There are seven deadly sins, and Las Vegas exists to provide one convenient location to sample all of them. The Heart Attack Grill is the place for gluttony.1 Its voluptuous “nurses” serve food that will kill you if eaten too often. There’s the 10,000 calorie Quadruple Bypass Burger, a banquet built with 4 half-pound beef patties, 20 strips of bacon, and 8 slices of American cheese. Pair one with an order of Flatliner Fries (fried in lard) and a Chocolate Butterfat Shake, and you’ve got Vegas’s version of a Happy Meal.
Too much happiness can be lethal. Some of the Heart Attack Grill’s most loyal customers learned this the hard way. While being interviewed on a business-oriented news program, “Dr.” John Basso, the owner, held up a plastic bag that contained the ashes of a recently deceased patron who was also a spokesman for the Heart Attack Grill. Basso challenged other restaurants to be as candid about the dangers of their menu items as he is about his.
The Heart Attack Grill’s business model is unusual. Normally, killing customers—or even harming them—is a bad idea. But health care providers that harm patients can make more money by doing so. They get paid once for delivering services that injure patients and a second time for treating the injuries they inflict. This may explain why health care providers harm patients so often.
How often? The matter is subject to disagreement, but, as Scientific American reported, “every time researchers estimate how often a medical mistake contributes to a hospital patient’s death, the numbers come out worse.”2 A decade ago, the Institute of Medicine made front-page news when it conservatively estimated that medical errors killed 44,000 to 98,000 hospitalized patients each year.3 Just four years later, HealthGrades, a private organization that rates health care providers, doubled the count, putting the annual death toll at 195,000.4 Then, in 2013, the Journal of Patient Safety published a meta-analysis of prior studies that said “preventable harm to patients” causes more than 400,000 premature deaths per year.5 (A 2016 study by researchers at Johns Hopkins University School of Medicine broke the upward spiral of estimates by putting the tally of error-related deaths at 250,000.)6 If we accept the highest estimate, medical errors are the third-leading cause of death in the United States. Only heart disease and cancer kill more often.7 Even if we use the lowest estimate, medical errors are still the eighth-leading cause of death in the United States, just behind diabetes and just ahead of suicide.
Fortunately, most injured patients survive. Of course, that means that the total number of injuries substantially exceeds the number of fatalities; “serious harm [is] 10- to 20-fold more common than lethal harm.”8 More than 6 million patients may be injured each year.9
In 2011, Health Affairs published a series of articles that collectively paint a portrait of medical injuries and their associated costs. One group of researchers found that adverse events occurred in a shocking 33.2 percent of hospital admissions, ten times as often as previously thought.10 Another study estimated that medical errors generated over $17 billion in direct medical costs in 2008 alone.11 A third study (based on the amounts people would pay to avoid risks to their health) found that medical errors entail annual social costs ranging from $393 billion to $958 billion.12
It has been clear for decades that medical injuries result in increased health care utilization. In 2007, an article published in the Journal of the American Medical Association estimated that quality problems in outpatient care and medical errors alone necessitated “116 million extra physician visits, 77 million extra prescriptions, 17 million emergency department visits, 8 million hospitalizations, [and] 3 million long-term admissions,” in addition to almost 200,000 deaths.13 How much larger would those figures be if we included hospitalized patients?
Is it possible that these figures are as high as they are because errors generate revenue and profits for health care providers? Apart from an occasional monster, health care providers don’t deliberately harm patients in order to make more money. They claim to take adequate precautions and remind us that bad things happen to their patients because their patients are sick. This is the Reverse Lake Wobegon Effect in action. For decades, it was the conventional wisdom regarding hospital-acquired infections (HAIs), which, according to a 2007 estimate by the Centers for Disease Control, afflicted 2 million patients a year, killed nearly 100,000, and cost over $25 billion to treat.14 Hospitals claimed they could do little to reduce these numbers because sick patients with weakened immune systems were easy prey for opportunistic bacteria.
