Food and Health
Arno G. Motulsky
Americans get bombarded with information about foods. Various food and food supplements taken to improve one's health are highly popular. Advisers to famous athletes tell us the virtues of certain dietary regimens.
Yet advice is often contradictory and the public has difficulty distinguishing scientifically based information from advice that has no solid basis.
What's a consumer to do?
Fortunately, we can answer that question with much greater confidence as a result of the widely publicized report on diet and health that was released earlier this month by the National Research Council. I chaired the committee of 19 experts that sifted through more than 5,000 scientific articles to prepare the report, which provides comprehensive information on the relationship between diet and chronic diseases such as heart disease, cancer and diabetes.
Although the report contained hundreds of statistics, the basic message to consumers was simple—namely, the current state of scientific knowledge allows certain conclusions about the relationship between diet and chronic illness. People can enhance their health by eating certain foods and avoiding others.
Conditions such as high blood pressure, coronary artery disease, obesity, osteoporosis and others are not caused just
by diet or any other single factor; they are multifactorial in origin. A person's genetic makeup, for example, plays an important role in "predisposing" to all these conditions. Age and gender may further modify one's predisposition; for example, women are at lesser risk for conditions such as coronary heart disease but at higher risk for osteoporosis and iron deficiency.
Ideally, one would like to be able to determine the unique genetic-biochemical-metabolic makeup of every person to identify who is at highest risk of developing a given chronic disease. Based on this information, a "tailor-made" dietary prescription may be possible sometime in the future. In fact, we already can detect persons at high risk for coronary heart disease by measuring blood cholesterol.
For now, however, we lack the scientific knowledge necessary to identify those at high risk for most chronic diseases. And we know that the majority of chronic diseases occur in the general population—not only in those at high risk. Our committee therefore concluded that if the population as a whole adopted our dietary recommendations, a substantial reduction in chronic disease in the United States would result. The high-risk approach—when possible—and the population approach are complementary.
We found very strong evidence that saturated fat and cholesterol intake affect the frequency of arteriosclerotic coronary artery disease. There are also excellent data showing that salt intake increases blood pressure. The evidence for dietary factors causing various cancers was weaker but still quite suggestive.
We found no data to indicate that dietary supplements such as the vitamins-minerals taken by many people has any effect on preventing disease unless prescribed for specific medical indications.
The best diet, we concluded, is one that is moderate in total and saturated fat, high in complex carbohydrates and fiber, low in added sugars, and moderate in protein, especially animal protein. In practical terms, this means most Americans should eat more fish and less meat, and select leaner meats and trim off excess fat. They should select low-fat or non-
fat milk and cheese, and eat more vegetables and fruits, especially green and yellow vegetables and citrus fruits.
Americans need to eat more cereals, breads and dried beans; use less oil and fat in cooking; and avoid fried food. They should restrict their consumption of egg yolks; salt; salty, smoked and preserved foods; and alcoholic beverages. Also, they ought to do a better job of checking labels for salt, fat and sugar content. We recommended these dietary patterns for children over the age of two as well as for adults.
To help Americans meet these goals, consumers should be provided more meaningful labels on their food purchases and restaurants should offer healthful menus. Along the same lines, nutritional education should be directed at all groups of society—not just the middle class.
The menu for a healthier lifestyle is clear. It's time more people put that knowledge into action.
March 21, 1989
Arno G. Motulsky is professor of medicine and genetics at the University of Washington.
* * *
The Spitting Image: Baseball Players and Chewing Tobacco
John C. Greene
A cheek full of chewing tobacco and a quick sideways spit. This image is as much a part of baseball as the peaked cap, the pinstriped uniform and the smack of bat against ball. The high visibility of smokeless tobacco use by the
heroes in our national sport is cause for concern because baseball players are such prominent role models.
Many of their fans are picking up the wrong message. Within the general population, a shocking 16 percent of young men between the ages of 12 and 25 have used smokeless tobacco in the past year. One-third to one-half of these are regular users, and use is increasing.
