same point with different metrics: average life expectancy at birth, infant and maternal mortality, healthy life expectancy at birth, and life expectancy at age sixty. In every category the United States ranks less favorably than the other wealthiest nations, without regard for continent.² The Commonwealth Fund has recently ranked health-care systems in eleven leading economies according to both quality and access. It ranks the United States eleventh.³

Yet another useful comparison is with our single most similar peer and using the most salient statistic. At birth, the life expectancy of a Canadian male is eighty years, four years more than that of a U.S. male. Female life expectancy is several years longer in both countries, and the gap between the two nations is about the same. For those who prefer to look a little more subjectively at “healthy life expectancy,” the picture is similar. The American newborn, without regard to gender, can expect sixty-nine years of healthy life; the equivalent Canadian, seventy-two.⁴ So, whatever the costs may be, we should not be smug about our health system. Despite the familiar boastful rhetoric to the contrary, our health care would certainly not appear from the outcome data to be “the best in the world.”

“Hey, wait just a minute,” someone may say. “The relative deficiency is not in our health care but in some differential genetic factors, demographics, or lifestyle choices.” The genetics are still largely unknown, but it is certainly not obvious that our genes are on the whole worse than those of the citizens of all those other countries. The demographic facts are more accessible. One fallacious argument I hear, based on demographics, is that we must have an older population than the others. If we did, we might be spending more than our peers and seeing higher death rates, but our lifetime results shouldn’t be worse. And, more important in rebuttal, we aren’t older than most of our wealthy peers. Our country’s median age is about thirty-seven, ranking us somewhere in the second quartile of all nations. Japan, at nearly forty-five, has a much grayer population and sits near the top with respect to outcomes. Canada, too, has an older population, and so do almost all of the European countries.⁵

Some of the difference may be attributable to higher levels of poverty in America, although the bottom decile incomes in our peer nations are not so different from our own lowest deciles. Another element could be the now-evaporating pool of uninsured citizens, who may have sought health care more reluctantly during the outcome sample periods than they would have in nations with universal coverage. This almost certainly biased our score on the Commonwealth Fund rankings. On the other hand, several studies suggest that even Americans who enjoy all the demographic advantages—white, insured, and college educated—have less favorable outcomes than similar individuals in other countries.⁶ Maybe we overeat more often, or smoke and drink more, or exercise less than the others, but that, too, appears to be less than obvious in the available data. If you travel, you have probably witnessed high rates of smoking and drinking elsewhere, and gym rats seem to predominate here in the States. I don’t wish to claim certainty about how much explanatory power some genetic, demographic, or lifestyle difference may have in shaping the results. It can be safely concluded, though, that if we have some adverse distinctions, our health-care system fails to compensate for them.

Now turn to the costs. Americans today spend about $2.9 trillion on health care annually, which is 18 percent of our gross domestic product.⁷ No other wealthy nation spends more than about 11 percent of GDP.⁸ The European average is less than 10 percent.⁹ To make matters worse, the costs here have been rising faster than inflation for many years. In the years from 1990 to 2008, it rose at an intimidating average annual rate of over 7 percent.¹⁰ President Obama took notice of this in 2009 when he said, “By a wide margin, the biggest threat to our nation’s balance sheet is the skyrocketing cost of health care. It’s not even close.”¹¹ The average cost to the system for one night in a hospital bed in Western Europe is less than $1,000. That bed you’d prefer to avoid costs over $4,200 a night here in the United States.¹²

If this isn’t story enough, a study by Deloitte Consulting suggests that we are underestimating the costs by the accepted measurements. The usual measurements exclude what Americans so generously spend on all manner of alternative medicines and practitioners. They similarly exclude weight loss, diet, and fitness programs taken on for health improvement; a good fraction of ambulatory services; nutritional supplements; health periodicals; and, most of all, supervisory care for the infirm and the elderly performed by other than for-pay professionals. The Deloitte team believes that the hidden costs may amount to almost another 20 cents for every dollar reported in the government’s National Health Expenditure Accounts.¹³ While the numbers that follow ignore the Deloitte findings, thus allowing the use of conveniently available data, I have no reason to doubt Deloitte’s estimates.

These nearly incredible costs, inexplicably, seem to be taken for granted now by many health-care policy analysts—as though they have always been a feature of our “best in the world” system. This is simply not the case. Take a look at the chart in Figure 4.1, which looks at the period from 1960 to 1986.¹⁴

 

Health Care as a Percentage of GNP

Figure 4.1

 

As recently as 1985, the United States was spending only 10 percent of GNP on health care, about the international standard today. In 1950, we were spending only 4 percent of GNP. You might want to look at the current spending disparity this way: In our economy, with a $16 trillion gross national product, the aggregate cost of health care is almost $3 trillion, and the extra eight points we spend that others don’t spend runs to just about $1.3 trillion. And that’s an annual number. Imagine what impact this much money could have if the overspending ceased and the difference became available for other private or public purposes. As a bonus, and not an insubstantial one, American manufactured products would be better able to compete in world markets. Not that I recommend it, but every couple of years of saving at this rate could cover a whole new war. Estimates are that the war in Iraq and the war in Afghanistan have together cost us about $6 trillion from inception.¹⁵ Or, better than that, we could restore the country’s infrastructure and rebuild the schools and bridges. Or, since nearly half of health-care costs leave your pocketbook in the form of taxes, we could have the mother of all tax cuts.¹⁶

A trillion is an incomprehensibly large number. One commonly used device to give it some intuitive meaning is the observation that a trillion dollars laid end to end would reach all the way to the sun and halfway back. A friend suggested another metric. Based on recently published New York City real property market value assessments, a trillion-dollar investor (if such existed) could purchase, without benefit of a mortgage, every single nonexempt building in the Big Apple—and still have more than $10 billion left over for redecorating them to suit.¹⁷ And that’s the amount I am suggesting we shouldn’t have to spend. And, I repeat, it’s an annual amount.

A direct quantification of total health spending, rather than just the excess amount, may also help in absorbing the reality at play here. Ultimately, whether through taxes or premiums, you and I are collectively paying the entirety of the health-care bill. On a per-capita basis, the total cost of American health care is close to $9,000 annually, or almost $23,000 for the average American family.¹⁸ You can multiply that $9,000 number by your own family size for a rough approximation of what you are spending every year. The next-highest number for per-capita spending I could find is Norway’s, at $6,000.¹⁹ And Norway, by the way, has the highest per-capita income of any other substantial country in the world, almost half again our own—so the pain there is doubly reduced.²⁰ Japan’s per-capita cost is about half as much as ours.²¹ The chart in Figure 4.2 sums up the situation visually.²²

 

Total Health Expenditure as a Share of GDP, 2010 (or Nearest Year)

Figure 4.2

 

We are the only true outlier in the upward direction. We’re number 1—and that’s not good. And what you spend each year on health care, directly and indirectly, is likely more than you spend on anything else. The average annual expenditure on housing is about $18,000, but that cost is per family unit, so it is less than the health cost total. Transportation spending is about $9,000 per year, but again this is per family. Food costs are almost $7,000 per family unit.²³ Nothing else is even in the ballpark of health care. These are, via government and private expenditures, your big outbound dollars.

 

---

 

There is little mystery to the sources of the costs. Five categories of expense explain the total fairly well. The first place many analysts look for excess is in doctors’ and nurses’ incomes.

 

The essential drivers of high cost in our system are well known. The apple of outsize spending can be sliced any number of ways, so various scholarly lists of where our costs exceed the other advanced nations’ differ. My list looks, in no particular order, like this: medical sector pay, technology and testing, prescriptions, administration and marketing, and end-of-life care.