The Quality and Safety Research Group at Johns Hopkins Hospital, led by Dr. Peter Pronovost, put the lie to this claim by proving that hospitals can protect patients from HAIs easily and cheaply. His team focused on infections suffered by patients with a central line who are treated in an intensive care unit (ICU). A central line (also called a central venous catheter) is used to give fluids and medication to critically ill patients in an ICU. It is typically inserted in the neck. As we discussed in the Introduction, an infection of the central line, called a CLABSI, for “central line–associated blood stream infection,” is often lethal. As Dr. Pronovost and his team observed, “Each year in the United States, central venous catheters may cause an estimated 80,000 catheter-related bloodstream infections and . . . up to 28,000 deaths among patients in [ICUs].”15
CLABSIs are almost entirely preventable, and the cost of preventing them is low. When inserting central lines, physicians and hospital staffers need only follow a simple, five-item checklist. They must:
They also have to avoid putting central lines in the groin and remove them at the earliest opportunity. That’s it. No special knowledge, expensive drugs, or fancy technology is required. As Dr. Atul Gawande, the health services researcher and author of a book on the benefits of using checklists in health care remarked, “These steps are no-brainers; they have been known and taught for years.”17
And these steps work not just at fancy academic medical centers, but everywhere. Pronovost’s group tested their checklist at 103 ICUs in community hospitals in Michigan. Before the trial, their average CLABSI rate was 7.7 infections per 1,000 catheter-days. After 18 months using the checklist, the average was down to 1.4 infections per 1,000 catheter-days. That’s a decline of 82 percent. The median infection rate was 0, meaning that at least half the ICUs reported no CLABSIs at all.18
Gawande described the Michigan experiment this way: “The results were stunning. . . . Over 18 months, the program saved more than 1,500 lives and nearly $200 million.”19 The improvement was so dramatic that Michigan’s “average I.C.U. outperformed ninety per cent of I.C.U.s nationwide.”20
The obvious question—obvious to us, at any rate—is: why weren’t Michigan hospitals protecting intensive care patients from CLABSIs before Pronovost showed up? Why weren’t hospitals everywhere doing so? When inserting central-line catheters, health care workers nationwide weren’t taking precautions that were known to work. Worse still, some hospitals had adopted checklists, but their personnel often failed to comply. In interviews conducted by the American Association of Critical Care Nurses, two nurses told of their experiences with physicians who refused to follow established protocols.
Inserting central line at bedside in ICU. Used checklist but surgeon refused maximal sterile barrier and in fact, ridiculed me and hospital staff for instituting (this precaution) when there is no “proof ” it works. Hospital does not allow RN to stop procedure so it was inserted without maximal sterile barrier. . . .
A cardiovascular surgeon was putting in an arterial line at the bedside. We have a checklist that must be completed for line placement that includes full barrier, washing hands, etc. The M.D. refused the sterile gown, mask, hat, and drape, and used only sterile gloves. The nurse offered the full barrier again telling him that all lines were put in with full barrier in our unit. He continued with the procedure. The bedside nurse did not feel empowered to stop the procedure. She later took the problem to the unit manager. No action was taken.21
These stories are far from unique. Before the Michigan experiment got underway, Pronovost asked nurses at Johns Hopkins to spend a month observing doctors and to record how often they completed each step when inserting central lines. “In more than a third of patients, [the doctors] skipped at least one.”22
Nor is the failure to follow known patient-protecting protocols limited to central lines. As Gawande observed, it occurs in many treatments:
A large body of evidence gathered in recent years has revealed a profound failure by health-care professionals to follow basic steps proven to stop infection and other major complications. We now know that hundreds of thousands of Americans suffer serious complications or die as a result.23
We know what works, and what works may be simple and inexpensive. But there is a widespread and deadly lack of implementation.
