What Doctors Don't Know.

HMOs aren't the only thing wrong with American Medicine

In Edgar Allan Poe's "The purloined Letter," the protagonists overlook the crucial clue that is sitting right in plain sight: a letter left casually on a desk. Today, a similar nearwillful blindness prevails among those who present the managed care industry as the leading threat to the quality of American medicine. Implicitly, this view defines high-quality care as consisting of immediate access to treatment. While that is obviously important, the critics ignore a separate issue whose equal significance should be glaringly obvious: What happens to the patient once he or she actually reaches the doctor or hospital?

It would be nice if good medical care simply consisted of preferring the physician's judgment over that of "insurance company accountants," as President Clinton put it in his 1998 State of the Union address. Indeed, the argument that the public is faced with a choice between medical decisions made by "good" doctors or "bad" bureaucrats (albeit private-sector bureaucrats) has been sounded by everyone from anti-managed care politicians to (surprise) indignant representatives of various physician organizations. Preserving physician freedom is presented as the way to "preserve" high-quality care; any other path leads inexorably to ruin.

Unfortunately for the health of patients, this story line is a gross oversimplification. There's no question that the business ethic of some health plans can dangerously distort medical decision-making and has sometimes done so. Yet the larger truth about our health-care system is at once more complicated and much more unsettling.

From ulcers to urinary tract infections, tonsils to organ transplants, back pain to breast cancer, asthma to arteriosclerosis, scores of thousands of patients are dying or being injured every year because the best scientific information on how to care for them is not being put into practice by physicians. If one counts the lives lost to preventable medical mistakes, the toll jumps even higher.

In the scientific literature, the struggle to put medical theory into practice goes by the genteel term "evidence-based medicine" Kenneth I. Shine, president of the Institute of Medicine, phrases the problem this way: "If we asked the question of whether physicians have based their practice on scientific principles," he says, "it is clear that the profession has been sorely lacking."

Put in Clintonian campaign terms, however, the problem is easy for a layperson to understand: "It's the doctors' decisions, stupid."

What Doctors Don't Do

Most patients would be surprised to know what their doctors don't know--or don't put into practice. Even the best-trained doctors go about their work with an astonishingly shallow base of knowledge concerning the link between what they do and how it affects a patient's health. For instance, more than half of all medical treatments, and perhaps as many as 85 percent, have never been validated by clinical trials. But even when there is scientific evidence about what works best, large numbers of doctors don't apply those findings to actual patient care.

A prime example of the latter problem is the treatment of ulcers, a common and painful condition that will afflict an estimated 25 million Americans at some point in their life. Back in 1988, the prestigious British journal The Lancet published the results of a clinical trial that showed that most ulcers were actually caused by a bacteria called H. pylori. Patients didn't have to spend their lives taking anti-ulcer medications to control acidic secretions; the bacteria could be eradicated and the ulcer cured.

A study replicating this research appeared in a major American journal in 1991. A consensus panel from the National Institutes of Health endorsed the anti-bacteria ulcer treatment in 1994. Yet today, half of all U.S. doctors still aren't testing their ulcer patients for H. pylori and prescribing the right therapy, according to the federal Centers for Disease Control and Prevention. Managed-care financial incentives have nothing to do with this problem. Indeed, if anything they should work in the opposite direction--it's cheaper to cure a patient than to keep paying for maintenance drugs.

But to truly understand the consequences of failing to apply evidence to practice, one need only look at treatment of heart disease, the number one killer of both men and women in the United States. Heart disease is common, and it is expensive (total health care costs for cardiovascular disease are more than $150 billion annually). To listen to the managed-care debate, one might think the most pressing problem in cardiac care is getting people directly to the hospital emergency room without waiting for pre-approval from their health plan. (And forget for a moment that only 1 of every 9 visits to the emergency room for chest pain results in a confirmed heart attack.) The medical literature, however, paints a very different story than the one heard on the nightly news. It's a tale of wasted money and of unnecessary patient deaths and complications that has nothing to do with health plans. To give just a few examples:

* Inappropriate invasive procedures are distressingly prevalent. Harvard University researchers found that one quarter of the heart bypasses, angioplasties and catheterizations (measuring blood flow and blood pressure in the heart) performed on elderly heart attack victims have no effect on patient survival. In other words, thousands of these expensive treatments--which pose a significant risk of complications or death--could probably be eliminated.