Medical sector compensation is a touchy topic. I know hardly any doctor or nurse who doesn’t feel underpaid. As one who deeply admires the dedication and sacrifices required in this grittier profession than my own, and where most all are paid less than a top business executive, I don’t enjoy arguing with medical professionals. The better comparison for this analysis, though, is not with business entrepreneurs, but with doctors in other nations. The facts are easier to ascertain than to interpret. Our doctors are, without a doubt, paid more than their counterparts in other nations, but the comparison is a little short of apples to apples. The most commonly cited distinction is that a higher percentage of doctors in the United States are specialists. This is certainly true, and it is just as indisputable that specialists give up more of their youthful years in training than general practitioners—a reasonable justification for higher pay. Were I a specialist, I would expect a reward for my hard-earned skills, and I might be edgy if anyone challenged my right to pretty darned good pay. This book, though, is not looking to blame anyone for high incomes but rather to raise questions about the system. So, it is fair to ask what our society gains from paying doctors, including specialists and generalists taken together, more than other countries do.

The mean income for a primary care physician in the United States is about $225,000 a year. The average specialist earns more like $400,000, which is the highest average pay in any of the major American professional occupations—which mine (business) is not considered to be.²⁴ The average physician in Norway, Sweden, Portugal, or Italy brings home less than $100,000 in equivalent purchasing power.²⁵ In the United Kingdom, Germany, France, and Canada, physicians’ compensation averages something closer to $150,000.²⁶ Our general practitioners earn more than the other countries’ generalists, and our specialists earn more than other countries’ specialists. The mix in the United States, moreover, favors specialists about three to two over generalists, which raises the combined average; elsewhere the general practitioners tend to outnumber the specialists.²⁷ Specialists are, understandably, harder to see than family doctors. Maybe that helps explain why the average member country in the OECD has about one and one-half times as many doctors’ visits per capita as the United States.²⁸ Society might be better off with a different mix of physician types, favoring more general practitioners, but I can’t say that with any confidence. Certainly, the incentives for an upcoming medical student are all the other way.

Incentives, in general, drive results. When incentives influence high-minded career choices so as to produce a less-than-optimal outcome for society, the appropriate public response would be a dispassionate evaluation of policies that might shift those choices. Where the system’s incentives and human nature interact to inspire skullduggery, though, indignation and demand for better enforcement are more suitable reactions. In an influential article in The New Yorker, Atul Gawande, a surgeon and Harvard Medical School professor, offered an example of how some doctors are gaining, not by investing in expertise, but by gaming the system.²⁹ The focus of his piece was McAllen, Texas, a poor town on the Mexican border. The average Medicare expenditure per patient there in 2006 was $15,000, twice the national average. It was also twice as high as in El Paso, another border town with similar demographics. And the expense differentials were of fairly recent vintage, almost all of them arising during the preceding fifteen years.

According to Gawande:

 

Between 2001 and 2005, critically ill Medicare patients received almost fifty percent more specialist visits in McAllen than in El Paso, and were two-thirds more likely to see ten or more specialists in a six-month period. In 2005 and 2006, patients in McAllen received twenty percent more abdominal ultrasounds, thirty percent more bone-density studies, sixty percent more stress tests with echocardiography, two hundred percent more nerve-conduction studies to diagnose carpal-tunnel syndrome, and five hundred and fifty percent more urine-flow studies to diagnose prostate troubles. They received one-fifth to two-thirds more gallbladder operations, knee replacements, breast biopsies, and bladder scopes. They also received two to three times as many pacemakers, implantable defibrillators, cardiac-bypass operations, carotid endarterectomies, and coronary-artery stents. And Medicare paid for five times as many home-nurse visits. The primary cause of McAllen’s extreme costs was, very simply, the across-the-board overuse of medicine.

 

The overuse, of course, led to higher take-homes for the participating doctors. What happened in McAllen was surely extreme and atypical, and in a recent update of the article Gawande indicates that much of the excess has now been curbed.³⁰ Still, we should assume that less visible bits of McAllen behavior arise nearly everywhere. That is just human nature. Doctors, like the rest of us, come in all stripes of morality and talent for self-justification. Where the incentive structure allows enrichment from a tactic like gross overutilization of medical services, a few will abuse that structure wholesale, and many more will take just a little advantage of it. In many cases, the harmful behavior need not even be undertaken consciously. The mind has a marvelous ability to rationalize any course of action that serves maximization of personal rewards, especially when the merits of decisions are as legitimately hazy in ethical terms as choosing the right amount and course of treatment for a sick patient.

You can assume that mini-McAllen behaviors abound, even in a profession with more idealism and generally higher ethics than in the population as a whole. Doctors are relatively free to “upcode” Medicare and Medicaid patients, logging in at the margin the more serious of several possible diagnoses and thus raising their own possible billing ranges. They can use, wherever allowable, expensive out-of-network providers to assist and consult on procedures. They can choose to actively treat a condition another physician would watch and wait on. They can design a follow-up visit schedule liberally or conservatively. Almost every doctor I have talked to about this confirms that end runs around price controls are easy for those who want to employ them. And these end runs have equal opportunity availability to cheaters and sincere believers in activist treatment alike.

The market for health care is especially vulnerable to this pattern, because it is a classic example of a market with few of the prerequisites for competitive efficiency. Health-care consumers almost never have sufficient price and quality information, or the skill and freedom, to select their vendors in a manner typical of simpler markets. We almost never know the details of what we ought to be seeking when we see a doctor; we don’t know much about the differing choices available through other suppliers of medical advice and treatment; nor do we know much about outcome data or probabilistic prognoses associated with the choices we make. On top of that, our emotions cloud everything when health is at stake. Few of us are likely to bargain fiercely, or even rationally, with someone on whom our lives may depend. We trust and depend upon our doctors and local hospitals in a manner very distinct from the way we feel about car dealers, or even automobile insurers like my own company. This is as it must be, and it is unlikely ever to change. And, on top of these factors, we are seldom paying out of our own pockets, as would be the case in most other markets for goods and services.

Those who wish to rein in the costs associated with seeing doctors sometimes suggest shifting much of the work to skilled nurses and physicians’ assistants. These professionals have had less expensive training than doctors, and their time is therefore charged at lower rates, and they can certainly perform a greater share of the tasks than they do today. I suspect this shift of effort is, in fact, a tide of the future. It seems inevitable that improved and easily portable medical records, along with better use of computerized diagnostic tools, will encourage a greater reliance on the nurses and paraprofessionals. How much that will actually affect aggregate costs isn’t worth hazarding a guess, but directionally an expansion of the medical service supply ought to help.

Another answer requires more profound change. The incentives to overtreat, whether through malice or compassion, are largely erased when physicians are paid on salary rather than in proportion to services rendered. Many already are, and the number is growing. Some years ago, an influential U.S. senator said bluntly, “Everyone knows that [salaried doctors] is the right model. The question is: How do you get from here to there?”³¹ The Cleveland Clinic, the Mayo Clinic, and Kaiser Permanente, no slouches in quality, all employ salaried doctors.³² So do a great many health maintenance organizations. If enough of the medical economy operated in this manner, I expect the income differentials between the United States and its peers would moderate—and there would certainly be a reduction in low-hanging fruit for the cheaters as well as the overly eager interventionists.