Again, Pronovost’s experiment proves the point. Michigan ICUs brought down their CLABSI rates because Pronovost went to great lengths to ensure that hospital personnel always took all five steps. When the Michigan Health and Hospital Association invited him to try out his checklists, he didn’t move to implementation directly. Instead, he asked personnel at participating ICUs to gather data on their own infection rates.
They found [that] the infection rates for I.C.U. patients in Michigan hospitals were higher than the national average, and in some hospitals dramatically so. Sinai-Grace [Hospital, in inner-city Detroit,] experienced more line infections than seventy-five per cent of American hospitals. Meanwhile, Blue Cross Blue Shield of Michigan agreed to give hospitals small bonus payments for participating in Pronovost’s program. A checklist suddenly seemed an easy and logical thing to try.24
Pronovost’s efforts to get ICU personnel to buy into his experiment didn’t stop there, either. Participating hospitals had to appoint project managers, who would roll out the checklists and be point persons for compliance; they had to empower nurses to stop doctors who missed steps (which was a radical realignment of authority); and they had to assign senior executives to visit their ICUs, listen to complaints, and solve problems.
The last step was truly extraordinary. Senior executives never show up at the sharp end of hospitals—the areas where patients are treated—but Pronovost was right to bring them in. Many ICUs had logistical problems. For example, few had adequate supplies of chlorhexidine soap or full-body sterile drapes at hand. Administrators could solve these problems and did so once senior executives were fully engaged. Hospitals convinced a medical supply manufacturer to put all the needed supplies into a single kit. Hospitals in Michigan bought the kits and ensured they were readily available in the ICU.
Why did the Michigan experiment succeed? Before the experiment started, CLABSIs occurred far too often because hospitals hadn’t even identified, much less solved, a business problem: how to deliver high-quality services consistently. They needed to design, implement, and monitor a process so that it worked correctly over and over and over again.
We say that this is a business problem because every type of business confronts it. If you want to compete with Starbucks, Einstein Bros. Bagels, or even the local coffee shop down the street, you’ll have to serve thousands of cups of coffee, each of which must be fresh, hot, and brewed correctly. To do that, you’ll need ingredients, coffee recipes, clean equipment, and trained employees. You’ll also need a process that brings everything together. Your ingredients—coffee, creamers, syrups, sweeteners, stirrers, disposable cups, and so on—will have to be in the right places when needed. They will also have to be stored in ways that preserve their freshness. Your employees will have to use the right ingredients in the proper proportions, check brewing temperatures and times, throw out old coffee, and keep everything clean and in good working order, while providing service that is fast and friendly. To maintain quality day in and day out, you’ll have to monitor everything, hire “secret shoppers” to sample the wares, assess customers’ satisfaction, discipline or fire workers who fail to follow directions, and so forth.
A person who runs any type of business will quickly discover that it’s hard to deliver high-quality goods, services, and experiences to large numbers of customers consistently over long periods of time. Consistency doesn’t happen by accident; it requires planning, implementation, and monitoring. It also requires feedback and learning. Few things work right the first time. Every mistake provides an opportunity to improve, so mistakes have to be identified and studied. Processes may have to be revised or redesigned many times before “getting it right” becomes automatic.
Many health care providers have problems with logistics. When Pronovost started his work at Johns Hopkins, one of the country’s great medical centers, doctors took all of the steps on the checklist only 38 percent of the time. Hopkins’ CLABSI rate was a shocking 19 per 1,000 catheter-days—“one of the worst records in the country.”25 Compliance was low partly because doctors had to “go to eight different places to get all the items needed . . . caps were stored in one place, masks in another, gowns in yet another. To make things even worse, many items were altogether missing.”26 Johns Hopkins was exposing patients to deadly risks because the hospital with some of the best doctors in the world was bad at business.
