An American Sickness: How Healthcare Became Big Business and How You Can Take It Back

by Elisabeth Rosenthal

“It’s now so dysfunctional that I sometimes think the only solution is to blow the whole thing up. It’s not like any market on Earth,” says Glenn Melnick, a professor of health economics and finance at the University of Southern California.

United States spends nearly 20 percent of its gross domestic product on healthcare—more than twice the average of developed countries.

Financial incentives to order more and do more—to default to the most expensive treatment for whatever ails you—drive much of our healthcare. The central mantra of “innovation” in the past decade has been “patient-centered, evidence-based care.” But isn’t that the very essence of medicine? What other kind of medical care could there be?

EACH MARKET HAS certain rules that are determined by the conditions, incentives, and regulations under which it operates. Currently, we buy and sell medical encounters and accoutrements like commodities, but how do participants in the marketplace make purchasing choices? Prices are often unknowable and unpredictable; there’s little robust competition for our business; we have scant information on quality to guide our decisions; and very often we lack the power ourselves to even choose.

The earliest health insurance policies were designed primarily to compensate for income lost while workers were ill. Long absences were a big problem for companies that depended on manual labor, so they often hired doctors to tend to workers. In the 1890s, lumber companies in Tacoma, Washington, paid two enterprising doctors 50 cents a month to care for employees. It was perhaps one of the earliest predecessors to the type of employer-based insurance found in the United States today.

The archetype for today’s insurance plans was developed at Baylor University Medical Center in Dallas, Texas (now part of Baylor Scott & White Health, since it merged with another health system in 2013, forming a giant healthcare conglomerate), which was founded in 1903 in a fourteen-room mansion by the Baptist Church. A devout cattleman provided the initial $50,000 in funding to open what was then called the Texas Baptist Memorial Sanitarium, “a great humanitarian hospital.” By the 1920s, more and more Texans were coming for treatment. When Justin Ford Kimball, a lawyer who was Baylor’s vice president, found out that the hospital was carrying a huge number of unpaid bills, he offered the local teachers’ union a deal. For $6 a year, or 50 cents a month, teachers who subscribed were entitled to a twenty-one-day stay in the hospital, all costs included. But there was a deductible. The “insurance” took effect after a week and covered the full costs of hospitalization, $5 a day, which is about $105 in 2016 dollars.

Within a decade, the model spread across the country. Three million people had signed up by 1939 and the concept had been given a name: Blue Cross Plans. The goal was not to make money, but to protect patient savings and keep hospitals—and the charitable religious groups that funded them—afloat. Blue Cross Plans were then not-for-profit.

Transformative technologies rapidly spread across the developed world. Abbott Laboratories made and patented the first intravenous anesthetic, thiopental, in the 1930s. Massachusetts General Hospital started the first anesthesia department in the United States in 1936. The first intensive care unit (ICU) armed with ventilators opened during a polio epidemic in Copenhagen in the early 1940s. Five dollars a day and a twenty-one-day maximum stay were no longer enough. Insurance with a capital I was increasingly needed. A private industry selling direct to customers could have filled the need—as it has for auto and life insurance. But a quirk of history and some well-meaning policy helped etch in place employer-based health insurance in the United States. When the National War Labor Board froze salaries during and after World War II, companies facing severe labor shortages discovered that they could attract workers by offering health insurance instead. To encourage the trend, the federal government ruled that money paid for employees’ health benefits would not be taxed. This strategy was a win-win in the short term, but in the long term has had some very losing implications.

Blue Cross and its partner, Blue Shield, were more or less the only major insurers at the time and both stood ever ready to enroll new members. The former covered hospital care and the latter doctors’ visits.

They accepted only younger, healthier patients on whom they could make a profit. They charged different rates, depending on factors like age, as they had long done with life insurance. And they produced different types of policies, for different amounts of money, which provided different levels of protection. Aetna and Cigna were both offering major medical coverage by 1951. With aggressive marketing and closer ties to business than to healthcare, these for-profit plans slowly gained market share through the 1970s and 1980s. It was difficult for the Blues to compete. From a market perspective, the poor Blues still had to worry about their mission of “providing high quality, affordable health care for all.” By the 1990s, the Blues, which offered insurance in all fifty states, were hemorrhaging money, having been left to cover the sickest patients. In 1994, after state directors rebelled, the Blues’ board relented and allowed member plans to become for-profit insurers. Their primary motivation was not to charge patients more, but to gain access to the stock market to raise some quick cash to erase deficits. This was the final nail in the coffin of old-fashioned noble-minded health insurance. Many of the long-suffering Blue plans seized the business opportunity. Blue Cross and Blue Shield of California was particularly aggressive, gobbling up its fellow Blues in a dozen other states. Renamed WellPoint, it is the biggest of the for-profit companies descended from the original ...