During the last two years, I have helped lead a team of researchers from the School of Dentistry at the University of California at San Francisco that has conducted an extensive study of smokeless tobacco use and its health effects on baseball players. Our team gathered data to help convince players to give up tobacco for their own health—and for the sake of their fans. We studied and provided physical examinations of more than 1,100 players from major and minor league teams, including most of the players from the Giants, Athletics, Angels, Cubs, Indians, Mariners and Brewers.
Our results so far are very disturbing. More than half the players we examined in the first year were either current or past users of smokeless tobacco. Of the 40 percent who used snuff or chewed tobacco regularly, nearly half had lesions in their mouths. Snuff is taken as a pinch and placed inside the cheek or lips; chewing tobacco is used in a bigger wad.
All of the players with lesions were offered biopsies, and these all were benign. Yet these white or yellowish patches inside the mouth can become cancerous with long-term exposure to tobacco, usually after two or three decades of regular use. Users of smokeless tobacco products also have a higher incidence of periodontal disease and gum recession. In addition, as much nicotine enters the bloodstream from using smokeless tobacco as from chain-smoking cigarettes, which leads to other medical problems. One of our research team members calculated that a third of the nicotine from snuff and half the nicotine from chewing tobacco is swallowed, which may contribute to ulcers, digestive problems and cancer of the stomach.
In short, anyone who thought smokeless tobacco products were a safe substitute for cigarettes should think again.
Many players are concerned about the image they are pro-
jecting to young people. One key player cited his responsibility as a role model as his primary motivation for quitting the habit. When he spotted a 10-year-old youngster chewing and spitting while attending one of his community lectures, he asked why he was doing such an awful thing. "Because you do," the boy replied. The player then made a deal with the youngster, promising to quit if the boy would. The boy agreed, and the player hasn't used tobacco now for six years.
A pitcher for another team hasn't managed to kick the habit yet, but he now uses snuff only in private. In public, he chews bubble gum. Yet another player had help from another quarter. His wife, who had been nagging him to give up smokeless tobacco, greeted him at the front door one day with a lighted cigar in her mouth. That did the trick; he hasn't used tobacco products since.
Similarly, many baseball teams no longer allow manufacturers of smokeless tobacco to leave free samples in locker rooms. The San Francisco Giants, for example, provide free sunflower seeds instead.
In other words, the baseball community is beginning to come to grips with its smokeless tobacco problem. However, the dilemma for players still exists. Smokeless tobacco remains a powerful tradition and many players derive satisfaction from chewing while contemplating action. They also have difficulty breaking the addictive habit.
These professional baseball players need help in quitting smokeless tobacco use, and others should be discouraged from starting. It is simply unacceptable that professional players remain more than twice as likely as other young men to use smokeless tobacco. They should be showing their young fans how to hit and pitch, not how to chew and spit.
August 13, 1989
John C. Greene is dean of the School of Dentistry at the University of California, San Francisco, and a member of the Institute of Medicine of the National Academy of Sciences.
* * *
Clearing Our Vision About Alcohol Abuse
Robert D. Sparks
You are watching television when a commercial comes on for a local hospital's alcohol treatment center. Someone in your family needs treatment, the voice says. Pick up the phone and get them into our inpatient program. They need help—and so do you.
The commercial hits home. Your sister has a drinking problem and you want to help her. But, even as you write down the phone number, you begin to have doubts. Your sister is not a chronic alcoholic who staggers down the street. No, she usually goes to work and appears fine. Isn't a residential program a bit excessive? Who will care for her family while she's gone? And how will she pay for the treatment?
These concerns are familiar to many. The loved one for whom they feel concern may be a parent, spouse, child, sibling or friend. Or, they themselves may have the problem. Seventy percent of adult Americans drink alcohol, and many of them have at least occasional problems ranging from quarrelsome behavior to drunk driving. Yet, when people seek to solve these problems, they often feel overwhelmed by the prospect of enrolling in a residential treatment center. Their problem—too much wine with dinner, too many beers on the weekend, or binge drinking—may not seem severe enough to warrant such treatment.