At least Johns Hopkins knew what its CLABSI rate was. Many providers have no idea how well they are doing relative to others. They think they’re above average—until an external review shows that they’re not. For example, the staff at the Saint Raphael Hospital in New Haven, Connecticut, thought they were doing a good job of preventing HAIs. The hospital’s ICU had a checklist protocol in place and an infection preventionist on its staff. Then the state of Connecticut began to require hospitals to report infections and released its findings. Turned out, Saint Raphael’s ICU had one of the worst HAI rates in the state. “[W]hen we started comparing to our colleagues, to our neighbors,” said Diane Dumigan, the infection preventionist, “we realized we were the outliers. It was an ‘aha’ moment.”27
For-profit businesses don’t like this sort of “aha” moment. The department head who finds out that Consumer Reports just ranked his company’s products a standard deviation below the competition knows it’s going to be a bad day. Indeed, by the end of the day, more than one person at the company may be out of a job. That’s why successful businesses monitor error rates, benchmark themselves against their competitors, and take other steps to ensure quality.
Because hospitals rarely take these steps, outcome quality can vary enormously from one to another. “In the first comprehensive study comparing how well individual hospitals treated a variety of medical conditions, researchers found that patients at the worst American hospitals were three times more likely to die and 13 times more likely to have medical complications than if they visited one of the best hospitals.”28 The findings should put to rest forever the belief that all providers are above average, or even that all are roughly the same. Paraphrasing Dr. Barry Rosenberg, one of the study’s coauthors, a New York Times reporter observed that, “if someone has a heart attack, the closest hospital could have a death rate of 16 percent, compared with one a little farther away, where the rate was 4 percent.”29
Rosenberg’s study didn’t connect individual hospitals to their results, and it seems likely that few hospitals know where they stand.30 Why? For most health care entities, unlike most non-health care entities, they simply can’t make money by measuring or improving quality.31 A coffee shop owner knows that quality is a key to success. Happy customers will buy lots of coffee and tell their friends about the great place they found. Unhappy customers will take their business elsewhere and publish negative comments on Yelp. In the health care sector, the connection between quality and revenue or profit is far weaker. Sometimes, there is no link because no one knows whether particular providers are good or bad. Sometimes, the connection is upside-down, meaning that providers’ revenues and profits are higher if they do a rotten job.
Regulators don’t monitor quality either. Although facilities that receive federal funding are subject to government oversight, “the law allows hospitals, ambulatory surgery centers, home health agencies and hospices to pay private, national accrediting organizations for such oversight instead.” These private accreditors “often [miss] serious deficiencies found soon after [their reviews] by state inspectors,” who examine a sample of hospitals and other facilities each year. They also keep their findings secret. When CMS floated a proposal to require accreditors to release reports about errors and other quality problems at specific facilities, hospitals and accreditors vehemently objected, arguing that disclosure would “adversely affect the collaborative efforts of accrediting bodies and healthcare organizations to improve patient safety and engage in continuous quality improvement.”32 As is always the case with such arguments, they offered no evidence supporting this claim—and the available evidence indicates the opposite.33
But CMS found the argument to be persuasive, and it abandoned the proposal. The head of the Leapfrog group, which advocates for quality and transparency in health care, criticized CMS’s decision to back down:
This is disgraceful, unfair to patients as well as employers and other purchasers of health care. . . . The public deserves full transparency on how the health care industry performs. Instead, transparency has been sacrificed to accommodate special interests that lobby to avoid disclosing embarrassing information about health care quality.34
Does anyone doubt that the outcome would be different if health care providers were treated like ordinary businesses?
Don’t believe us that providers often fare better financially when they treat patients poorly? Consider CLABSIs again. In 2013, Pronovost published a study of billings for infected and uninfected patients in matched physical condition, all of whom spent time in the ICU at Queen’s Medical Center in Honolulu, Hawaii.35 For patients with private insurance who contracted CLABSIs, the charges averaged $495,000. The bills for uninfected patients were only one-fifth as large: $100,000. Infected patients generated much larger profit margins too: $55,000 per patient for those with CLABSIs versus $6,500 per patient for the uninfected.
