Science on Trial: The Clash of Medical Evidence and the Law in the Breast Implant Case
by Marcia Angell, M.D., copyright 1996 by Marcia Angell, to be published June 1996 by W.W. Norton & Co. (London, New York).


A few years ago I would have been surprised to be told that I would one day write a book about breast implants. I had no particular interest in the subject, and it would have seemed unlikely that I would develop one. To be sure, as a physician I was aware that cosmetic surgery was a part of health care. But I considered it a relatively minor part, dwarfed by the many more crucial matters facing doctors and patients -- life-threatening diseases, unremitting pain, profound disability. Nor would my responsiblities as executive editor of the New England Journal of Medicine have been likely to involve me int he study of breast implants. My job is to help evaluate the many reports of medical research sumitted to the Journal, of which only about 10 percent are selected for publication. Reports of breast implant research were rarely among the submissions and, until recently, never among the papers selected for publication. Not very much research was being done on the subject, and what there was wasn't very good. So how did the idea for this book come to be?

The short answer is that I was drawn to the issue by two manuscripts submitted to the Journal, one in 1992 and the second in 1994. In 1992 David Kessler, the commissioner of the FDA, banned silicone-gel-filled breast implants from the market because they had not been proven safe. The ban caused widespread alarm among the 1 million to 2 million women who already had implants. It also led to a torrent of lawsuits against the manufacturers of breast implants. Shortly after Kessler announced the ban, he submitted a manuscript to the New England Journal of Medicine explaining his reasons. In reviewing the manuscript I began to realize that the matter was more complicated than I had suspected. Kessler's decision involved not just science, but public opinion, politics, and the law. I thought the consequences of the FDA decision would be far more wide-ranging than Kessler seemed to believe, and I wrote an editorial to that effect to accompany his article. That was my first exposure to the subject.

Two years later, in 1994, Dr. Sherine Gabriel of the Mayo Clinic submitted to the New England Journal of Medicine her report of the first epidemiologic study of whether breast implants increase the risk of certain diseases and symptoms. By this time, thousands of lawsuits had been filed alleging that they did. Some women received huge damages from sympathetic juries (the record was $25 million). For every jury verdict, many more cases were settled for the plaintiff out of court. The situation became unsustainable for the breast implant manufacturers, and so they agreed to set aside $4.25 billion to meet the claims of all women with breast implants once and for all -- the biggest class-action settlement in history. Yet the Mayo Clinic study, published shortly after the class-action settlement was announced, did not show a link between breast implants and disease. I was struck by the discrepancy between the legal findings and the scientific evidence. Why were the courts so sure, when the scientists were not at all certain? I suggested some possible answers to this question in another editorial, which accompanied the Mayo Clinic study. That was my second foray into the subject, and by this time I was hooked.

But the most important reason I decided to write about the breast implant controversy is because it illuminates important themes in American life. It richly illustrates the interplay of regulation, litigation, commercial interests, media coverage, and consumer pressure in the public confrontation of a major health issue. In the decade leading up to the FDA ban, the safety of implants was debated in the courts, in medical journals, and in the popular media. The debate occurred against a political backdrop of controversy about the relative excesses of big business and government regulation. Were large corporations like Dow Corning, the leading manufacturer of breast implants, knowingly foisting off dangerous products on the public or were they responsibly meeting consumer demands? Should the FDA have moved more promptly to pull implants off the market or should women have been allowed to make their own choices? After the ban, the focus shifted to the courts, where matters of scientific fact were decided not by scientists, but by lawyers, juries, and judges. On what basis did they reach their conclusions? And what effect did public opinion have on the outcome? The media covered the controversy fully, if not always accurately, and the public took a lively interest in the matter. Did public opinion drive the FDA's actions and the courts' conclusions? And how was the unfolding of the breast implant story affected by greed and the expectation of personal profit? Vast amounts of money, in addition to the $4.25 billion in the class-action settlement, are still at stake in the thousands of individual lawsuits that have been filed and will be filed.

What fascinated me most about the contradictory opinions and accusations, the frenetic legal activity, and the huge sums of money at stake was the question of evidence. As I investigated the subject, I realized that the breast implant controversy is simply one example of the difficulty we Americans have in dealing with scientific evidence, particularly on matters of health. We depend greatly on science and its technological fruits, we like to talk about what research shows and what it doesn't, we think we understand risks, but when it comes to the recurrent medical scares that sweep across the land like locusts, all our sophistication goes out the window. Just give us the conclusion, tell us whom to blame, and don't bother with the evidence.

In this book, I will use the breast implant story to illustrate the broader themes that concern me. How do scientists reach conclusions about questions of health risks, such as whether breast implants increase the chance of developing a disease? How does the legal system reach its conclusions about the same question? Do the differences in methods account for the vast differences in outcome? What are the essential distinctions in the way science, the law, and the public regard evidence, and what are the consequences for our society?