* Proven drug therapies, meanwhile, go unused. A New Jersey study found that just one-fifth of eligible heart attack victims received beta blocker therapy that could have increased their chances of survival by as much as 43 percent. Other studies have found much the same problem. By one estimate 18,000 people die each year from heart attacks because they did not receive effective interventions. Meanwhile, one-third of heart-attack survivors in one study left the hospital without their doctor telling them to take an aspirin a day to help prevent another attack.

* Some hospitals are performing heart surgery even when their doctors haven't performed the operation often enough to keep their skills up. There is a long and well-documented connection between a hospital's performing a high volume of bypass surgeries and increased patient survival. Yet nearly one-third of the 1,023 U.S. hospitals performing open-heart surgery in 1994-95 performed fewer than the 200 cases per year that groups such as the American Heart Association recommend as a minimum, according to Medicare data. In California, 56 percent of the hospitals were below the 200-case minimum, a recent study found. It makes a difference. In New York State in 1995, one hospital that performed only about half the recommended number of bypasses had a morbidity rate over twice the state average. Traditionally, bypass surgery has traditionally been a prestigious and highly profitable service. That may or may not explain why the profession has not enforced its own standards.

Reports like these led to a damning indictment of the quality of care by the prestigious National Academy of Sciences' Institute of Medicine. The IOM report, in the September 16, 1998 issue of the Journal of the American Medical Association (JAMA), cited some of the above examples and concluded that "serious and widespread quality problems exist throughout American medicine ... very large numbers of Americans are harmed as a direct result." The IOM consensus report added pointedly, "Quality of care is the problem, not managed care."

The Making of a Non-Scandal

One would think that the glaring disconnect between research-based medicine and everyday practice would be a national scandal. The evidence, after all, is in plain sight. Moreover, some big names in the policy world are even starting to talk about the problem.

Earlier this year, the presidential Advisory Commission on Consumer Protection and Quality issued an extraordinary indictment. The commission wrote: "Exhaustive research documents the fact that today, in America, there is no guarantee that any individual will receive high-quality care for any particular health problem." American medicine, it added, was plagued by "overuse, underuse and misuse." In other words, the heart disease examples cited earlier are the rule, not the exception.

The press, however, glued to the politically hot "patient bill of rights," has virtually ignored other causes for the wildly varying quality of patient care. The reaction to reports on deaths due to preventable medical mistakes has been ho-hum. Earlier this year, when the Centers for Disease Control and Prevention released a report stating that 50,000 people die each year from infections caused by their hospitalization, The New York Times put the story on page A12. A close reading of the story revealed that the last time the CDC looked closely at infections in hospitals was nearly 25 years ago! What went unstated in the report is that no governmental or private agency requires hospitals to have either a standardized definition of an infection or report on its infection rate to anyone.

So where is the outrage? In large part, the absence of scandal may be due to the equal absence of identifiable victims. The classic medical expose is about "bad" doctors. A man who has had the wrong foot amputated by a surgeon or a woman who has been sexually molested by her internist knows something went wrong. And, indeed, incompetent or impaired doctors remain a problem. Public Citizen Health Research Group estimates that one percent of doctors nationwide "deserve" some disciplinary action each year, yet Federation of State Medical Boards data show an average of just 3.84 actions per 1,000 physicians in 1997. Moreover, patients in a state like Florida may wait as long as two years for their complaint even to be heard.

Similarly, health-plan members with access problems often know when they've been mistreated, as was the case with the Washington representative of a health-care organization who was told that being taken unconscious to the ER was no excuse for not getting pre-authorization approval.

By contrast, the heart-attack patient who doesn't receive beta blocker therapy or isn't told to take an aspirin on discharge most likely won't have a clue that his life was put at risk. That's particularly true if some sort of therapy is prescribed and the doctors and nurses appear to be working hard and doing their best. (Of course, a few patients don't take any chances. An older physician I know who lives in a rural area carries beta blockers and aspirin in his wallet, along with instructions to the hospital on their use, just in case.) Similarly, the patient who receives an unnecessary angioplasty has no way of knowing that it was unneeded--except, perhaps, when some health plan reviewer objects, and the patient becomes outraged that the HMO is rationing care. And in the case of hospitals doing a low-volume of bypass surgery, it's hard to prove that any individual patient was harmed.

Perhaps the best example of a quality lapse where the victim is unlikely to complain involves autopsies. A recent JAMA study of autopsy results found that doctors failed to diagnose cancer properly in a shocking 44 percent of cases. Yet the frequency of autopsies is at a record low, in part because an unperformed autopsy is likely to generate a grateful family, not a dissatisfied one ready to complain to its local TV news team.