In 1993, before the Blues went for-profit, insurers spent 95 cents out of every dollar of premiums on medical care, which is called their “medical loss ratio.” To increase profits, all insurers, regardless of their tax status, have been spending less on care in recent years and more on activities like marketing, lobbying, administration, and the paying out of dividends. The average medical loss ratio is now closer to 80 percent. Some of the Blues were spending far less than that a decade into the new century. The medical loss ratio at the Texas Blues, where the whole concept of health insurance started, was just 64.4 percent in 2010.

The framers of the Affordable Care Act tried to curb insurers’ profits and their executives’ salaries, which were some of the highest in the U.S. healthcare industry, by requiring them to spend 80 to 85 percent of every premium dollar on patient care. Insurers fought bitterly against this provision. Its inclusion in the ACA was hailed as a victory for consumers. But even that apparent “demand” was actually quite a generous gift when you consider that Medicare uses 98 percent of its funding for healthcare and only 2 percent for administration.

By the late 1980s, Providence had hired professional coders to translate doctors’ exams into medical bills. Physicians were given stock phrases to use to describe their exams and told what procedures to perform to ensure better revenue—instruction that became commonplace at many hospitals. The doctors began receiving statements each month that showed how much money their examinations brought in, relative to those of their colleagues. Relations between the administration and the doctors became increasingly testy. The physicians in Dr. McCullar’s group asked to see what the hospital had billed Medicare on their behalf. Some doctors were concerned about overcharging. Others wanted to make sure they were getting their fair share. Providence refused and threatened to fire those who insisted on disclosure.

By the late 1990s the hospital said it no longer wanted to pay a salary to doctors in the ER and clinics; instead, it would treat them as independent contractors. “That turned us into a business also,” Dr. McCullar said. “We negotiated contracts that stipulated what percent of revenues we deserved.” Providence’s marketers also required the doctors to attend what Dr. McCullar calls “charm school.” The seminars were run by a group of physician consultants who had started the Foundation for Medical Excellence in Portland. “I tried to get the administration to look harder at quality outcomes—like infection rates—but they didn’t, and that’s more expensive.” With canny marketing, Providence grew its business, aggressively advertising cardiac care and a state-of-the-art “stroke center.” A new building was adorned with marble columns and featured a fountain with jumping salmon; after more renovation, granite countertops were added and expensive art was displayed. “It became like the Providence Marriott,” Dr. McCullar said. “I was embarrassed to have poor patients come here.”

In 2013 it had revenues of $2.6 billion and about $2 billion in assets. Its CEO is paid about $3.5 million a year. Yet it still describes itself as “a not-for-profit Catholic health care ministry” continuing “a tradition of caring that the Sisters of Providence began more than 158 years ago.” It lists nuns from Providence Ministries as its “sponsors.” The senior vice president of the health system received a papal medal from John Paul II, the Cross Pro Ecclesia et Pontifice, for his service after he retired in 2001.

The hospital decided it would no longer pay these physicians a fixed salary; instead, they would be compensated in proportion to the relative value units (RVUs) of the care they dispensed. RVUs are a measure of productivity used to determine medical billing.

Generalists like Dr. W. are assigned RVUs primarily according to the complexity of their exams and treatment plans, which are coded on a scale of levels 1 to 5. A simple level 2 visit may yield $60; level 3, $120; level 4, $210; and so on. “What started to happen is lots of pinkeye was billed at a level 4,” he explained. “It may not sound like much—but it adds up.” There was financial incentive: colleagues who were coding expansively could make twice as much—over $300,000 instead of $170,000.

Because many prestigious medical centers have all their medical staff on a salary, they contend that their physicians and surgeons therefore have no incentive to upcode or perform unneeded tests and procedures. But many of the well-known health systems (including the Harvard-affiliated Partners HealthCare, the Henry Ford Health System, Duke Health, and Baylor Scott & White Health) tie those salaries to physicians’ RVUs or sometimes offer “productivity bonuses” based on them. A small number even deduct money from a doctor’s salary if his or her RVUs are too low. In...

Insurers, particularly Medicare, tried to prevent upcoding by spot audits of charts, but hospitals provided helpful assistance to doctors. “It’s like a candy store—all you have to do is check the right boxes,” one doctor told me. “We had software for dictation, and it would even say, ‘You need to check two more boxes for level four.’” A questionnaire filled out by a patient can be labeled as a “health needs assessment,” which is a billable item. Injecting a patient’s bum knee with steroids can be coded as “surgery” costing $1,200. If the...