This staggeringly large profit differential meant that Queen’s Medical Center had “no incentive to invest in prevention.”36 Indeed, if Queen’s Medical Center had taken steps to prevent CLABSIs, it would have lost money. Stated differently, hospitals have a “perverse incentive to have more line infections,” because the system pays more for patients with CLABSIs than for those who remain uninfected.37
This problem—that providers cannot make money by measuring or improving quality—is not limited to central lines and CLABSIs. Gawande’s research group at the Harvard School of Public Health recently published a study of the billing records for over 34,000 hospital inpatients, each of whom had one of nine common surgical procedures in 2010. More than 1,800 patients, 5.3 percent, experienced at least one potentially preventable complication, such as an infection at the surgery site or a pulmonary embolism. The patients with surgical complications generated $8,084 more in profit on average after the hospitals’ variable costs were netted out. Privately insured patients with complications were especially profitable, generating an average of $39,017 more in profit.38 Another study of a Michigan hospital found that every 1 percent decline in surgical complications cost the hospital $1.2 million.39 Finally, a 2016 study found that patients who had major surgery at low-quality hospitals cost Medicare thousands more than those treated at top-ranked hospitals because they required more postoperative care and were more likely to be readmitted.40
Perverse incentives like these permeate the health care sector. In one study, four programs designed to improve quality of care (by managing high-cost pharmaceuticals, improving diabetes management, reducing smoking, and providing workplace wellness programs) all ended up costing providers money.41 As three economically minded commentators observed in the New England Journal of Medicine, “It’s hard to create a favorable [return on investment] for reducing volume in a system dominated by fee-for-service payments for delivering care.”42 Our politically controlled payment system pushes providers to increase volume and punishes them for reducing volume, even when the services they eliminate are low quality or harmful. Even the most conscientious provider is unlikely to pursue money-losing quality improvements with enthusiasm.
Perverse incentives may also explain why it took the staff at Miami’s Jackson Memorial Hospital ( JMH) almost 20 years to gain control over an epidemic of HAIs caused by Acinetobacter baumannii.43 These infections are incredibly dangerous. A. baumannii is antibiotic resistant. Patients who contract it have an “all-cause” mortality rate of roughly 40 percent.44
JMH’s staff finally reduced its HAI rate by changing the hospital’s culture, not by applying sophisticated medical knowledge. Weekly emails to the hospital’s C-Suite, including maps that “showed not only how the infection was spreading but who was spreading it,” finally got the administration to take action.45 Once supported from above, nurses got serious about protecting patients. They spoke up when they saw people doing things that were “dangerous for infection control, such as failing to wash their hands or not cleaning the patient rooms properly.” They became “much pickier about their rooms being cleaned,” and demanded re-cleanings when rooms “[weren’t] up to snuff.” They maintained “zero tolerance” for doctors who violated infection-control practices and reminded physicians to take off their lab coats and wash their hands before examining patients. Infection rates plummeted.46
When a report of the results of JMH’s anti-infection efforts was submitted for publication, one reviewer observed that the article was “more management than science and entail[ed] organizational culture change.”47 Exactly. Many of the most important problems of the health care sector are business problems, not medical problems. They persist because our politically controlled, third-party payer system has taken away any “business case” for improving quality. That’s probably why JMH’s C-Suite took as long as it did to address the epidemic of HAIs that was stalking its ICU. It’s great that infections were finally brought under control, but it shouldn’t have taken two decades to build a medical culture that protected patients from an avoidable peril.
Why do these perverse incentives exist? The problem is not that doctors, physicians’ assistants, nurses, or other medical personnel intentionally and deliberately injure their patients. The dedication of health care professionals, who often work under difficult and unpleasant conditions, is awe inspiring. The problem is that health care providers operate in environments that do not subject them to financial pressures to improve and often pressure them not to improve. Fee-for-service payment arrangements—the kind of payment arrangements that government encourages and that lobbyists for health care providers defend—reward providers for quantity not quality, including services that are unnecessary, ineffective, or even harmful.48 Unsurprisingly, quality is often mediocre or worse. Inconsistent quality is not a medical problem. It is a business problem, and it requires business solutions. But, if investing in quality leads to lower revenue and profits, few will measure quality or improve it. Absent a business case for quality, quality won’t be measured and faulty delivery systems will remain unchanged. When it comes to delivery systems, health care is like any other business. Providers respond to incentives, and they need better incentives to improve.