I chaired an expert committee that recently examined this situation for the Institute of Medicine of the National Academy of Sciences. After studying medical reports and programs across the country, we concluded that no single approach works for everybody. Some people benefit from short-term counseling while others need specialized inpatient programs. Still others may overcome their problems without any formal treatment at all.
The challenge is to match individuals with the right pro-
grams. Yet, too often, the options available are not the ones needed.
Make no mistake; specialized hospital and residential treatment programs often do an outstanding job helping people to overcome alcoholism. As valuable as they are, however, they are not required for everyone. In fact, as many as a third of the people enrolled in hospital programs could be dealt with as effectively in programs that are much less costly. The same is true of Alcoholics Anonymous. It deserves all the praise it has received for helping alcoholics remain sober. Yet AA is not ''the answer'' for everyone who ever had a problem with alcohol.
Research has shown that many persons with less severe alcohol problems could change their behavior in several ways, such as through an intensive one-time counseling session or even by watching a video on drinking problems.
Does this mean society needs fewer hospital or residential treatment programs and AA meetings? No, it needs more of them. But it also must recognize that it is missing a tremendous opportunity by failing to broaden the base of treatment options for those now unserved.
Doing so would be a blessing not only for the individuals involved, but also for society generally. There is a widespread misconception that most car crashes, broken lives and other problems from alcohol are caused by the 10 percent of the adult population with serious drinking problems. However, the 60 percent of the adult population that consumes alcohol in more moderate amounts, by the sheer weight of its numbers, accounts for more problems. Offering them appropriate counseling or treatment options would reduce this toll.
It also would be a better way to allocate resources. Hospital programs typically cost several thousand dollars. When the treatment is appropriate and effective, this money is well spent. But for every two people treated successfully in this way, someone else might respond just as well, if not better, to a less expensive method. Insurance companies, government programs and others should be more flexible in paying for these options.
Overall, adopting a more flexible approach probably would be more costly because it would make treatment available
to many more people. Yet, failing to do so is like saying that society will not treat people with broken fingers because it is too busy taking care of those with broken arms. The social cost of such a policy in terms of accidents, lost productivity and broken lives is simply unacceptable. People with alcohol problems need more treatment choices.
July 1, 1990
Robert D. Sparks is president emeritus and senior consultant with the W.K. Kellogg Foundation.
* * *
Howard W. Jones, Jr.
As the director of the nation's oldest center offering in vitro fertilization therapy for infertile couples, I weary of saying, "Sorry, your chances are not the best."
Our center has helped more than 500 couples bear one or more children. Yet an estimated 4.4 million women of childbearing age in the United States experience difficulty conceiving a child, and many of them might become pregnant with better treatment methods. Research to develop these methods is being hampered by a continuing moratorium on federal funding for human embryo and fetal research.
About 200 centers in the United States offer in vitro fertilization and embryo transfers to infertile couples. But success rates have been low, as was reported recently by a committee of the Institute of Medicine and the National Research Council on which I served. The procedure is appropriate as initial treatment for 10 percent to 15 percent of infertile couples, and in 1988 it resulted in a live birth only 12 percent of the time.
Clinics like ours will never be able to help every infertile couple, but we could improve the odds considerably by learning more about the biology of human reproduction. Just as a brain surgeon studies brains, researchers in our field need to study sperm, eggs, pre-embryos, embryos and fetal tissue. Computer modeling, studies of animals and other methods can answer some of our questions, but we must investigate early prenatal development in humans directly if we are to truly understand it and learn why it does not always occur as desired.
Such research raises many difficult ethical and social issues, such as whether embryos are human, how "spare" embryos from fertilization procedures should be handled and whether fetal tissue from abortions ought to be used in research. These questions are profound, and they have been addressed in more than 85 reports from 25 countries.