I will begin with a brief review of the breast implant controversy to provide the narrative. To set the context further, I will describe in some detail the implants themselves, as well as the history of breast augmentation and the more recent use of implants for reconstruction after cancer surgery. In the remaining chapters, I will widen the focus to the themes illustrated by the breast implant story: the role of regulation and litigation in American life, the nature of scientific evidence, the way in which legal evidence is different, the distorting effect of greed and corruption on the usual process for deciding medical matters, the public's all-or-nothing response to health news, and the ripple effects of a paroxysm like the breast implant controversy.

Finally, a brief personal note to give the reader some idea of the point of view I bring to this subject. I consider myself a feminist, by which I mean that I believe that women should have political, economic, and social rights equal to those of men. As such, I am alert to discriminatory practices against women, which some feminists believe lie at the heart of the breast implant controversy. I am also a liberal Democrat. I believe that an unbridled free market leads to abuses and injustices and that government and the law need to play an active role in preventing them. Because of this view, I am quick to see the iniquities of large corporations. I disclose my political philosophy here, because it did not serve me well in examining the breast implant controversy. The facts were simply not as I expected they would be. But my most fundamental belief is that one should follow the evidence wherever it leads. From time to time, it is important to look up and see where you have been led and who is there with you, but you should not turn back. That's what this book is all about.

Marcia Angell
The Breast Implant Story: Once Over Lightly

By the time FDA commissioner David Kessler decided in 1992 to ban silicone-gel-filled breast implants, an estimated 1 million to 2 million American women already had them.1 (The precise number is unknown.) According to Kessler, the manufacturers had not fulfilled their responsibility to demonstrate the safety of the implants, and he therefore had no choice but to take them off the market. Thus ended 30 years of the easy availability of breast implants (for those who could afford the surgery). Most women with implants had simply wanted to enlarge their breasts, but about 20 percent had obtained them for reconstruction after mastectomy for breast cancer.2 A 1990 survey commissioned by the American Society of Plastic and Reconstructive Surgeons found that over 90 percent of women with breast implants were well pleased with the results. Nevertheless, many people were jubilant about the FDA ban -- including advocates of tough government regulation, women who believed breast implants had caused them to become ill, and feminists who thought it was about time someone put a stop to women being pressured to conform to male fantasies. What follows is the story of the breast implant controversy. It is an astonishing story, with implications far beyond the question of whether or not silicone-gel-filled breast implants are safe. As executive editor of the New England Journal of Medicine and as a woman, I was aware that breast implants were controversial on many counts, but I had not given the matter much thought until Dr. Kessler submitted an article to the Journal explaining his reasons for banning the devices. The article was important and we were happy to publish it.3 Still, I was troubled by the likely consequences of Kessler's action, as well as by some of his arguments. He seemed disdainful of women who wanted breast implants for purely cosmetic reasons, and his decision, though welcomed by many women, struck me as a little patronizing. More important, how would the sudden ban strike the million or so women who already had implants? Would they accept Kessler's legalistic argument that he was simply responding to the lack of evidence adduced by the manufacturers? I thought not. Far more likely, they would see the FDA ban as proof that the implants were extremely dangerous.

And that is exactly what happened. Despite Kessler's weak assurances that removal of the implants was unnecessary, women rushed to have them taken out, sometimes by the same plastic surgeons who had implanted them in the first place. One woman who could not afford the fee attempted to remove her own implants with a razor blade; she survived the ordeal, and was pictured in newspapers looking greatly relieved after a surgeon had finished the job.4 Most notably, the small trickle of high-stakes court cases that began in the early 1980s swelled to a torrent.5 Compassionate juries awarded multimillion-dollar damages to women with a variety of complaints that they and their lawyers attributed to breast implants.6 If Kessler thought the ban was the end of the matter, he was wrong.

What did we know about the safety of breast implants? Kessler was right about this: at the time the FDA made its decision to ban them, we knew next to nothing. Incredibly, there had been no systematic studies of the effects of breast implants. We knew, of course, that there could be complications from the surgery itself (as with any surgery), including infections and hemorrhage. We also knew that in many women the tissue around the implants scarred excessively, distorting and hardening the breasts, and that these contractures could be uncomfortable. And finally, we knew that in a significant percentage of women (the best estimates are around 5 percent) an implant ruptured, releasing silicone gel into the surrounding tissues and flattening the breast.7 But these local complications, unpleasant as they were, were not the basis for most of the alarm about breast implants, nor were they the focus of the multimillion-dollar lawsuits.