Coding creep grew ever bolder—until it made no sense at all. If needle sticks and eye infections are coded as level 5, what was a crushed chest from a car accident or a heart attack? A Center for Public Integrity (CPI) investigative series in 2012 found a huge increase between 2001 and 2008 in Medicare billing for levels 4 and 5 visits among emergency room patients who were sent home, from a quarter to nearly half of all patients. Meanwhile, the proportion of level 2 visits decreased by about half, to just 15 percent. More than 500 of the 2,400 hospitals in the database billed the two most expensive codes for more than 60 percent of pati...

Facility fees were a logical outgrowth of a period of rapid scientific progress in medicine, which allowed many treatments to move to an outpatient setting. Improvements in anesthesia, pain medicine, minimally invasive surgery, and biopsy techniques meant that many procedures and operations could be safely performed without an overnight stay. New medicines to quell the severe nausea of chemotherapy meant patients could receive treatment at an infusion center. Because hospitals had traditionally charged a day rate for inpatients, it made some sense that insurers (including Medicare) had largely accepted their new practice of charging facility fees for major outpatient care as well by the turn of the century.

Dr. Ronald Anderson, a Pennsylvania rheumatologist, can inject a bursa with painkillers in his office for about $80. (Bursas are fluid-filled sacs that reduce friction around joints. When they become temporarily inflamed it is called bursitis.) But when one of his patients had the condition treated by an orthopedist at a surgicenter with ultrasound guidance, the bill was almost $5,000, most of which the insurance company paid.

As departments underwent serial financial review, hospitals did away with loss leaders and enhanced their most profitable offerings: orthopedics, cardiac care, a stroke center (revenue from expensive scans), and cancer care (revenue from infusions). They erected electronic billboards promoting the short wait times in the ER—a bizarre notion if, as they all aver, people shouldn’t use high-priced emergency rooms to get elective care.

Since it was a complicated new therapy with unpredictable finances that few patients would need, Medicare reimbursed generously. In short order proton beam therapy was being used on a far wider range of tumors than had ever been intended, despite little evidence that it was superior to cheaper options. Every hospital wanted one and new machines came with billing tips as well as elaborate calculations about how long it would take to recoup the investment. (See Rule 1: More treatment is always better. Default to the most expensive option.) The National Cancer Institute and the American Cancer Society tried unsuccessfully to tame the frenzy with science, to no avail, even though one study of thirty thousand patients with prostate cancer found that “proton beam therapy provided no long term benefit over traditional radiation therapy, despite far higher costs,” according to a report in the Wall Street Journal.

In deals brokered by consultants like Innovative Health Strategies (and backed by private equity investors), hospitals “sold” their patients for $40,000 to $70,000 per head to the big commercial players.

But some corners of medicine yielded little revenue no matter how hard consultants squeezed. Dr. Gene Dorio, a geriatrician, was practicing at a small California hospital that went bankrupt in the early 2000s. Business professionals were brought in to “turn it around.” In 2006 the hospital announced plans to close its transitional care unit, where elderly patients could gain some strength before discharge. The unit was popular and provided invaluable healthcare, as far as patients and doctors were concerned. Dr. Dorio, a member of the Medical Executive Committee, organized a protest by the hospital’s elderly patrons in the lobby, to no avail. There was no revenue stream to support it, and it closed in 2008.

The median cost to a hospital for each full-time resident in 2013 was $134,803. That includes a salary of between $50,000 and $80,000. Federal support translates into about $100,000 per resident per year. Researchers have calculated that the value of the work each resident performs annually is $232,726. Even without any subsidy having residents is a better than break-even deal.

Founded in the 1980s by a medical anthropologist and a statistician, Press Ganey bills itself as continuing “to lead the patient experience industry,” hired by an estimated 50 percent of all U.S. hospitals. In 2015 it filed with the Securities and Exchange Commission for an initial public stock offering. Customer satisfaction is important and predicts repeat business, but it does not necessarily indicate medical quality. Studies have determined that such surveys have only a “tenuous” link with patient outcomes. Physicians hate them. But Medicare pays hospitals a bonus for performing well on patient surveys. “So if a patient asks for a test and it won’t hurt, they’ll get it,” one doctor told me. “It’s good for Press Ganey scores. It takes more time and trouble to explain why they don’t need the X-ray.”