Many nations have been more successful than ours in confronting the conflicting viewpoints and agreeing how to proceed. Americans have tended to polarize these issues rather than doing the hard work of devising research guidelines acceptable to most people. The Bush administration announced not long ago that it would continue to deny federal funding for human embryo and fetal research, and some members of Congress now are questioning the legality of the decision. Such wrangling needs to be replaced by people coming together in a less politicized manner, both within and outside government, to search for some consensus. As matters now stand, researchers are leaving the field and millions of infertile couples are frustrated.
From a scientific standpoint, the research opportunities have never been brighter. Our committee identified dozens of promising avenues of study, from the development of sperm and eggs to the process of embryo implantation. It listed several current clinical practices, such as stimulating and obtaining eggs from in vitro patients, that might be improved at clinical centers. It also called for closer investigation of technological advances such as freezing eggs and embryos.
A related scientific field illustrates what might be achieved with increased research on human reproduction. During the past two decades, scientists have made tremendous ad-
vances in techniques to assist conception in animals through in vitro fertilization and embryo transfer. Sharp ethical and scientific differences between humans and other animals prevent these methods from being applied directly to the treatment of human infertility. Yet the progress achieved with animals is pertinent, and it contrasts with the slow pace of research on human reproduction. Increasing this pace could lead not only to better treatment of infertility, but also to improvements in contraception, food production and efforts to sustain endangered species.
It is entirely proper in a democratic country that ethical questions involving science are resolved by society as a whole rather than by scientists alone. But, as one who confronts the heartbreak of infertility every day, I wish more Americans would recognize what we could accomplish by pursuing this research. Instead of saying "I'm sorry" so often, people like me could say, "Congratulations! You're pregnant."
June 3, 1990
Howard W. Jones, Jr., is professor of obstetrics and gynecology at Eastern Virginia Medical School and director of the Jones Institute for Reproductive Medicine, in Norfolk.
* * *
Who Is Going to Deliver Baby?
Roger J. Bulger
Having a baby should be one of life's most memorable experiences. For many American women, it's just that—but for the wrong reasons.
Some have a baby by Cesarean section but wonder later whether the procedure was really necessary. Others discover that their gynecologist has stopped delivering babies
because of the fear of being sued and skyrocketing malpractice insurance rates. Still others lack health insurance and may have difficulty getting care at all.
Our country's obstetrical care system has delivered millions of healthy babies and saved countless young lives with its medical miracles. Yet it faces increasing problems, one of the most serious of which is the continuing controversy over medical professional liability.
Nowhere has the malpractice issue hit harder than in maternity care. Seventy percent of obstetricians-gynecologists reported in a 1987 survey that at least one claim had been filed against them. They were more than twice as likely as other physicians to be sued; their annual malpractice insurance premiums in some cities now exceed $100,000.
A committee of the Institute of Medicine of the National Academy of Sciences concluded in a study released this past week that the malpractice situation helps explain why so many women across the country, especially those from rural areas and inner cities, cannot find anyone to deliver their babies.
Obstetricians generally practice in metropolitan areas; in rural areas, most babies are delivered by family physicians. About 2,500 certified nurse-midwives also practice across the country. Significant numbers in each of these groups have eliminated or limited their obstetrical practice because of liability concerns.
The trend has made it difficult even for some women with adequate health insurance to obtain obstetric care. Low-income women, however, have been hurt the worst. Not only do they generally lack health insurance, but they also are more likely to smoke, have poor diets and otherwise be at risk of experiencing complications. Many obstetrical providers have cut back on services to these high-risk women and reduced their Medicaid caseloads because they fear being sued for any bad outcomes, even those that are unavoidable.
The day-to-day practice of obstetrics also has changed. Some of these changes have been beneficial, but others are more questionable. Notably, there is considerable evidence that routine use of electronic fetal monitoring in normal pregnancies is both expensive and unnecessary, occasionally leading to inappropriate Cesarean sections. When a defec-
tive baby is born, a physician is far less likely to be sued for performing an unnecessary C-section than for not doing one. Obstetric technologies, such as fetal monitors, should be assessed more systematically before they become accepted as standard practice.
Whether high malpractice insurance premiums constitute a real economic burden for obstetrical providers is open to debate. But what is certain is that the premiums will remain high so long as obstetrical disputes are settled through our existing legal structure. The tort system now in place has resulted in increased costs, reduced availability of services and an erosion in trust between doctor and patient.