Before we beat up on the doctors too much for high pay, though, let me remind you that this is only one of the factors pushing our health-care costs to exceed international standards—and it cannot be the largest. So don’t look to slashing doctors’ pay as a way to recoup all of that excess money you are spending. There are roughly a million doctors in this country, including both holders of an MD degree and other recognized medical certifications.³³ Multiply this number by the most we could consider their pay to exceed foreign peer pay, and we could imagine a comparative overage in cost approaching $100 billion. This is without judgment as to whether the differences are justified by higher-quality professional skills or rational consumer choice. The total represents less than a point of GNP, and closer to half a point. Remember that the United States is spending a dozen times that much in excess of its peers. So, there is work to be done, and possible savings for you from better incentives, but there is no magic savings bullet here.

I haven’t mentioned nurses so far. There is a reason for that. You might ask as a starter, as I did, whether nurses’ pay might be a major contributor to the high costs of U.S. health care. This does not seem to be the case. One OECD study shows, in fact, that nurses earn only slightly more in the United States than their overseas counterparts on a purchasing power parity basis. When compared to pay in other professions requiring postgraduate work, a nurse’s relative pay in the United States is about average for the countries sampled by OECD.³⁴ Another source ranks our nurses’ pay lower than that by comparison and, incidentally, asserts that Russian nurses are paid only about one-eighth what nurses in the United States receive.³⁵ I trust, for the sake of those Russian caregivers and their patients, that this last piece of data was in error.

 

---

 

Overutilization of testing is well known as a source of excessive cost. One driver of the excess is misdirected competition among hospitals that causes each of them to purchase and offer the most sophisticated technology regardless of normal supply-and-demand considerations. Another part of the problem is defensive medicine in response to threat of litigation.

 

Technology can save lives, and properly used, it can doubtless save money in just about every sector of the economy. There is virtually no limit, though, to what it can cost us all if its application is unrestrained. This is particularly so with respect to medical testing technology. Unlike doctors’ incomes, this is a source of expense where expert opinion is nearly unanimous about savings to be found by eliminating excesses. I was unable, to my surprise, to find a reliably sourced estimate of the annual total of spending by Americans on medical tests, but an educated guess would put the aggregate cost at something like half a trillion dollars a year. One trade journal reported in 2009 that diagnostics account for 10 percent of all health-care costs, but other sources suggest a much higher number.³⁶ A loose estimate used by writers in the field is that one-third of the testing and diagnostic cost total is wasted.³⁷ What we can say for sure is that America’s health-care system incentivizes substantial unnecessary testing costs.

The American Board of Internal Medicine Foundation has compiled reports of twenty-five medical specialty societies and determined that the routine use of 130 different medical screenings, tests, and treatments should be scaled back. The societies jointly released lists of these tests in a campaign they called “Choosing Wisely,” which urges patients to question certain testing recommendations.³⁸ There are now about eight thousand distinct diagnoses.³⁹ If we tested for all of those that are nonzero possibilities in a patient, we would break the health-care bank. The newest budget buster is genetic testing, which friends at Cold Spring Harbor Laboratory tell me is now just on the verge of yielding clinically useful results. A medical journal estimates that molecular and genetic tests will soon, though, represent as much as one-third of all diagnostic testing costs.⁴⁰

A personal story recently added to my education on misaligned incentives in the medical testing arena. Over the course of my early sixties, I noticed a mild impairment of hearing in the upper frequency ranges. So, I thought I ought to track it and asked for a routine hearing test, the kind where you raise your finger whenever you hear a beep. The test confirmed what I knew and added useful quantification. Then, about a week later, I received a letter from the hospital that had administered the test, suggesting an MRI. The logic was that my two ears had less than identical results, and thus a scan for a rare type of inner-ear tumor would be wise. Most people, presumably, would have gone for the test promptly, but with excesses in health spending on my mind, I decided to do some research.

A key part of my academic training involved the use of a tool called Bayesian statistics, to which I have come to be devoted. I won’t explain the statistical science here; it suffices to say that this is an ideal technique for examining the need for that MRI. The statistical technique follows intuitive logic. The rarer a tumor is in the whole population, the lower are the odds that it will be found in any individual, regardless of any test indication. And a test statistic is only as indicative as the deviation it shows from a normal result. My test asymmetry was hardly outside the normal range at all, and the potentially suspected tumor is extraordinarily uncommon (and not particularly dangerous even when it occurs). The Bayesian probability calculation of the odds an MRI would find a tumor yielded 0.00003, likely similar to the odds of my being killed jaywalking on the way to the test—which is the way my wife predicts I will eventually be taken from her. I declined the test.

The hearing loss story provides a number of lesson springboards. One takeaway is that a patient can be easily persuaded to accept a recommendation for an expensive test. Anyone less of a doubter by nature, or any less steeped in statistical training, would have gone for that MRI. Published estimates differ widely, but sources peg the cost of an average MRI in the United States in a range from $1,000 to $2,500, perhaps excluding any accompanying doctors’ fees.⁴¹ In my case, you—or at least those of you who share an insurer with me—would have helped pay for this. It is hard to blame people for accepting recommendations from doctors, and it is easy to see the systemic bias on the part of the patient toward overemployment of tests rather than underemployment. The bigger lesson, though, relates to the hospital’s incentive to recommend the test. The hospital in question had already paid for its MRI machinery and committed to all of the attendant support costs. The only way it can get its money back is to use the machine, preferably to near capacity. The hospital has no incentive at all to go light on usage. This is a fine hospital, by the way, and Boston should be glad to have it available. But multiply this case by millions, and the country has an overspending problem that you have to finance. It should be obvious that hospital competition in American health care seldom looks anything like the wonderful Adam Smith version of competition that disciplines costs so effectively. Here, it too often means acquiring the most sophisticated technology on the block and, when in doubt, recommending its use for patients—who are generally without the expertise, courage, or financial motivation to question expert advice.

The United States is the undisputed leader in this peculiar race to run the most tests. It is likely at or near the top for usage of all 130 tests and screenings identified by the Board of Internal Medicine Foundation, but MRIs are the ones for which the best data is easily available. In a 2014 study, the United States was shown to provide 105 MRIs per thousand residents per year.⁴² Spain’s number was 65. France, with more typical first-world statistics, provided 82. Canada recorded only 54. If the United States really runs MRIs at the rate of 30 million per year, and they cost about $2,500 apiece, the cost to you is running something like $75 billion annually.⁴³ And that’s just one of the 130 potentially overused tests. I probably get a PSA test (for prostate cancer) too frequently, an admittedly emotional reaction to having been a pallbearer for a close friend after his untimely passing from that disease. I can’t hold others to a higher standard. As Scientific American reports one tactful expert as having stated, though: “A lot of these screening tests are fishing expeditions. They’re very low-yield.”⁴⁴

An extreme and novel version of the case against excessive testing was put to a small group of us at Cold Spring Harbor by a Dartmouth Medical School professor. He argued not only that testing is costly, but that it is often more harmful than helpful to patients.⁴⁵ Every medical test is calibrated to have a characteristic balance of false negatives to false positives. The false negatives, where an illness is present but missed by the test, are considered far and away the greater threat to the patient—and to the doctor or hospital. A failure to diagnose a serious condition can lead to a health crisis for the patient, and a lawsuit for the medical professionals. False positives, even though they can lead to unnecessary anxiety or potentially dangerous overtreatment, are considered by the designers and providers of tests to be a much lesser threat. The result is that tests are calibrated to generate more false positives than most people understand.

