Are Large Hospitals the Problem with US Healthcare?
Two “distinguished” healthcare analysts have examined the medical system in the USA and conclude that the REAL problem is . . . large hospitals
Is the main problem with the US healthcare system that hospitals have gotten too large since the 1990s?
That seems to be the remarkable conclusion of two of the nation’s most distinguished health-policy analysts, David Dranove and Lawton R. Burns. Dranove is an economist and Walter J. McNerney Distinguished Professor of Health Industry Management at the Kellogg School of Management at Northwestern University. Burns is a sociologist and James Joo-Jin Kim professor of health care management at the Wharton School at the University of Pennsylvania. Dranove and Burns’s joint effort, Big Med: Megaproviders and the High Cost of Health Care in America, purports to offer special insights into the pathologies of the American healthcare system via an interdisciplinary approach.
I. From Hospitals to Integrated Delivery Networks (1811–2000)
The following is a summary of the complex evolution of hospitals in the US from 1811 to 2000, broken into four parts.
A. Hospitals (1811–1960)
The hospital story in the US begins in the early nineteenth century with the founding of Massachusetts General Hospital in 1811. More than a half century later, New York Presbyterian was founded in 1868 and Presbyterian Hospital was founded in 1893 (not 1883, as Dranove and Burns claim) to become no less than the flagship facility of today’s massive University of Pittsburgh Medical Center.
As Dranove and Burns aver, American hospitals in the nineteenth century were institutions where the sick were isolated from the rest of society or the injured or infected had their limbs removed. Hospitals only slowly evolved from dungeons of death and dismemberment to oases of life-saving treatments after the steady arrival of many innovations: the first and increasing use of nitrous oxide and ether as anesthetics (1844–46); the first use of vaccines for cholera, anthrax, rabies, typhoid fever, and plague (1879–97); the development of ABO blood typing and the first successful blood transfusion (1901–07); the use of insulin for the treatment of diabetes (1922); the first blood bank (at Cook County Hospital in Chicago, 1937); the use of penicillin from first dose to a production level of 650 billion units per month (1942–45); the first cardiac pacemaker (1952); and the first kidney transplant (1954).
B. Hospitals and Their Economic Enemies (1960–93)
The second era in hospital history consisted mostly of hospitals fighting or adapting to five major nemeses: health maintenance organizations (HMOs), price and supply controls, outpatient surgery centers, and selective contracting, ending with the Health Security Act of 1993.
HMOs. HMOs offered care that was prepaid and provided by staff working in teams. There was also an emphasis on illness prevention in some of them—hence the name “health maintenance organizations.” By 1950, they had caught on mostly in the West Coast states of California and Washington. Despite this limited success, hospitals and doctors saw them as serious threats to their incomes and profits and thus swung into action, lobbying state legislatures to forbid HMOs from advertising in any medium. HMO physicians came under severe persecution: many saw local medical societies revoke or deny them membership while some hospitals cancelled or denied them admitting privileges. Nevertheless, the industry slowly grew with the help of the HMO Act of 1973 and several regulatory and court victories.
Price controls. The back-and-forth fight between governments and the healthcare sector over the level and rate of healthcare spending came after the passage of the Medicare and Medicaid programs in 1965. Medicare overnight became the top buyer of health services in the nation while Medicaid became one of the largest components of each state’s budget. With federal and state budgets experiencing immediate stress and hospitals being a prime factor behind spending (more than a third by 1970), governments targeted hospitals. According to Robert Murray and Robert Berenson’s 2015 Hospital Rate Setting Revisited: Dumb Price Fixing or a Smart Solution to Provider Pricing Power and Delivery Reform?, the State of New York implemented fixed payments per day in 1971. Seven other states, Medicare, and private insurers copied the program, and hospitals responded by increasing the number of days per stay (i.e., patients stayed longer for care.)
Outpatient surgery centers. Until the early 1970s, surgery usually entailed a stay of at least a few days in a hospital. Outpatient surgery was an innovation for quick, minimally invasive surgeries (e.g., tonsillectomies and hernia repairs) that allowed a patient to be discharged the same day or early the next. The first independent facility was Surgicenter, founded in Phoenix, Arizona, in 1970. Patients liked ambulatory surgical centers (ASCs) because they were smaller, quicker, and easier to check in and out of than hospitals. Public and private payers liked their lower charges. According to Dranove and Burns, by 1980, a decade after Surgicenter was founded, about two hundred ASCs were operating in the US, and by 1988, more than a thousand.