Several states have begun experimenting with alternatives to tort law, such as by providing no-fault compensation for certain birth defects. These experiments offer one way of easing the malpractice dilemma in obstetrics; similar efforts are needed around the country.
The most immediate need of all, however, is to provide uninsured women with greater access to maternity services. Extending additional immunity to physicians at government-financed clinics, contributing to liability coverage for Medicaid providers, and expanding the National Health Service Corps are possible ways to encourage doctors to provide obstetrical services.
The current situation in obstetrics cannot be allowed to persist until the malpractice issue is resolved for medicine generally, a process that could take many years. Too many women and babies are not being served. States need to address immediately the deterioration in maternity services for the poor that have been worsened by liability concerns.
Having a baby should be memorable for the right reasons. It's time we gave birth to a maternity-care system that makes this possible for many more American women.
October 15, 1989
Roger J. Bulger, president of the Association of Academic Health Centers, chaired a committee of the Institute of Medicine of the National Academy of Sciences that studied medical professional liability and the delivery of obstetrical care.
* * *
Accidents Are Not Always Accidental
Susan S. Gallagher
They are the deaths that fill our local evening news: A family suffocated in a fire. A child drowned in a neighbor's swimming pool. The teenager who shot himself. The young mother crushed in a car accident.
Events like these are so common they appear inevitable. But suppose the mother's car had been equipped with an airbag? What if the teenager's parents had locked up their gun, the pool had been surrounded by a locked fence, and the family had installed a smoke detector?
Every year about 150,000 people in the United States die of injuries. One in four Americans is injured annually. This is a staggering toll, yet accidents often are perceived—incorrectly—as random events that cannot be anticipated or prevented. Case in point: After the recent series of crashes, deaths and injuries on military ships, headlines read "Navy Calls Accidents String of Bad Luck."
Injuries are so common that they lack the aura of scientific mystique associated with cancer or AIDS. When we hear about someone receiving an organ transplant, we are fascinated by the patient's story. Rarely do we ask whether the death of the organ donor was due to injury and might have been prevented.
Similarly, when people are severely injured, their families usually focus on the need for rehabilitation and long-term care—not on preventing similar accidents. Contrast that with the tradition of supporting medical research on heart disease, cancer or whatever disease killed a loved one. Most important, a life lost to injury is almost always obvious and mourned. But it is rarely evident—and celebrated—when a life is saved through prevention efforts.
Dramatic progress has been achieved during the past three decades in our ability to prevent injuries and mitigate their
impact. Experts in epidemiology, medicine, engineering, biomechanics and other disciplines have shown how to save tens of thousands of lives each year. But neither scientists nor the larger community have mustered the political will to apply this knowledge and implement proven strategies.
Congress, responding to a report of the National Research Council and Institute of Medicine on this needless suffering, established a unit at the Centers for Disease Control to oversee research, practice and training on injury control. Last year it more than doubled the program's budget to $21 million. These steps deserve applause but must be expanded drastically to tackle a problem costing society $158 billion a year.
A decade ago 26 states rescinded laws requiring motorcyclists to wear helmets; motorcycle fatalities increased 40 percent. Airbag technology has been available for 20 years; only now are manufacturers introducing airbags in automobiles. Technology also exists to make self-extinguishing cigarettes, which could prevent an estimated 2,000 deaths and 6,000 burn injuries annually. But legislation requiring manufacturers to produce them has stalled. Even simple and inexpensive measures such as smoke detectors have not been universally implemented.
Some existing safety efforts, meanwhile, have not been proven effective. Research suggests that traditional driver education programs have not been as useful as portrayed; some alternatives, such as graduated licensure of beginning drivers, should be tried. Better programs also are needed to reduce cases of lead poisoning in children and to make amateur competitive sports safer.