As testing abounds and false negatives are minimized, medical professionals tend to define illnesses by lower thresholds over time. What was once considered normal blood pressure, for example, is now considered treatably high. Diastolic pressure below 120 used to be recorded as normal; now it’s 80. The systolic pressure threshold was 160; now it’s 120, according to the American Heart Association.⁴⁶ As a consequence, thirteen million more Americans undergo expensive treatment for hypertension, without clear evidence of appreciable benefit in the marginal situations. Cholesterol and diabetes thresholds have been reduced, again with the result that treatment intensity and frequency have been increased but without strong evidence of patient benefit. Growths once ignored or unseen are now classed as cancerous or precancerous. Since many suspicious cases would never have developed into full-fledged illnesses, the medical profession considers that it caught and cured a large number of cancer cases when, in reality, patients were not actually going to suffer from them. The triumphant cure and survival rate data you now read from time to time necessarily reflects an inseparable mixture of undeniably genuine progress and remedies for problems that never were. But it is used to support more and more testing. Even if a net benefit is derived from the tests, it is hard to see where any limits lie.

One defense some doctors offer is that overtesting is just a natural protective reaction to an overly litigious legal system. This point deserves your sympathy. An excess of tort actions will predictably draw an expensive defensive response. When I was the Massachusetts Insurance Commissioner, and one of the state’s primary health-care regulators, I questioned the costs of defensive medicine and a tort system in which less than half of the money—after costs of medical and legal defense, plaintiff attorney charges and contingent fees, and insurance—ever got to a patient. I proposed that the state legislature consider a change in the definition of actionable negligence. Negligence was too often found simply because of an unfortunate or tragic outcome that elicited jury sympathy for injured victims or their families. My proposal was to encourage an expansion of first-party insurance to cover most of the adverse outcomes without the necessity of a lawsuit, with malpractice awards limited to carefully defined cases of gross negligence. The sole response my proposal earned was a cartoon in the Boston Globe portraying a nattily attired trial lawyer reading my proposition and opining to a colleague, “I think I’m going to be sick.”

The change in malpractice laws would still make sense, and for all the same reasons. You are paying a great deal more than you should for defensive medicine and for all the extra testing and treatment it inspires. You are paying again for the malpractice insurance and other direct costs of the tort system. And you are getting, even if you are an unfortunate victim, very little of that cost back in exchange. One published study in the Journal of the American Medical Association estimates that 28 percent of all testing and treatment is ordered, at least in part, for defensive reasons, and that 13 percent of the costs arise to satisfy defensive motives.⁴⁷

 

---

 

Prescription drugs are more costly in the United States than elsewhere and more costly than they need to be. A portion of the excess is driven by irrational preferences of many patients, and some by the behavior of large pharmaceutical companies.

 

Almost 10 percent of the nation’s total health-care bill is for prescription drugs.⁴⁸ One needn’t doubt their importance to question both their quantities and their costs. It is easy to find examples of overuse, especially among the elderly. The dynamics of prescription often contradict the usual balance of influence between doctor and patient. Whereas testing and treatment decisions are dominated by the physician’s views, the doctor is all too often guided by the patient with respect to prescriptions—and the patient is seldom well informed. When a patient asks a physician to order major surgery the doctor thinks unnecessary or unwise, the physician’s medical ethics, decency, and good sense provide strong barriers to going forward. This is much less true with prescriptions, where the medical risks and costs are so much less that the patient’s demands are often given the benefit of the doubt. Some years ago, my wife’s aunt, at an advanced age, appeared increasingly confused. The family discovered she was taking at least a dozen prescription medicines every day. After family-inspired consultation with a caring doctor, she reduced the number to two, and she immediately improved. I don’t have all the facts, but I knew this aunt well enough to wager she had insisted on all of those drugs, probably using several physicians so that none had a full picture.

The price of drugs is also an issue. As is the case for other medical services, the prescription market ill fits the competitive model due to information asymmetries. On top of this, the patent laws are designed more to protect manufacturers and stimulate research spending than to limit system costs. What many view as excessive patent protection allows drugs to be priced where only third-party payers can afford them. The right balance for patent law is hard to discern, but there are easily found examples of the huge commercial value patent protection offers. The price of Lipitor, the cholesterol reducer I take, dropped by 85 percent when its patent protection expired. Before that, though, Lipitor’s economic lifetime sales had totaled over $140 billion, making it the world’s biggest-selling drug.⁴⁹ It’s no wonder that drug companies are fighting and lobbying hard for longer periods of protection and for the right to exclude outside parties from challenging patents before issue. The French nonprofit organization Médecins Sans Frontières (of which Doctors Without Borders is the affiliate in this country) has said that if the drug companies get their way in now-pending international negotiations, their proposals “would make it extremely difficult for generic competitors to enter the market, keeping prices unaffordably high, with devastating public health consequences.”⁵⁰

Eleven of the twelve cancer drugs approved by the Food and Drug Administration in 2012 were priced for annual costs of over $100,000.⁵¹ During the past decade, the price of the average cancer drug regimen has doubled.⁵² In this last case, a number of oncologists have vocally rebelled, but to no avail. One artifice that the pharmaceutical giants are accused of using is pulling a drug off the market as its patent protection nears expiration and replacing it with a new “improved” version, thus starting the clock all over again. In a victory for the opponents of this tactic, an Indian court recently ruled that a new version of the cancer drug Gleevec was not a true innovation but merely a patent extension tactic.⁵³ This helped clear the way for mass production of a generic version selling for about 4 percent of the cost of the original branded drug.⁵⁴

As with testing and hospital utilization, there is also occasional outright abuse. One of the uglier questions is whether some doctors’ prescriptions are influenced by financial rewards from manufacturers. It is fittingly illegal for a pharmaceutical company to pay a doctor a direct kickback, such as an outright commission on the doctor’s scripts for the company’s drugs. While it is possible this kind of blatant conflict may still be enabled through elaborate disguises in rare cases, this is unlikely to be a major systemic problem. Fees paid to doctors for publicly promoting their drugs, however, are legal. And the line between appropriate and conflicted promotional behavior is blurry. The nonprofit news outfit ProPublica alleges that crossing the line has resulted in settlements of hundreds of millions and sometimes billions of dollars in cases involving indirect kickbacks and improper marketing.⁵⁵ As with expensive testing, this is an arena in which the human mind—even the minds of people in the noble medical profession—can be wonderfully elastic when it comes to self-justification. I am not sure anyone can know, including the doctor involved, whether a physician favors a drug he or she is paid to promote as a result of the payment or based on independent, expert judgment. It is unknowable, even when all the objective facts are on the table, whether a doctor leaned toward prescribing a drug he or she promoted in a speech because of an honorarium or felt comfortable providing the endorsement because of a well-founded and preexisting confidence in the drug’s efficacy. The effects on the patient and the medical system can be identical, though, when honoraria and biased prescription behavior are linked subconsciously as when direct commissions are paid.