But it's not only the lack of easily-identifiable victims that's kept evidence-based medicine off the public radar screen. There are also political and economic reasons. Take the American Medical Association, which for a time actively sought out HMO horror stories for the media and Congress. While the AMA's scientific arm is a leader in the fight for evidence-based medicine, its political arm is devoted to protecting doctors' autonomy. Sometimes, that's propatient, as when the AMA battles unreasonable managed-care rules. Other times, it's pro-pocketbook of doctors, as when the AMA defines quality of care as forcing plans to contract with "any willing provider." That kind of arrangement takes away a plan's ability to compete on quality, because it's no longer free to decide which doctors will be part of the plan.

Public discussion of evidence-based medicine might lead some to wonder whether unchecked physician autonomy is really such a good thing. Those whose memories stretch back more than two news cycles might even remember that the Health Maintenance Organization Act of 1973 emerged out of widespread public disgust with a fee-for-service system then run by doctors.

Indeed, in 1976, a time when less than 10 percent of the population belonged to an HMO, the Times ran a five-part, page-one series about medical errors, overprescribed drugs and other failings of American medicine. The date of those articles should be noted by the nostalgia-mongers who loudly contrast today's system with the days of "Marcus Welby-style medicine? Nineteen seventy-six was the year that Marcus Welby, M.D. finally left the prime-time airwaves. Even in Marcus Welby's time, Marcus Welby-style medicine was not a reality for patients; it was just a TV show.

While the physician groups have an economic agenda that can cause them to slight evidence-based medicine, some consumer groups have a political one. Liberal organizations like Public Citizen seem to want to discredit managed care in order to pave the way for Canadian-style national health insurance. While that's a worthy goal (and one this magazine has long supported) it's unlikely to succeed in the current climate. More to the point, addressing the legitimate problem of access to care by the millions of Americans without health insurance says nothing about the quality of care received once access is assured. Conservative groups such as the National Center for Policy Analysis are as obsessed as the liberals with access and financing issues. But unlike the liberals, they want to replace managed care with a national system of medical savings accounts, another policy fantasy.

Given these ideological and economic interests, today's medical scandals follow a predictable script. For example, the refusal of some health plans to allow some breast cancer patients to stay in the hospital overnight after a mastectomy created a national uproar. Yet there is persuasive medical evidence that suggests a one-night stay can be perfectly safe--if health plans and hospitals choose the patients carefully and put the proper support services in place. More to the point, if breast-cancer victims are the real concern, then we should consider the lack of attention given a study appearing about the same time as the "drive-through mastectomy" debate erupted.

This study, by Dartmouth Medical School researchers, found that "geography is destiny" when it comes to care for breast cancer (as well as a number of other conditions). For example, fewer than three out of every 200 breast-cancer patients in Rapid City, S.D. received breast-conserving surgery versus a mastectomy, while in Elyria, Ohio, 96 out of 200 did. The two procedures have roughly equal therapeutic efficacy.

Which is worse: being given a too-short hospital stay after a mastectomy or unnecessarily losing a breast to surgery in the first place? To even pose that question exposes the shallowness of viewing high-quality care as synonymous with deference to individual physician judgment.

Since women's health issues are so politically attractive, here's another example of practice variation every woman will understand. When family practitioners in Washington State were asked about treating a simple urinary tract infection, one of the most common of female complaints, eighty-two physicians came up with an extraordinary 137 different strategies! It defies common sense to believe that every patient of each of those doctors is receiving equally high-quality care.

(Speaking of non-scandals: why is hospital "dumping" of poor patients without insurance--which once made front-page news--now relegated to the trade publications? Are the only certifiable "victims" the victims of health plans?)

The good news is that there are, in fact, some successes by health plans, hospitals and doctor groups in implementing science-based medical practice. The efforts of United Healthcare, a national for-profit plan, were even profiled recently in The Wall Street Journal. There are others. In Boston, for instance, Harvard Pilgrim Health Care has pioneered a program to involve asthmatic children, their parents, and the plan's doctors in applying "best care" The program cut inpatient admissions and emergency room visits by kids by more than three-quarters. Not only that, the children reported a significant increase in their ability to function at school or home.

Yet many organizations hesitate to share their successes at making use of medical procedures that are already on the books. A moment's thought makes it obvious why: A hospital or doctor who excels at using scientific advances that have already been in the medical literature risks being greeted by the question: What were you doing before this? The questioner might even be a lawyer.