Injuries never will be eliminated. But consumer products and physical environments could be made less hazardous, such as with designated traffic lanes for bicyclists or with detachable bases to reduce baseball sliding injuries. Stronger safety standards in automobiles could be enacted and existing ones enforced more vigorously. More could be done to change social norms involving dangerous behaviors, as has been done with drunk driving. Litigation can help pressure manufacturers to stop making unsafe products. Most needed are new injury control programs on the local level, training
programs for professionals, and better data systems to identify where the problems are.
In other words, we need to stop accepting injuries as a fact of life—and death. As former Surgeon General C. Everett Koop pointed out, "If some infectious disease came along that affected one out of every four children in the United States, there would be a huge outcry and we would be told to spare no expense to find the cure—and to be quick about it."
August 19, 1990
Susan S. Gallagher is a senior scientist at Education Development Center Inc. in Newton, Mass. This article is adapted from a longer version she co-authored with Ilana Lescohier and Bernard Guyer in Issues in Science and Technology.
* * *
Changing Behavior to Limit the Spread of AIDS
Heather Miller and Marshall Becker
Anyone who has tried to stop smoking knows it is not easy to change a habit that is pleasurable but dangerous. Yet some people do give up smoking and other behaviors involving fundamental urges, compulsions and addictions. Their success deserves more attention as the AIDS epidemic continues to claim lives.
Theoretically, the spread of the HIV virus that causes AIDS could be virtually stopped by altering the behaviors known to transmit it: unsafe sex and intravenous drug use. Over the past several decades, social and behavioral scientists have accumulated valuable knowledge about modifying unhealthy
behaviors. Their approaches and insights ought to be applied to altering AlDS-related behaviors, as appropriate.
All too often, however, drug use and unsafe sex are portrayed in moralistic terms. People are told to ''just say no,'' even after it's clear they are unwilling or unable to do so. Clinging to such a simplistic approach in the face of deadly consequences and the availability of more effective strategies exhibits moral mean-spiritedness.
One lesson from the behavioral and social sciences that might be applied to the AIDS epidemic is that many people respond more effectively to a choice of options than to a single, dictated course of action. The evidence also shows that most individuals are better at modifying selected behaviors than at changing their entire lifestyle.
For example, many dieters find it easier to reduce the size of portions than to give up certain foods altogether. Similarly, some intravenous drug users who cannot obtain or accept treatment immediately, for whatever reason, may find it easier to stop sharing drug paraphernalia or to sterilize injection equipment than to practice abstinence.
To be effective, behavioral intervention programs need to consider people's health beliefs and perceptions of their own efficacy. The programs should incorporate knowledge about how people make decisions and why they take risks. Efforts to promote behavioral change also should be based on current theories of relapse prevention, social norms, community organization and environmental restructuring.
Studies have shown the importance of providing accurate information in language that the intended audience can understand, delivered through the mass media and other routes that take reading skills into account. Messages should be realistic, based neither on moral judgments nor on excessive levels of fear. In the case of AIDS, this means overcoming any reluctance to discuss sexual behavior in a clear, explicit and comprehensible fashion.
Because our society is diverse, interventions that are effective for one group may not work with others. Therefore, multiple strategies are needed to help different groups, and they must be repeated often—both to remind people about the risk of AIDS and to reinforce desired behavioral changes.
Regardless of what we do in the future, many people already are involved in high-risk behaviors, and some sexual and drug experimentation is likely to continue. Data indicate that the majority of Americans become sexually active in their teens. Rather than engaging in wishful thinking, we must find the courage to heed these data and begin the difficult process of discussing sex and drug use with adolescents in a manner that will help them act to protect themselves. Any program that fails to go beyond exhortation and come to grips with this need consigns millions of Americans to possible death.
Similarly, we must be prepared to explore innovative strategies, such as needle-exchange programs, to save the lives of those who are not yet ready to elect—or who cannot obtain—more traditional treatment for their drug problem.
We are eight years into the epidemic and there is no indication that an AIDS vaccine or cure is near. Given the dire consequences of this disease, public officials, health professionals and others must begin making better use of findings from the behavioral sciences. They also must make a substantial commitment to research efforts that will expand this understanding rapidly.