If a pharmaceutical company tracks how much a doctor prescribes of its products, and then—without a formal contract or fixed percentage arrangement—offers to the most profitable and loyal doctors light-duty consulting contracts, exotic travel, expensive meals, or speaking fees for delivering company-approved speeches, the effect is essentially the same as a commission. This is what ProPublica says is commonplace.⁵⁶ Without necessarily accepting that extreme estimate, one can certainly conclude that something is amiss when half a dozen large pharmaceutical companies—with names you know and trust—have entered into settlements with authorities for practices just like these. ProPublica’s database details approximately $4 billion in improper payments to doctors by seventeen pharmaceutical companies, mainly revealed in connection with litigation and enforcement actions.⁵⁷

American doctors write about four billion prescriptions a year, or thirteen for every citizen.⁵⁸ It is estimated that as many as one-fifth of all Americans are actively using five or more prescription drugs at any time.⁵⁹ This does not count, of course, nonprescription drugs, herbal medicines, and diet supplements. Drug companies earned profits of over three-fourths of a trillion dollars in the past decade.⁶⁰ They are aided not just by market forces but by actions (and inactions) of your Congress designed to help them out. There are many examples. The legislation establishing the Medicare drug plan specifically forbids the federal government to bargain on drug prices; elsewhere in the world government buyers regularly negotiate for the same discounts any large private buyer would receive.⁶¹ In most of our peer countries, when a patent has expired and cheaper generics can replace the expensive patented drug on the shelves, the original manufacturer is forbidden to pay generic suppliers to withhold product—but not here. The Federal Trade Commission has estimated that such “pay for delay” tactics cost consumers over $3 billion a year.⁶² Ours is one of a few countries that fail to limit branded drug advertising directly aimed at physicians when generics of like kind and quality are available. This may be part of why the pharmaceutical companies spend $24 billion on marketing directly to doctors, eight times what they spend on consumer advertising.⁶³

You as the ultimate funders of the system face the same two problems with respect to drugs as with physicians’ services: dysfunctional incentives and market failure due to lack of meaningful competition. Patients have a great deal less knowledge than fear of their medical conditions, and too many can be unreasonably demanding. Doctors, for understandable reasons, are more inclined to accommodate and please their patients with respect to drugs than procedures. And some doctors lean in their drug recommendations toward the wishes of pharmaceutical companies that reward them for promotion of their products. All of this can happen, moreover, without much sacrifice of individual patients’ interests or contradiction of conscience. Add nearly full insurance to the picture, and the market failure is nearly complete. The interest sacrificed is mainly that of the public collectively, who must eventually pay the costs.

 

---

 

Another source of excessive cost is end-of-life care. For reasons of moral complexity as well as hospital financial incentives, far more is spent on marginal postponement of death than makes societal sense. At a minimum, we should elicit and obey the wishes of the patients themselves.

 

The philosophical issues surrounding end-of-life care make patient protection arguments look simple. Start with these questions: How much would it be worth to you to live a year longer? How would that price be influenced by your quality of life or age at the time or by your balance sheet? How much would it be worth to you to see a loved one live a while longer? Does the answer differ if someone else is paying? Who should decide when to stop spending money on life extension and let nature take its course: The patient? The patient’s family? The most extreme member of the patient’s family? The patient’s doctor? A hospital staff? Or only God? Could you bring yourself to “pull the plug” even if you were rationally convinced that recovery of decent life quality was no longer possible? Suppose you were 90 percent convinced? Do patients you might label as irresponsible, such as continuing alcoholics with ruined livers, deserve the same interventions accorded to model citizens? How about an alcoholic whose drinking has precipitated the condition but has sworn never to touch a drop again?

All of us know stories that give us pause. There was a 60 Minutes feature about a woman with advanced heart and liver disease who spent the last two months of her life commuting between a nursing home and the local hospital. Although she had signed a living will that forbade extraordinary measures, more than two dozen specialists showed up at her bedside.⁶⁴ You presumably paid for that. A New York Times piece told of a patient who remained in the hospital for a year following a liver transplant. “He had never, ever been told that he would have to live with a ventilator and dialysis,” his doctor said. When the doctor inquired whether the patient might prefer to go home and receive hospice care, the surgeon on the case asked him to leave the hospital and not to come back.⁶⁵

I use a personal trainer to help me stay fit. Her mother was kept alive in a hospital setting for more than a year after she had lost her facility to recognize her family members or to communicate in any way. One of her adult daughters, likely for religious reasons, insisted on maximum care even when the other siblings were ready for a loving farewell. I don’t know the cost of her care, but we can estimate from hospital bed costs that even without intensive care supplements, it had to be in the $2 million range. Few serious students of health care doubt that extreme measures taken as a part of end-of-life care are significant contributors to the $3 trillion health-care bill we all share.

Readily accessible hard data establishes that a large portion of the health-care dollar in this country is consumed in end-of-life care. It is widely stated that about one-third of health-care costs for the elderly are attributable to care for patients near death.⁶⁶ The National Institutes of Health confirms, more specifically, that one-third of the Medicare portion of health-care expenditures is devoted to treatment and care in the last month of life.⁶⁷ I have seen higher estimates of total end-of-life costs, calculated on a more broadly inclusive basis. Even the NIH estimate, though, represents about one-tenth of total medical costs in the United States. Three-fourths of these costs result from ventilator use, resuscitation, and other extreme care techniques at the end of life.⁶⁸ That’s more than a point and a half of GNP, or about a quarter of a trillion dollars a year.

It seems obvious that this is too much, but the philosophical complexities, when combined with the economic incentives that influence hospitals, have rendered solutions out of reach. The chief executive of a major and well-respected hospital put it concisely and candidly to the New York Times: “If you come into this hospital, we’re not going to let you die.”⁶⁹ Hospitals vary all over the lot in average spending on this kind of care. The Mayo Clinic is near the low end, averaging about $50,000 per terminal patient, whereas some hospitals spend about twice that. The most expensive providers offer noticeably more intensive care unit days and specialist consultations than lower-cost hospitals.⁷⁰ Debates about how long someone should live in the ICU aside, it is not even clear that the high-cost, most activist hospitals actually prolong life. A report by Dartmouth College researchers determined that in its sample the highest-cost hospitals had just about the same end-of-life outcomes as the less interventionist.⁷¹

The most emphatic assertion I have to offer about end-of-life care is that it is time to start listening to the patients themselves. Dr. Atul Gawande, whom I quoted before, describes a better model than the national norm, and one already proven by use in this country. He cites the case of La Crosse, Wisconsin, where end-of-life hospital costs are in the range of half the national average, and the life expectancy exceeds the national mean.⁷² In 1991, apparently, local medical leaders began a campaign to solicit end-of-life decisions in advance from physicians and their patients. Patients admitted to a hospital, nursing home, or assisted-living facility began routinely to complete a form that asked them, among other questions: (a) Do you want to be resuscitated if your heart stops? (b) Do you want aggressive treatments such as intubation and mechanical ventilation? and (c) Do you want tube or intravenous feeding if you can’t eat on your own? By 1996, 85 percent of dying residents in La Crosse had such forms in place, and doctors almost always followed their instructions. The results have been genuinely impressive.

A related improvement in the current state of affairs, also cited by Gawande, would be achieved if more people simply died at home. Again, this should be a choice routinely offered to the patient. My own mother chose that course. She wanted to die with dignity in her own bed, in her own clothes, and surrounded only by the family who loved her. I doubt she would have lived more than a few days longer if we had applied extreme methods. Fearful that a hospital would take over the decision process, she refused ever to enter one—or even allow a doctor visit—as the end approached. Hospice provided her the dignified passing she selected for herself. Gawande even suggests that counter to conventional wisdom, patients might actually live longer in familiar and loving environments than in a hospital ICU bed.⁷³ He notes with approval that although in the last half of the twentieth century the percentage of Americans who died at home dropped from over 50 to 17, the trend has now reversed. In the past five years, the number of people choosing hospice or home settings has returned to almost 50 percent.⁷⁴ This is a cultural shift that can only bring good.