Many health care professionals and medical ethicists will find these assertions jarring, if not reprehensible. They draw a sharp distinction between health care providers and businesses of other types, and they believe that people trained in medicine should have the final say over how health care businesses function. Doctors, nurses, and other medical professionals should not have laypersons interfering with their judgments or telling them how to work.
This view has two problems. First, it has been an epic failure. Patients are routinely hurt because quality problems go unaddressed. Medical professionals have been in charge of service delivery for a century, yet the evidence of high rates of preventable injuries and deaths is overwhelming. Second, this view fails to recognize the limits of medical training. Medical professionals study biology, chemistry, physiology, and pathology. They also have extensive clinical training. These experiences give them a leg up when it comes to diagnosing illnesses, selecting treatments, and performing procedures. But they receive little or no training in quality improvement. Ask a newly minted doctor what “Six Sigma” means and you’ll probably get a blank stare. You’ll get the same response if you ask about W. Edwards Deming, Joseph M. Juran, or James Reason, the lions of the quality improvement movement. Medical schools do not teach students how to reduce error rates by designing reliable delivery systems.
To the contrary, medical schools are famous for breeding attitudes that are antithetical to quality improvement. Instead of treating medical errors as opportunities to discover and address defects in delivery systems, medical schools teach doctors to criticize themselves and the people around them for being imperfect. Instead of learning how to draw on other providers’ knowledge, doctors often intimidate nurses and other assistants, causing valuable information to be ignored. The culture of medicine is famously punitive. It is also contemptuous of outsiders who lack medical degrees.49
The results are plain to see. Medical errors and preventable adverse outcomes are everywhere, as are unrealized opportunities for quality improvement. As one quality improvement researcher put it, in health care, “the low-hanging fruit isn’t just low-hanging fruit; the fruit is lying on the ground, and we have to be careful not to trip over it.”50
The history of anesthesia safety demonstrates the value of bringing in outsiders to improve quality. “In the 1970’s and 1980’s . . . 1 in 6,000 administrations of anesthesia resulted in death; and serious brain injuries were even more frequent.” Then, over about a decade, the mortality rate dropped “to one in 200,000 administrations”—a 97 percent reduction.51 Why? Because the leaders of the American Society of Anesthesiologists “set out to systematically identify and address the root causes of mistakes.”52 One of their most consequential decisions was to invite Jeffrey B. Cooper to study their errors and identify equipment-related causes.53 Cooper wasn’t a doctor. He was an engineer trained in critical incident analysis, and he had given a talk entitled “The Anesthesia Machine: An Accident Waiting to Happen” in 1974. By making anesthesia safer, Cooper helped protect millions of patients from harm. Even though he wasn’t a doctor, he showed anesthesiologists how to consistently deliver high quality and safe care.
One would hope that regulators, particularly those at the U.S. Department of Health and Human Services (HHS), would welcome efforts to make health care safer. But not always. Federal regulators at HHS’s Office for Human Research Protection (OHRP) shut down Pronovost’s Michigan experiment in 2007 because they believed that the use of checklists qualified as human subjects research and that the experiment had been carried out without review by an institutional review board (IRB). The IRB would have required the informed consent of the patients and the hospital personnel involved.54 OHRP’s action delayed similar experiments at ICUs in New Jersey and Rhode Island that were in the early stages of implementation.