Despite these barriers, I've seen a number of hospitals that are dedicated to operationalizing better care. In Pittsburgh, for instance, Forbes Regional Hospital used evidence-based clinical guidelines to improve the recovery time of seriously-ill pneumonia patients and dramatically cut the death rate of those who were less ill. In South Carolina, changes in heart-attack care at the Anderson Area Medical Center saved the lives of an estimated two patients a month who previously would have died. In Los Angeles, Cedars-Sinai Medical Center's systematic review of the medical literature on a heart condition in infants called Tetralogy of Fallot found that one treatment meant to prevent respiratory infections in these newborns actually increased the infection rate!

The common thread in all these improvement stories is this: There are no "bad guys" The doctors involved were neither evil nor incompetent. They are your doctor and mine--which may be the most powerful explanation of all for why the problem of evidence-based medicine has attracted so little attention. Whatever the failings of doctors in general, "to distrust one's [own] doctor is to be vulnerable in the most fundamental and undesirable of ways," as one medical sociologist wrote. "The image of the doctor in America continues in large part to be an idealization that reflects people's hopes rather than their actual experiences"

Yet "good" doctors don't follow best practice for a variety of reasons. There is a failure to keep up with an overwhelming flood of medical literature, a reluctance to get caught up in new therapies that may be fads and uncertainty over which patients are best-suited for which therapy. Certainly, financial incentives can play a part in overuse, underuse and misuse. But ideologues who believe there is a perfect payment system that guarantees high-quality care will be disappointed. The problem of medical mistakes and a failure to practice evidence-based medicine is international. And, of course, there remains the problem of treatments where the evidence is ambiguous or simply unavailable.

A Better Way

When it comes to protecting patients, the same kind of pressures being brought on health plans should also be applied to the actual providers of care. While President Clinton has declared that passing a patient bill of rights will be a top 1999 priority, legislation that focuses only on speedy access to care will ultimately fail to truly protect patients. Providing every patient with the most appropriate and effective care--"doing the right thing and doing the right thing right"--must be just as much a priority as guaranteeing access to that care.

Those physicians who are working hard to bring change to our present system need financial and political encouragement, just as those who are reluctant to change need a mixture of financial and political incentives to rethink their position. Take, for instance, the connection between low volumes of surgery and higher mortality rates. If the patient bill of rights guarantees the right to go to the emergency room without delay, then how about the right to know, before the anesthesia takes effect, how many times the hospital or an individual surgeon has performed heart surgery? Medicare rule-making could go a long way to encourage this type of disclosure.

Similarly, what about a system of disclosing hospital infection rates that parallels the ratings used by health departments for restaurants? Right now, hospital lobbies across America display a certificate of accreditation from the Joint Commission on Accreditation of Healthcare Organizations, a group whose workings few patients understand. On the other hand, the difference between an "A" and a "C" grade on a certificate rating infection control would be an instant conversation-starter for patients and their doctors.

At the same time, legislators must be careful not to undermine internal hospital improvement efforts. Organizing information in a more systematic way, particularly through the use of computers, will both give doctors useful data at their fingertips and allow outside overseers to control better. Software now available can provide doctors with treatment protocols based on the latest science or a database of drug interactions designed to protect patients against possibly deadly adverse reactions. Other software, involved in electronic medical records, can help doctors and hospitals track the outcome of care in different groups of patients and continuously improve it. Unfortunately, the news stories highlighting legitimate worries about privacy of electronic medical records have not given equal play to the improvements in patient care that electronic medical records can bring.

Meanwhile, those researching how to apply evidence more effectively to practice could use far more funding than their present few crumbs. While Congress and the president heap billions more onto NIH's plate, the little-known Agency for Health Care Policy and Research makes do with a budget in the $100 million range. Funding applications work in medicine is like buying bullets and boots for the military; it's not glamorous, but it can mean the difference between success and failure.

Evidence-based medicine may be moving toward a higher profile. The Washington-based National Coalition on Health Care and the Boston-based Institute for Healthcare Improvement has just launched a quality initiative called Accelerating Change Today (ACT) for American Health. The explicit aim is to put best practices into common practice. Similarly, several corporate coalitions including some of the most prominent names in American business have quietly been holding talks about coordinating their efforts in a national effort to upgrade standards of care.

As the cost of medical care starts to accelerate again, it is becoming more urgent than ever to provide only that care which is the most appropriate and most effective. The alternative is grim: buying whatever care is cheapest. It's about time we started measuring the performance of health professionals and of health plans alike--and insisting that both groups be held accountable for results.

Michael L. Millenson, a Chicago based consultant, is the author of Demanding Medical Excellence: Doctors and Accountability in the Information Age (University of Chicago Press).