Though welcome change may be in the wind, at the time of this writing Medicare still does not reimburse for advance care planning relating to end of life.⁷⁵ Combine this with the low percentage of doctors who receive geriatric training in medical school, and the falling numbers of geriatric specialists (down by one-third since 1994), and it spells wasted opportunity. The venerable National Academy of Sciences’ Institute of Medicine has recently recommended that Medicare and private insurers reimburse for planning. The American Medical Association has already set up a billing code for an hour-long conversation between doctor and patient and the filling out of an advance care planning form.⁷⁶ I hope most of you would agree that this is not the ghoulish “death panel” so many are afraid of. The NAS’s Institute of Medicine has also called for more general restructuring of Medicare and Medicaid to eliminate “perverse financial incentives.”⁷⁷ President Abraham Lincoln set up the National Academy to advise the President and Congress on matters requiring more than a layman’s scientific view. This would be a good time to take the Academy’s advice. As the Institute’s chief executive put it, “Patients don’t die in the manner they prefer . . . The time is now for our nation to develop a modernized end-of-life care system.”⁷⁸

Remember that the human brain did not evolve for making complex, long-term decisions, and neither did the Congress. But, on end-of-life care, we and our elected leaders must eventually make some choices. Modern medical science cannot extend people’s lives indefinitely, but it can all too easily devise ways to spend without reasonable limit on attempts to do so at the margin. Stalling death is a lot easier than extending a fully conscious life of decent quality—and a lot less useful. I doubt most patients want that outcome, and it is not even clear that life is actually extended when extreme treatments are applied. Even if these techniques do have a marginal extension benefit, moreover, and even if patients and their families favor the most extreme forms of intervention, society simply can’t afford to pay for them as they expand indefinitely in cost and sophistication. Ask yourself whether it would really be a better world if everyone got to spend a few days or weeks in an ICU at the end of life—and you were accordingly poorer.

 

---

 

The easiest excess to identify is in the costs of administration. This source of waste alone would justify a single-payer system, but there is an even better reason for establishing a unified system. Competition among hospitals and other providers can never provide reasonable cost discipline, even side by side with a government program. The full unification of regulation is essential if our country is to resolve the discrepancy between costs and outcomes.

 

Recall that our country’s health-care costs are about eight points of GNP higher than the costs in peer nations. If you are tallying my estimates of what the suspected drivers of extravagance might add to the bill, you are aware that much of the gap still remains to be explained. I have saved the most obviously wasteful contributor to the gap for last: the manner in which we administer our health-care system. The United States has somehow found itself with few or none of the benefits we admire in market solutions—while paying all of the costs inherent in a corporate approach to the marketing and administration of health care. This is a bad deal.

A study conducted at the City University of New York looked at health administration costs in the economically most advanced countries, and its lead author wrote, “We were surprised by just how big the differences have grown. The United States is in another league than every other country.”⁷⁹ The study estimated that administrative costs accounted for more than 25 percent of U.S. hospital expenditures, a figure not remotely approached elsewhere.⁸⁰ The report also found that those countries operating under a single-payer health system had the lowest administrative spending levels.⁸¹ A separate study, published in the New England Journal of Medicine some time ago, pegged overall administration costs at between 19 and 24 percent of total spending on health care. An update in 2003 raised the estimate to 30 percent.⁸² The share of the total health-care labor force devoted to administration had grown by 50 percent, despite the introduction during that interval of numerous managed care systems and massive increases in the automation of information technology. Today, it appears that at least one in every four health-sector employees works only at an administrative or clerical desk.⁸³

The two largest components of what we can group together as administrative costs are marketing and record keeping, including patient billing and reimbursement management. Social Security, while not a medical program, shares many of the same tasks and operates with an overhead cost of less than 1 percent. It has substantially no advertising and relatively uniform record keeping. The Medicare program is more expensive to operate, but not much, spending something under 2 percent of its total budget on administration.⁸⁴ There is virtually no advertising in the Medicare budget, and although Medicare’s records are more complex than those of Social Security, the requirements across the spectrum of providers and patients are relatively uniform. Critics of Medicare like to say the true cost is more like 6 percent. The difference between the trustees’ measure of overhead and the critics’ measure arises from the addition in the latter of the administrative expenditures incurred by the insurance companies that participate in Medicare Parts C and D. Whether or not that is a fair inclusion actually matters little. Private insurer administrative costs, including marketing, are generally estimated at 17 percent to 20 percent of premiums. The recent Affordable Care Act carries a provision aimed at limiting private insurer spending on administration to 20 percent of premiums, and 15 percent in some group plans.⁸⁵ Even if this can be fully and effectively enforced, which I doubt, it would mainly keep matters from getting worse; it is hardly a cure for the high costs you already bear.

I have long argued in conversation on this topic for a single-payer approach, though not usually to much avail. While organizing thoughts for this book, I recognized some major omissions in my accustomed support case for a single payer. Opponents quickly saw these holes as powerful defenses against any argument for change, and not unfairly so. One flaw was that the phrase single payer doesn’t fully or properly describe the solution that makes the most sense. The other is that we already have what look like single payers in Medicare and Medicaid, and they haven’t effectively held down costs other than in administration. Both points are worthy of address.

Single payer is the phrase in common use today, but it is in fact a poor descriptor of what is needed. The preceding pages have covered a number of the sources of this country’s unusually high costs: provider incomes rightly or wrongly exceeding international levels; copious utilization of sophisticated diagnostic testing equipment; failure to constrain either utilization or prices of pharmaceutical products; spending on nonproductive administrative paperwork as well as abundant advertising by insurers, providers, and their suppliers; and uncontrolled entitlement expenditures on end-of-life care. Having a single agency act as the monopoly bill payer for the entire U.S. health-care system would help greatly in curtailing administrative paperwork and reducing some marketing costs, but that transformation alone—as wrenching as it would be—would do little to address the other problem areas.

What we really need is a combination of a single payer, a single negotiator, and a single regulator. This might be best called a unified, or single, health-care administrator. Indeed, creating a single-payer format would by itself generate a bounty of savings in paperwork. The economies arising from freeing providers and their staffs of multiple forms and compliance with diverse and occasionally irrational justification regimens represent the simplest of all health-care savings to visualize. If the single payer was also a single insurer—at least for basic care, responsible for collecting premiums and handling all aspects of claims reimbursement—the savings would be that much greater. Insurer operating expenses, including marketing and advertising, do little or nothing in this imperfect market to improve the quality of care. This, however, is about where the single-payer notion, narrowly defined, reaches the limits of its usefulness.

To realize any savings in the pharmaceuticals area, we also need a single negotiator. The country would be well advised to rid itself of the plethora of laws and procedures meant principally to protect drug companies—and then bargain for the best prices a massive customer can get. The consumers’ side of the market can never by its nature be a paradigm of effective competition. The suppliers’ side, though, would rise eagerly to the challenge in the right environment. If more innovation and research are needed—an argument to which we should all be open—direct incentives for entrepreneurial and nonprofit activities would be unquestionably more cost-effective than subsidizing overall profit levels that benefit mainly pharmaceutical industry shareholders. Here the model can be borrowed from virtually any one of our first-world peers with minimal danger of unintended adverse consequences. It is a certainty that a single purchaser could bargain for, and obtain, better drug prices than we pay today. And an increased use of generics would save a great deal more if we would just cease to protect powerful patent holders beyond all reason. The Commissioner of the FDA has suggested that even the curtailed use of generics under current rules has generated more than $1.2 trillion in savings over a recent ten-year period.⁸⁶ The benefits of applying available market power on behalf of the public are not limited to pharmaceutical purchasing. A single negotiator could be a potent public advocate and bargaining representative in negotiations with other types of health-care providers as well, including medical device manufacturers, physicians’ organizations, and hospitals.