Because ASCs posed a serious threat to hospitals’ inpatient surgery services, hospitals fought back in two ways: first, they opened their own ASCs connected to their main campuses. Second, while these new facilities were contiguous to the main hospitals, many were built with separate entrances to mimic the convenience of the independent centers. This convenience, wedded to close proximity to a hospital should complications arise, was the carrot dangled to patients, which was effective. The stick was wielded against insurers: all outpatient surgeries must be done at our hospital or we will not do business with you. Coming from hospitals with good reputations, these were serious and effective threats. After many insurers caved to these demands, independent ASCs sued hospitals, contending that hospitals were abusing their market power to hinder local competition. Hospitals argued that their markets were not local, and courts bizarrely and continually agreed with them, dismissing the lawsuits.
Selective contracting. Despite large insurers’ having five to seven times the revenue of the largest hospitals, insurers had little power over hospitals and physicians. Before the 1980s, insurers rotely paid all bills with nary an objection. Patients had few if any copayments and few if any obstacles to continually patronizing the costliest physicians and facilities. Only enrollees in stingy HMOs faced narrow provider networks. Dranove and Burns report that in this environment, “every provider, even small community hospitals and solo physicians, could price like a monopolist.”
Selective contracting in the 1980s changed this: in response to states’ no longer requiring insurers to reimburse all licensed providers, insurers formed preferred provider organizations (PPOs). PPOs arranged networks of providers, and enrollees seeking care outside their network paid high costs; thus, providers had a strong incentive to be network members. In return, insurers demanded and usually received significant discounts for letting providers into their networks. Enrollees gained lower-cost coverage but little or no change in access. Dranove and Burns report that PPO enrollment skyrocketed from close to zero in 1983 to approximately 28 million by 1987.
C. Hospitals Become Integrated Delivery Networks (1993–98)
The early years of the 1990s were a watershed. The hospital industry, especially after selective contracting, felt that it had lost the upper hand in its relationship with insurers. More and more, the answer seemed to lie in vertical integration, meaning that hospitals merged with physician practices, insurance plans, or both.
The year 1992 saw the election of Bill Clinton. First Lady Hillary Clinton and her healthcare advisor Ira Magaziner led the crafting of the Health Security Act, “Act” being a misnomer because the bill never passed Congress. The plan consisted of three main components: the establishment of regional state health alliances, the creation of insurance plans by local insurance firms and providers, and prices for services that would be set by the alliances. As Dranove and Burns state, “While the Clinton Plan died on arrival in Congress in the spring of 1994, it nevertheless frightened providers into a variety of integration efforts.”
One result of these efforts was that in just a few years hospitals turned into massive health systems, or, in the parlance of industry consultants and executives, integrated delivery networks (IDNs). There was horizontal integration, where hospitals bought or merged with other hospitals; vertical integration where hospitals bought physician practices and the physicians became hospital employees; and vertical integration where hospitals created their own health insurance plans. By one estimate, while only 16 percent of hospitals were part of an IDN in 1994, 50 percent were part of an IDN by 1998. So effective was the threat of the Clinton Health Security Act in driving consolidation that, while from 1980 to 1993 about three hundred hospitals merged, in 1994 alone there were over six hundred mergers. Then from 1995 to 1998 there were over five hundred mergers per year. After this, the merger craze lost steam.
D. Integration Fizzles (1998–2000)
After the fog cleared in the wake of the integration bacchanal came the mundane task of providing healthcare services. This proved rocky at best: in the eternal hunt for the economies-of-scale unicorn, costs soared and revenues nosedived. This was seen nowhere better than at three particular systems: the Allegheny Health, Education, and Research Foundation (AHERF) in Pittsburgh; Detroit Medical Center in Detroit; and Allina Health in Minneapolis. While each of the three is a complex case study in and of itself, the upshot is that while AHERF declared bankruptcy in July 1998 owing about $1.3 billion in debt in the largest healthcare nonprofit bankruptcy in US history, the two other systems performed little better than barely surviving.