We have identified specific behaviors that can transmit AIDS. It's time we put to use our knowledge of what works best in facilitating and sustaining changes in these risky behaviors.
November 12, 1989
Heather Miller is a study director of the National Research Council's Committee on AIDS Research and the Behavioral, Social and Statistical Sciences. Marshall Becker, professor of health behavior and associate dean of the School of Public Health at the University of Michigan, serves on the committee.
* * *
The Dilemma of AIDS Drug Experiments
Robin Weiss and Theodore Cooper
One of the most difficult aspects of the AIDS epidemic for medical researchers is having to turn down requests from AIDS patients who volunteer to participate in experiments of new drugs.
From the researcher's perspective, the applicant may fail to meet certain criteria, such as a prescribed set of symptoms, and therefore be ineligible. Yet turning down such requests from dying patients may appear cold and insensitive. After all, if the patients are likely to die, why not let them at least try the drug? It could help them and, if nothing else, it will give them hope. Even if the drug fails, its use may produce knowledge to help others. What, then, is the harm in letting people participate?
As we learned during our recent study of AIDS and HIV infection for the Institute of Medicine and the National Academy of Sciences, medical researchers, physicians, AIDS patients and others have been struggling with the ethics of this difficult situation. Our committee concluded that the most compassionate course is for drug trials to continue to follow rigorous scientific guidelines.
Experiments to compare one treatment with another in humans are called clinical trials. Intuitively, it seems a simple matter to administer a drug to a patient and then see if the patient improves. However, unless the drug has dramatic results, it can be difficult to tell how it has affected the person's condition when the illness itself has spontaneous ups and downs. If two different drugs are given to two people with the same disease, differences in outcome may be due more to differences between the two people than between the two drugs.
Over the years, scientists have sought to diminish these kinds of ambiguities by developing strict guidelines for drug experiments. When comparing two or more drugs, for ex-
ample, scientists take care to follow strict eligibility criteria when they recruit volunteers, assigning patients randomly to different groups. The groups are made as similar as possible, but they receive different treatments. In many drug trials, the control group receives a placebo. For most AIDS drug trials involving persons with symptoms, however, the control group should receive the drug AZT, which has proven effective. This way everyone has a chance to benefit from therapy.
Any difference between the groups may affect the validity of the results. For instance, if the patients in one group happen to have less severe illness at the beginning of the experiment, then this, rather than the experimental drug, may account for an improved outcome at the end of the test.
Another rule of clinical trials is to try to use the right number of test subjects. This is done by calculating the smallest number of people needed to show a statistically significant difference between the experimental groups. The goal is to avoid exposing more people than necessary to a potentially toxic drug and to complete the study in the shortest possible time.
This leads to one of the most crucial points about clinical trials, which is that they are experiments, not treatment. Three things can happen when testing a new drug: it can make no difference in the patient's health; it can be beneficial; or it can make the patient worse by accelerating disease or exerting toxic effects that make the patient sicker. In the case of AIDS drugs, it is frequently overlooked that people can be made sicker by experimental therapy. This is what happened with the drug suramin, which looked promising but proved in clinical trials to have serious toxicity and no clear benefit.
Allowing wide distribution of unproven therapies or relaxing entry criteria for AIDS drug trials, therefore, is not a benign action. Trials that are larger than necessary or sloppy in design and execution will delay the effort to find meaningful therapy for HIV infection and AIDS.
A number of researchers have been studying alternatives to traditional randomized trials, and their work warrants close monitoring. The suffering caused by AIDS has been so severe that innovations must be pursued vigorously and any
effective new drugs made available as widely and quickly as possible, perhaps through treatment demonstration projects. The federal government needs to provide more funding to support clinical trials, and it should work with industry to develop better systems for gathering, analyzing and sharing test results.
Our committee also called on researchers to make a greater effort to extend their trials to a wider variety of new drugs and to such untapped populations as women, IV drug abusers and pediatric patients.