Even with a unified purchaser and administrator, however, there would be gaps in the effort to moderate the fraction of our GNP spent on health care. Excessive utilization by the public of prescription medicines, wasteful and duplicative purchasing by hospitals of costly testing equipment, and, most of all, unbounded spending on end-of-life care all need to be addressed. Efficient administration and the use of available bargaining power alone are insufficient tools for these challenges. As out of fashion as the word regulator may be in some quarters, a degree of regulatory oversight is indispensable if the systemic savings sought are to be anything like the annual trillion dollars or more that lie within our reach. I am nowhere near wise enough to offer precise detail, or withstand cross-examination, on just how the regulation should be applied, other than to urge that it include support for the general savings propositions described earlier in this chapter. I can be sure of this, though: the hardest task will come as a preamble to any specific form of regulatory intervention. Without a major cultural shift on the part of the public and a profound appreciation for the benefits freeing up a trillion dollars annually could yield the American people, regulation will not have the prerequisite consent of the governed.

The other debate weakness in the accustomed case for a single payer is highlighted when an opponent asks why Medicare and Medicaid don’t already provide all the benefits attributed to a single-payer system. It is generally acknowledged that these programs control their own administrative and marketing costs much more effectively than their care outlays. We have all seen attention-grabbing reports, moreover, of Medicare and Medicaid fraud. Among the reasons for the limited cost containment success in these programs is that they were intentionally designed to lack some of the tools necessary for greater impact. They are true single payers within their spheres, but they are shackled as negotiators and all the more constrained as regulators.

A parallel handicap is that Medicare and Medicaid sit side by side with a private system where incentives for the care providers too often favor higher costs. The incentives for the private insurers in this complex web, while more complicated than those of care providers, are not a whole lot better. In self-insured group plans, medical costs are mainly charged back to employers while the private insurer takes an administration fee that rises proportionally with treatment costs. When the insurer’s revenues vary on a cost-plus basis with spending on care, there is little incentive for tough cost scrutiny. And insurers fear losing their corporate accounts if they are strict with excesses. Insurers in risk-bearing plans do have short-run incentives to be frugal, but they are at even greater risk for account loss when they try to tighten the reins. Their scale in the long term is determined, moreover, by total health-care market size, which in turn is an upward-sloping function of care utilization.

When the private portion of the market sets the standards for frequency of testing, drug usage and cost, and compensation for every variety of medical services, it is extremely hard for any accompanying single-payer programs to be more parsimonious. Since 1997, Medicare statutes have provided for a “sustainable growth rate” limitation on payouts, tying the expansion of some reimbursements to the growth rate of gross domestic product. Congress, under pressure, has modified these rules almost annually. When, in 2011, the Medicare Payment Advisory Commission of Medicare suggested cutting some physicians’ fees, enforcement was postponed until at least 2015 and has now been postponed again. The single payers, moreover, can do nothing to alleviate the need for providers to maintain multiple accounting and billing systems for their private or shared patients, so little or no advantage can be taken of available efficiencies in this category of cost. Administrative spending runs to roughly $650 per patient in the average hospital.⁸⁷ Doctors’ offices, too, have to maintain costly redundant billing and reporting systems, with staff to go with them. These tend to be fixed costs, reduced only slightly by the simultaneous presence of Medicare and Medicaid, where accounting is standardized. It simply isn’t fair to judge the existing single payers by the results of a split system in which more than half the costs are private.

Another problem with today’s Medicare and Medicaid is that, like the private system, they allow costs to be largely invisible to the patient. A cost doesn’t go away, of course, just because it is spread over large numbers of people in taxes and premiums. It is a better bet that when purchase of a service is removed from the direct consequences of price, both cost and utilization will expand. The think tank the Cato Institute has noted correctly that not only do patients overuse those resources that appear to be almost free, but producers of new equipment can sell their products into an unusually price-insensitive market.⁸⁸ Private insurance has a wide variety of copayment requirements, deductibles, and subscriber contribution requirements, but few are particularly effective. Medicare has an annual deductible for hospital stays, but no copay for the first sixty days of care. It has a gently means-tested annual deductible and a percentage copayment for doctor’s services. The Medicaid program allows individual states to set all three variables, but limiting federal statutes and the need status of the Medicaid clientele make this a tough arena in which to demand meaningful market incentives. When all is said and done, none of the incentives for careful consumption of health care are strong enough. Part of the problem is deep within our political ethos. Americans seem to resent accurate user charges when the good being purchased is seen as a necessity. Health care is the archetypical example.

Some observers understandably look at our first-world peers and insist that the favorable experience of other nations with full coverage shows that there is little need to hold patients responsible for any substantial portions of their charges. Their observations about the peer systems and their results are correct enough, but the conclusion may not be equally valid. The United States has too long enjoyed the habits of overutilization and overspending to permit their curtailment without the enforced discipline of contributory first-party payments. If someday doctors and patients have learned to care enough about aggregate health-care expenditures to maintain a sensible balance between costs and usage, it might be a celebratory event to restore at that time fully reimbursed health care for all—without substantial patient contributions. In the interim, we will need deductibles and copayments, and all the other forces we can muster, if we wish to bring about a meaningful change in national spending on health care.

Annual subscriber contributions have almost no beneficial effect on utilization behavior at the margin. Their usefulness is simply in spreading the financing cost of health care between out-of-pocket costs and insured costs. Deductibles that are applied on an annual basis have only a mild incentive impact in most cases, and none at the point of purchase after the deductible (patient-paid) amounts are exhausted in any year. It is principally copayments, owed by the patient every time a service is consumed, that can restore the beneficial effects of a free market. An economically powerful copayment schedule is about the only tool available to restore some market efficiency to health-care utilization, and a universal and steeply means-testing copayment schedule would be required to make it either effective or just. The obstacle here is the public, which unwisely seems to prefer to pay by the less visible mechanisms of taxes and insurance premiums—even if the true cost is higher that way. Until this preference changes, the government programs will be politically constrained to accept what failed markets deliver. And as long as our private and public systems operate side by side, we are condemned to overspend.

 

---

 

Demographics are currently unfavorable for health-care spending reductions regardless of systemic changes. In the short run there is no way to prevent further increases in the share of GNP devoted to health care. Other countries, though, have models more likely to moderate costs in the longer run.

 

If the prior pages have implied that better public policies can radically bring down the absolute costs of health care in any near-term time frame, I should clarify my view. Better use of incentives and administrative efficiencies would be a force in the right direction, and they can curb the expanding fraction of GNP devoted to health care in the long run, but the trajectory of health-care costs in the next decade or so will be upward no matter what we do. That is because demographics count, too. As the baby boomers age, the ratio of expensive health-care beneficiaries to healthy funders of private and public programs is sure to rise. So, the utilization of health services is bound to grow faster than the economy—and by no small margin. Adding to the upward cost pressure may be the impact of the Affordable Care Act, whose proponents made a stronger case on fairness grounds (and for which reason I was a supporter) than they did for its cost control features. Its illness prevention features are, simply put, unlikely to compensate for adding an additional subpopulation, likely of less-than-average good health, to an existing system in need of serious repair. Some experts defended the law as a first step; universality first, then efficiencies. I will feel a great deal more comfortable if that sequence turns out to be more than wishful thinking.