II. History, Causality, and Integration
Dranove and Burns’s history of the industrial organization of the US hospital industry from 1993 forward is unquestionably second to none. A sweeping and thorough account of recent history, though, in no way necessarily evinces a sound etiology of industry dysfunction. The next section will juxtapose Dranove and Burns’s version of events with the classical free-market perspective, noting the differences and implications.
III. Methodenstreit over Healthcare
The “method struggle” in healthcare that is actually occurring is over how to best treat the sector’s economic malfunctions: high and volatile costs, costly access, exorbitant insurance premiums and high deductibles, pricey and too tightly controlled prescription drugs. The mainstream health economics and policy approach is to ignore the supply-and-demand factors that led to each of these problems and propose new interventions to alleviate them. A second approach is to examine the state of the industry before these malfunctions arose, analyze what led to each one of them over time, and then base reform on the resulting facts and knowledge. Dranove and Burns clearly embrace a version of the mainstream approach.
A. Was All Well before IDNs?
Clearly the two accounts (Dranove and Burns’s and the free market version) are not commensurate in their ability to explain the changes in prices, costs, output, and affordability of healthcare over time. The first question that Dranove and Burns’s account elicits is, “If US healthcare had few problems before IDNs (driven by the Clinton Health Security Act) appeared in the early 1990s, what spurred the creation of the Clinton Health Security Act?” Puzzling is Dranove and Burns’s account of the 1991 rise of Pennsylvania US Senate candidate Harris Wofford and 1992 presidential candidates Bob Kerrey and Paul Tsongas. All three candidates proposed significantly changing the US healthcare system in terms of improving affordability, access, or both. Instead of explaining why so many politicians made healthcare a campaign issue in 1991 and 1992 before the IDN wave came about, Dranove and Burns just drop the three candidates’ names on the way to introducing Bill and Hillary Clinton and the Clinton Health Security Act. The obvious answer to the question of why healthcare was a major campaign issue in the political races of 1992 is the American public’s widespread dissatisfaction with the pre-IDN US healthcare system. Dranove and Burns overwhelmingly focus on the high and runaway costs, limited access, and mixed quality of the post-IDN sector instead.
B. Kenneth, What Is the Frequency?
And then there are some seriously head-scratching statements about markets. “Before 1960, government largely stayed out of healthcare markets” is a stunning statement, given all the state actions in the wake of the Flexner Report (1910) and the establishment of Blue Shield and, later, Blue Cross. And then there’s “As much as we hate to admit it, there are limits to what we can accomplish through unfettered markets. We are even tempted to revive certificate of need (CON), despite its many shortcomings.” Unfettered markets? The current IDN-dominated sector? That is indefensible, especially given the authors’ own theory that mere proposed intervention (in the form of the Clinton Health Security Act) drove the initial IDN wave in the early 1990s. More of these contradictions and confusions are littered throughout the book, which is part of the reason why it is such a difficult book to read and follow. The experienced reader begins to wonder, Is it me or is it them?
While Dranove and Burns implicitly promise that their book will provide special insights into the pathologies of US healthcare because they are taking an interdisciplinary approach, that assurance could not be more unfulfilled. Chapters 1–6 (at most) form a worthy contribution to the history of the evolution of the American hospital from 1811 to today, with the chronicling of the IDN era from 1993 to the present particularly seminal.
However, selective history, regardless of its merits per se, in no way necessarily and accurately identifies the causes of the US healthcare industry’s many serious economic malfunctions. Thus, again, megaproviders are another symptom rather than a root cause of industrial disease. In that sense, Big Med is another all-too-typical contemporary analysis in health economics and policy. What is a priori ruled out is true market allocation of goods and services. New interventions and de facto central planning inevitably become the only acceptable solutions. Further, if Dranove and Burns are the “good guy” free marketeers, who needs interventionists and their state-monopoly Medicare for All? As Francis Jeffrey famously wrote in 1814 about William Wordsworth in the Edinburgh Review, “This will never do.”
[A longer version of this article was published in the Quarterly Journal of Austrian Economics.]