Of course, hospitals could have continued to use their existing, defective procedures without obtaining IRB approval. That wouldn’t have been human subjects research or, indeed, research of any kind. It would, however, have infected thousands of patients every year and killed many of them. Hospitals could also have implemented checklists on their own, and that would have been fine too. Again, that wouldn’t have been research. But when hospitals, wanting to do better but not knowing how, brought in an academic consultant to systematically measure their error rates and show them how to do better, that was somehow dangerous human experimentation requiring bureaucratic review and informed consent.
Gawande, who condemned the OHRP’s action, noted the likely consequences of OHRP’s approach:
If the government’s ruling were applied more widely, whole swaths of critical work to ensure safe and effective care would either halt or shrink: efforts by the Centers for Disease Control and Prevention to examine responses to outbreaks of infectious disease; the military’s program to track the care of wounded soldiers; the Five Million Lives campaign, by the nonprofit Institute for Health Care Improvement, to reduce avoidable complications in 3,700 hospitals nationwide.55
Of course, research scientists should treat human subjects properly when evaluating the effectiveness of treatments and procedures. But efforts to get providers to take known patient-protecting steps simply don’t require the type of oversight IRBs provide. Fortunately, after being pressured by patient safety advocates, OHRP reversed course.
Patients desperately need there to be better incentives for providers to improve health care delivery. Gawande observes:
there are hundreds, perhaps thousands, of things doctors do that are at least as dangerous and prone to human failure as putting central lines into I.C.U. patients. It’s true of cardiac care, stroke treatment, H.I.V. treatment, and surgery of all kinds. It’s also true of diagnosis, whether one is trying to identify cancer or infection or a heart attack. All have steps that are worth putting on a checklist and testing in routine care. The question—still unanswered—is whether medical culture will embrace the opportunity.56
We’d bet on “probably not.” When ICUs in Michigan cut CLABSI rates to near zero, hospitals lost hundreds of millions of dollars in revenues. Unless we improve providers’ incentives, so they can actually make money by improving quality, we expect errors and other quality problems will keep harming and killing patients.
That’s the prediction of Drs. Gawande and Pronovost too. Despite the remarkable results the Michigan experiment produced, despite the successes other experiments with checklists in hospitals have achieved, and despite the common use of checklists in other industries, the health care sector has yet to deploy them in a systematic way, let alone require their use the way the airlines do. In a 2007 interview, Gawande asked Pronovost how long it will take for doctors and nurses to use checklists routinely. Pronovost’s reply: “At the current rate, it will never happen.”57
In the end, though, we think that Gawande misdiagnoses the problem by blaming the culture of medicine. “If someone found a new drug that could wipe out infections with anything remotely like the effectiveness of Pronovost’s lists,” he wrote,
there would be television ads with Robert Jarvik extolling its virtues, detail men offering free lunches to get doctors to make it part of their practice, government programs to research it, and competitors jumping in to make a newer, better version. That’s what happened when manufacturers marketed central-line catheters coated with silver or other antimicrobials; they cost a third more, and reduced infections only slightly—and hospitals have spent tens of millions of dollars on them.
But hospitals wouldn’t use checklists, he said, because “the prospect pushes against the traditional culture of medicine, with its central belief that in situations of high risk and complexity what you want is a kind of expert audacity. . . . Checklists and standard operating procedures feel like exactly the opposite, and that’s what rankles many people.”58
Of course there’d be rapid adoption of an infection-killing drug. Hospitals would buy it for $100 and charge patients $1,000. We bet that silver-coated catheters were a bonanza too. Whatever hospitals paid for them, they almost certainly billed insurers considerably more. That’s why hospitals were in no hurry to use checklists in 2007. Error reduction reduced their revenues. If government allowed error reduction to be profitable, checklists would be in use everywhere.
Hospitals made it a priority to take the recommended infection-preventing steps when, in 2008, the federal government and private payers stopped paying them more for treating patients who suffered HAIs and other preventable events and errors.59 The question is why, for so many decades, American hospitals allowed high HAI rates to persist while claiming there was nothing they could do about them. At least “Dr.” Basso is honest about the perils of eating at the Heart Attack Grill.