But like nearly all prestigious American medical centers, the UPMC is nonprofit, so it pays almost no U.S. property or payroll taxes. Until the 1960s, most hospitals and doctors had to do charitable work. Laws and codes of ethics said sick people should be treated even if they couldn’t pay. By 1969 most Americans were insured, so the IRS defined a new standard for hospitals that wanted to keep their tax-exempt status: these institutions had to provide “charity care and community benefit.”

Not-for-profit hospitals are now just as profitable as capitalist corporations, but the excess money flowing in isn’t called “profit”—it’s “operating surplus.” Charity Navigator, a group that rates nonprofit organizations based on their governance and use of donated funds, doesn’t even rate not-for-profit health systems because they function on such a different model.

It’s probably fair for hospitals to be able to count services for low-income patients (Medicaid or uninsured) as “charity care and community benefit,” because that practice brings in less than the cost of treatment. But since 1986, hospitals that care for large numbers of low-income people have already been compensated in other ways. They buy all their pharmaceuticals at a discount, through a federal program. Likewise, Medicare gives them so-called disproportionate share payments, essentially bonuses for treating higher numbers of poor people, who tend to be sicker and less able to pay bills. At least three of the twenty hospitals in the UPMC system collected these payments in 2014.

In 1986 Congress, in cooperation with the American Medical Association, commissioned a respected health economist, William Hsiao of Harvard, to create what was ultimately called the resource-based relative value scale (RBRVS). The goal was to scientifically determine what each physician service was really worth. The researchers on Dr. Hsiao’s team would assign a work value to each of the seven thousand codes then in the American Medical Association’s Current Procedural Code Manual.

Dr. Hsiao’s team calculated a work value in a new currency called relative value units (RVUs), based on (1) the work/time spent by a doctor for the visit or intervention, (2) the overhead incurred in rendering the service, (3) the cost of training required to learn to perform the service, and (4) the malpractice expenses involved. That RVU score was then multiplied by a conversion factor that was adjusted annually and varied slightly by location to determine a dollar payment. It was an elegant but complicated algorithm and—it turned out—one ripe for manipulation.

Attempts by Medicare to rein in inflation at the meetings are resisted. After Medicare decided to pay surgeons an all-inclusive rate for spine surgery (rather than paying separately for each step), representatives from the American Academy of Orthopaedic Surgeons and the North American Spine Society saw to it that the valuation of the bundle increased at the RUC meeting in April 2013.

Dr. Michael Zapf, an office-based podiatrist in California, has been approached by at least four surgicenters in his area to buy in as an investor/owner. “They guarantee a return on investment of more than one hundred percent,” he told me. How? The doctors sign up with insurance plans, but the centers themselves do not participate in any insurance networks, so the facility fees are not constrained by insurers’ negotiated rates. That allows them to bill $40,000 to $50,000 for what Dr. Zapf called “a simple surgery” like a bunion removal, when the standard office fee would be $3,000 to $4,000.

Groups of pathologists, anesthesiologists, radiologists, and ER physicians (PARE) followed, creating limited liability companies (professional LLCs) and becoming corporate contractors who sold their “physician services” to their former employers. Many did so even as they continued to work within the hospital. Emergency room physicians were the last to jump on board. When I worked as an ER doctor in the mid-1990s, we were all hospital staff. By 2014, 65 percent of the nation’s five thousand hospitals had contracted out their emergency department staffing/management function, according to Merritt Hawkins, a physician staffing company.

For those in the industry, it was a win-win situation: hospitals no longer had to buy malpractice or health insurance or figure out how to staff vacations. Doctors could charge what they felt they were worth. But for patients, this meant the proliferation of separate bills for these doctors’ services, from companies with mysterious return addresses in distant states. Then, around 2010, many of these doctors’ groups, who worked at in-network hospitals, simply stopped contracting with any insurers at all, leaving unsuspecting patients with tens of thousands of dollars in surprise medical bills.

Patients will typically only vis...

On a busy day, have you ever thought, “I wish there were two of me”? Over the last twenty years many doctors have managed that feat, at least for billing purposes. They do it with physician extenders. Physician extender is an umbrella term that refers to the trained ancillary personnel who help doctors and surgeons care for patients, including nurse-practitioners, surgical technicians, physician assistants, and midwives. In many countries they practice independently, but in most of the United States, according to law, they typically work under the auspices of an MD.

To pay a nurse near $150k and the ghost doctor another $500k to do the same task is just an example of how the medical community is pilfering.” Though a few states allow nurse-anesthetists to practice independently, anesthesia societies have fought vigorously to prevent that, citing safety, of course. In 2000 a press release from the American Society of Anesthesiologists targeting members of Congress ominously declared that “seniors will die” if nurse-anesthetists are not supervised by anesthesiologists.