In the end, however, the first rule of these experiments must continue to be that they are sound scientifically. Doing so is actually the most humane—if often difficult—course of action if we are to bring this terrible epidemic under control as soon as possible.
October 4, 1988
Robin Weiss, director of AIDS activities at the Institute of Medicine, was study director of a committee of the Institute and the National Academy of Sciences that examined the AIDS crisis. Theodore Cooper , chairman and chief executive officer of The Upjohn Company, chaired the committee.
* * *
Identifying What Works in Medicine
Samuel O. Thier
The U.S. national health care bill exceeds $600 billion annually. But many are questioning whether the care—including diagnostic tests, surgical procedures, drug therapies and other medical interventions—always is truly effective.
Consider hip fractures, which an Institute of Medicine committee recently examined. About 260,000 Americans,
mostly elderly, suffer hip fractures each year. As many as half of them require care for the rest of their lives, at great expense to their families and society. Direct medical care costs total about $6 billion per year. Yet no clear consensus exists among physicians about certain aspects of the treatment of hip fractures, such as how long patients should be hospitalized, which surgical options to use, or the most effective sequence of surgical interventions.
Vast anecdotal evidence exists about dealing with hip fractures, but the plural of anecdotes is not data. A more rigorous evaluation might analyze how patients have responded to different kinds of treatment, the number of days they were hospitalized, mortality rates, costs and a host of other questions. Focusing payment on the most effective approaches might reduce costs. Cutting the bill by even a small fraction could yield substantial savings, perhaps freeing up resources for other needs such as prenatal care, drug treatment, basic care for the uninsured or AIDS. Even if no money were saved, care would be improved.
So, too, across the spectrum of medical care. Studies have called into question whether coronary bypass operations, Cesarean sections and other common procedures are overused. Some procedures that are useful in specific situations have become accepted across the board—for example, the use of electronic fetal monitoring in obstetrics. Although the value of this technology in normal pregnancies is unproven—and perhaps even negative—clinicians feel pressure to apply it generally. They risk being accused of negligence if they fail to use the devices and something goes wrong.
Too often, care is unnecessary or inappropriate, or appropriate care and preventive services are not provided. Accepted medical practices vary from city to city. Even within a hospital, one physician may be much more likely than another to order a costly test or procedure even though evidence of its effectiveness is lacking. Unnecessary expenditures are rarely the result of deliberate fraud, and professional autonomy is important for physicians. But as the cost and complexity of choice in any clinical circumstance continue to expand, physicians need to know what works.
This presents both a challenge and an opportunity for the medical community and society generally. Our institute
has been studying medical practices for treating hip fractures, breast cancer and heart disorders, and others are looking at chronic pain, bedsores, cataracts, depression, sickle-cell disease, prostate enlargement and other conditions. Yet even with these studies and increased interest among federal officials, the knowledge base is likely to remain meager for years to come. And it will not be easy to translate findings into actual changes in the day-to-day practice of medicine.
Evaluating medical procedures more systematically is one of the best—and most overlooked—ways of improving health care. In recent years, our country's familiar health care system has been swept by a host of changes, from proliferating health maintenance organizations to hospital emergency rooms overwhelmed by uninsured patients. Some valuable changes have improved quality and efficiency while controlling costs. Yet many efforts have focused on either cutting benefits or limiting patient access to the system, raising troubling questions of equity and quality.
Studying the effectiveness of our actions offers a promising alternative for controlling costs while, at the same time, improving care. Medical professionals should participate actively in this process, viewing it as a way to improve their craft rather than as an intrusion on their prerogatives. Medicine is a learned profession, and one of the responsibilities ceded by society to any learned profession is to set and enforce standards. Although the diversity of human beings makes it impossible to create a single rule book for clinical practice, more can be done to separate the accepted from the truly effective.
If we physicians do not take the lead in doing this, then others with less training will do it for us. As health care costs approach $2 billion a day, Americans need to know that everything being done in their hospitals and doctors' offices is truly worthwhile.
October 28, 1990
Samuel O. Thier is president of the Institute of Medicine of the National Academy of Sciences.
* * *