The Affordable Care Act does contain a few cost buffers, and recent numbers indicate a welcome slowdown in the pace of health-care spending. The good news is that the improvement is not just in the total but in the government programs as well.⁸⁹ This suggests that the savings are not just a recessionary effect, since recession should mainly impact the private side. The bad news is that much of the improvement is due to one-time effects that cannot be expected to continue as restraints. And, in any case, the trend still points upward. There are so many factors influencing health-care spending that it would be an error to draw any long-term conclusions from short-term results. The tide should always be distinguished from the waves, and the tide is still shifting toward cost increases. No policy response can prevent this for many years. I will be surprised if the percentage of our GNP consumed by health care does not soon approach 20 percent, even if the public learns to treat the issues more rationally than it does today. The opportunity before you does not, alas, include a quick reversal in the direction of costs. The threat is that without rational policy action, the share of GNP consumed by health care won’t stop at 20 percent and won’t improve as the demographic pull subsides.

Other countries have done a better job than we have. Some of the features of each are potentially applicable in our culture and some are not. Some are attractive models and some are not. I am not suggesting we copy any one of them wholesale. The summary that follows is mainly to show that there is no standard model. We must pick and choose from the available menu of regimes a set of methods consistent with our culture and history—and that will work for us. England has a national system funded by taxes.⁹⁰ Every citizen is covered, and other than a small copayment for dentistry, eyeglasses, and prescriptions, there are no bills. The means-tested market incentives I would like to see are largely missing. On the other hand, while most primary care doctors are private contractors reimbursed on fixed schedules, most specialists are salaried by the state. Many of the hospitals are state owned.

Switzerland has a hybrid universal coverage system, where private insurers are prohibited from making a profit on basic care but permitted to do so with respect to advance care.⁹¹ Taiwan, widely praised for its modest administrative costs, has a national system with copayments and universal coverage.⁹² The Japanese utilize health care even more than Americans—there are more doctor visits and more prescriptions—yet they spend half as much as we do per capita. Prices for all medical services are heavily regulated, and there is a 30 percent point-of-purchase copay for those who can afford it, varying downward by income and age.⁹³ Germany, the home of social insurance, has a system with diverse elements that can be confusing to an outsider. Universal coverage is offered, with means-based premiums and by private insurers—but they are nonprofits. The insurers bargain with doctors on prices, while the government helps impose price discipline through a global budgeting process. There is also a parallel for-profit private system available to a relative few.⁹⁴ Looking at all of these approaches simultaneously, about the only conclusion one can draw is that the health systems employed by our peers are all different in design, but all seem to share at least these features: universal coverage, effective cost restraints, and superior outcome statistics. It is hard to believe we don’t have something to learn from them.

The most common argument I hear against borrowing anything from the approaches to health care of other wealthy countries is that they have longer waiting periods for services than we do. I have read that in the United Kingdom, Australia, and Canada, the average wait for elective surgery is more than a month, and over a quarter of the nonemergency patients in one survey reported waiting over four months. All three of these countries say they will invest extra resources to address the waits.⁹⁵ My assumption is that there is validity to the comparison. Our system probably works faster in these situations. If it saved enough money, though, I would accept a longer waiting period for a hip or knee replacement and couldn’t possibly object to a wait for purely elective surgery. Delay is less tolerable for more serious or potentially progressive conditions, and I haven’t found reliable disaggregated data to see how the waits in other countries vary by type of surgery. Even if delays occur elsewhere where they ought not to, though, please recall that we do not enjoy better health outcomes. So the impact of any such waits can hardly be devastating. Outcomes are more important than velocity of treatment. And our system is costing us over a trillion dollars each year in costs the others don’t bear. The American public is not all that good at understanding or accepting tradeoffs that require giving up a luxurious accustomed entitlement, but sometimes tradeoffs simply need to be made.

 

---

 

The easy thesis about health care is that our woefully ineffective system of provider competition should be replaced with a universal single-payer, single-regulator approach. In the meanwhile, there are numerous valuable halfway measures that can be taken to improve efficacy in specific areas of the system, with a goal no less ambitious than world-class outcomes at costs similar to those of our peers.

 

The inescapable fact is that you and I pay far too much for the health care we receive, more than a trillion dollars a year too much. The care is good, but outcomes are less favorable than in peer nations with lower costs, and a cost excess with twelve zeros is simply too large to ignore. It should be a national goal to catch up with our peers in outcomes and, at the same time, cut what you pay every year by one-third. Some of the savings would appear in your taxes, some in your health insurance costs, and some in your out-of-pocket expenditures. It is there for the asking.

We Americans tend to be proud of our ways, and quick to defend them. But as Uwe Reinhardt, an economics professor at Princeton University, asked, “How can it be that ‘the best medical care in the world’ costs twice as much as the best medical care in the world?”⁹⁶ That’s the right question. There are multiple drivers of the cost overage. There is no easy thesis to offer with respect to doctors’ incomes. Fraud should be fought, of course, but don’t look there for massive savings. Only a few physicians are intentionally behaving dishonestly; most are just doing what the system calls for—and remain more altruistic than the rest of us. We might well be better off with a mix of general practitioners and specialists that looks more like the rest of the world’s mix, and we might want to let paraprofessionals do more of the work, but neither of these changes would make a serious dent in the overall spending pattern.

There is no easy response to burgeoning technology costs either, as long as hospitals are dysfunctionally competitive and our tort system ensures a surfeit of defensive medicine. Technology will continue to improve, and its costs have no natural upper limit. There will similarly be no ready path to reducing the huge expense for end-of-life care until America is ready to face the realities of the situation and cut back. We can make some real progress, though, by insisting that patients’ choices about their last days be stated in advance and binding on the medical profession. There is a somewhat clearer logical path to follow with respect to medications. Even if we continue to want too many drugs, especially in elder years, we can take steps to temper the costs. There is no reason for government programs not to use the normal powers of large purchasers to bring prescription medicine costs into line with other nations’ costs. Only the politics of campaign contributions need to be overcome.

Excessive spending on administration is the easiest target. There is no more powerful first step the country could take to cut unnecessary costs than to establish a single-payer system for the distribution of health insurance and the handling of its accounting. A unified system would begin paying dividends immediately by eliminating costs for advertising and the maintenance of multiple, incompatible record-keeping and billing systems. Even more important, it would gradually permit the building of useful efficiency and quality incentives into a health-care system now plagued with perverse incentives.

The money you are unnecessarily paying today for below-par health-care outcomes is coming from your wallets, but the victims of the misdirected spending include generations unborn. The economics of health care are such, given relative propensities to illness, that spending on health care always transfers wealth from the age cohort currently working to the elderly and leaves less to be spent on the young. Ask yourselves whether, if forced to choose, you would rather live in a society that tilts toward spending the next dollar on the young or the very old. If nothing changes in health care, the next dollar is already spoken for. If we stop spending thoughtlessly, though, we can treat our seniors at least as well as they would be treated elsewhere in the world and still save money. Not all expenditures and not all goods produce the same amount of economic benefit. Spending on the young and on infrastructure represents an investment that will pay later returns. Spending on too many tests, too many drugs, and too much intervention in the last days of life is just spending.

The institution of a single-payer health-care system, accompanied by the appropriate cost disciplines, presents an opportunity with few historical parallels in terms of national financial impact. The average American household can save itself thousands of dollars each year, our products can be more competitive in world markets, and our society can amply fund additional investments in its future—all without a sacrifice in health outcomes. Repairing the health-care system in this country is, measured in dollars, worth more to your family and our country than any other conceivable national policy change.

It’s not even close.