In Mr. Jordan’s case, Medicare at first paid both the anesthesiologist and the nurse-anesthetist, each getting the approved rate of about $700. When he called Medicare auditors to complain about what seemed like double billing, they rescinded the physician’s payment. So Old Dominion sent him a bill, which he refused to pay, adding, “Either they wrote it off or they’ve sent me to collection and I don’t care which it is.”

MEDICINE IS A BUSINESS. It won’t police itself,” said William Sage, a doctor and a lawyer who is a professor of health law at the University of Texas. “People had a lot of faith in the American medical profession—that they would act differently than other businesses—but they were wrong.”

Most modern pharmaceutical firms grew out of nineteenth-century small businesses, which sold tonics and potions of often-dubious benefit. They relied on a little bit of science and a heavy dose of marketing.

To get products approved, firms had to create applications to prove that medicines were “safe and effective,” meaning more effective than doing nothing at all. That standard was never refined to include the more modern question: Is the product more effective than the dozens of other treatments for a particular condition that are already on the market? Equally important, the FDA yardstick for approval did not include any consideration of price or measure of cost-effectiveness—a metric that virtually all other countries now use as they consider admitting new drugs to their formulary.

To speed the arrival of generics, the act dictated that applications from makers of copycat drugs to sell their products no longer had to include fresh clinical trials to demonstrate safety and efficacy, but could rely on prior studies done by the makers of the brand drugs. The generic manufacturers’ abbreviated new drug application (ANDA) merely had to show that the chemical compound was the same as the branded drug at the same dose, produced the same levels of medicine in the body (bioequivalence), and violated no patents, which had to have expired.

Despite episodes like these, drug advertising is now a constant in our lives. The Supreme Court has protected drug advertising under the guise of free speech. We are one of two countries that allow it, along with New Zealand. Media companies, and particularly cable television stations, are ever more dependent on it for survival; healthcare advertising stayed robust even through the Great Recession.

Pharmaceutical companies like to say it takes well over $1 billion to bring a new drug to market: the costs of the basic science, developing a new compound, figuring out the right dose, and the FDA process of human testing for safety and efficacy. (Many companies also include opportunity costs—the profits that could have been made by investing the money elsewhere—in the estimate.) In some cases, that is likely true. But academic studies have placed the actual average scientific research and development costs for a new drug at between $43.4 million and $125 million. It is unclear how much of PhRMA’s typical $1 billion estimate is for testing markets, advertising, and promotion.

The FDA granted Vanda’s Hetlioz the green light in January 2014. To his consternation, Dr. Sack was an expert on the panel that had unanimously recommended the approval a few months before. “I came out thinking our system of drug approval doesn’t make any sense,” he told me two years later. “Because of the FDA’s criterion we were following to determine efficacy, I was forced to vote yes, but it made me deeply uncomfortable.” It was not demonstrably better than melatonin itself. FDA panels don’t consider the cost of the products they review. When Hetlioz received approval, Dr. Sack assumed it would be maybe $10 or $15 a pill. In fact, Hetlioz was later priced at $96,000 a year, $8,000 a month, or $267 per dose. “So now we have a prescription medicine for over two hundred dollars a day that is not as effective as melatonin used properly—which you can buy for six dollars over the counter for a bottle of one hundred,” Dr. Sack said. “It is eye-popping that we could get in this situation. And that it would be advertised on TV.”

Determined and creative, Vanda adopted a strategy for expedited approval as a first-in-class drug to treat Non-24, an orphan disease. Instead of testing tasimelteon rigorously in the general population, the testing could involve dozens rather than tens of thousands of patients, and could last six months instead of years. It could more readily focus on highly fungible surrogate endpoints.

Vanda and the FDA originally agreed that nighttime total sleep would be the proper metric of success for the study, but Vanda had backed away from that. Instead of focusing on all nights, they included only the 25 percent of nights when subjects’ sleep times were lowest. On those nights, patients who took the drug slept fifty-three minutes more. Vanda also selected two other surrogate measurements to prove its drug’s efficacy: a urine test of melatonin and a “Non-24 quality of life scale,” designed by Vanda, which measured only days 112 to 183 of the study. “Changing the endpoints of an ongoing clinical trial—especially without the FDA’s blessing—is a sign of trouble,” Mr. Feuerstein wrote. “The story Vanda tells of its sleep-disorder drug doesn’t have a happy ending.” But it did.