Instead, a growing number of Americans had come to believe that breast implants could cause devastating effects on the rest of the body. In particular, silicone-gel-filled implants were said to be responsible for a constellation of disorders known as connective tissue diseases. These diseases -- which include systemic lupus erythematosus (SLE), rheumatoid arthritis, scleroderma, and polymyalgia, among others -- are thought to involve a disturbance in the immune system that turns the body's protective defenses against itself. The result is an autoimmune disease -- that is, a prolonged civil war within the body that can produce profound weakness and fatigue, along with variable damage to the joints, skin, and internal organs. It was theorized that silicone, leaking slowly from the implants, provokes an immune reaction that then somehow turns into an autoimmune process. This theory, not the local complications, was the basis for most of the alarm and the largest lawsuits. Earlier worries that implants could cause cancer were largely eclipsed by the concern about connective tissue disease, in part because scientific studies were showing the cancer scare to be unfounded.8 It is harder to study connective tissue disease than cancer, because it is more difficult to pin down or rule out the diagnosis.

Eventually, the breast implant manufacturers, losing one product liability case after another, agreed to the largest class-action settlement in the annals of American law.9 Class actions are lawsuits in which the claims of many people are decided in a single court proceeding brought by representative plaintiffs. Through a class-action settlement, defendants can limit their losses by paying a set amount to be distributed among all those with a claim. In April 1994 the major manufacturers agreed to pay $4.25 billion to women with breast implants, a billion of which was explicitly set aside for the lawyers involved. Although nearly any woman with implants would be entitled to something under the terms of the settlement, women were permitted to opt out if they thought they could do better in court on their own. By June 1, 1995, 440,000 women had registered to participate in the class-action settlement (as much as a third of all women with breast implants in this country), but, with vigorous encouragement from plaintiffs' attorneys, about 15,000 (half Americans, half foreigners) had opted out to seek higher damages individually.

The implant manufacturers could have warned of a link between breast implants and connective tissue disease in the informational pamphlet they are required to put in each package of implants. If they had done so, they would have had less trouble. In addition to dissuading many women from considering implants, such a warning would have disarmed those who got implants anyway, since they could no longer have claimed that the manufacturers had misled them. A product liability suit is less likely to be successful if the buyer (in this case, the plastic surgeon, who is expected to inform the patient) has been warned. But the manufacturers steadfastly maintained that the implants were safe, despite the lack of scientific evidence in either direction. To be sure, there were many individual stories of connective tissue disease developing after the placement of breast implants, but these reports alone ("anecdotes," in scientific jargon) do not constitute evidence that the implants caused the disease. They could well have represented pure coincidence. Since connective tissue disease can develop in women with or without implants, the only way to demonstrate that implants actually cause the disease is to show that the risk is significantly higher in women with breast implants than it is in those without implants. To do so requires epidemiologic studies -- scientific surveys of the incidence of disease in samples of different groups. But it was not until a few months after the class-action settlement was announced that the first such study of breast implants and connective tissue disease was published.10

How could the law have been so far out in front of the evidence that huge amounts of money were changing hands and $4.25 billion more was promised? And whose hands were they? Many of those who welcomed the FDA ban believed that the law was not in front of the evidence, but that the evidence had been largely suppressed by rapacious implant manufacturers interested only in their profits and by the plastic surgeons who made a very good living implanting the devices. However, not even this conspiratorial theory does justice to the real complexity of the story. As suspected, the story is full of greed, but this greed is hardly limited to implant manufacturers and plastic surgeons. Plaintiffs' attorneys, doctors, researchers, and even implant recipients themselves are exploiting the lucrative opportunities. Four billion dollars is a lot of money, and there is even more to be had.

The breast implant story shows us some of the worst in human nature; it also reveals the weaknesses of a number of important U.S. institutions, including the media, segments of the legal and medical professions, and the courts. It raises the question of how we know what we think we know about the causes of disease, and, in so doing, it calls into question the systems we have devised for finding out. How does the public decide whether silicone breast implants or any other product is harmful? How do researchers and doctors decide, as compared with judges and juries? These issues can be considered within six broad themes that pervade and shape American society. All of the themes are clearly sounded in the breast implant controversy.

The first concerns the place of regulation in American life. The FDA, as perhaps the most visible of the governmental regulatory bodies, faced a virtually impossible dilemma in deciding whether or not to ban breast implants. Whatever its decision, the FDA would be strongly criticized. If implants were left on the market, the FDA would be seen as caving in to the manufacturers. If they were banned, at least a million women might wrongly conclude that they were at risk for serious disease. The letter of the law requires that manufacturers of devices demonstrate their safety before they market them. But breast implants had been on the market for a dozen years or so before that law was passed and another dozen afterward before the FDA decided to ask for evidence of safety. The FDA's decision to call for evidence after all that time is open to a number of political interpretations. Some believe that the manufacturers had a free ride during most of the 1980s because of the anti-regulatory philosophy of the Reagan administration. On the other hand, it is hard to escape the conclusion that the change in the FDA's attitude toward the end of the decade might also have reflected the mounting public concern about the issue, which in turn was largely driven by the publicity surrounding a few spectacular court cases. To the extent that this latter interpretation is correct, the FDA responded like a political body, not a regulatory body.

In addition, after David Kessler came to the FDA in 1991, the controversy took on the character of a battle of wills. Kessler clearly believed in a strengthened FDA. The manufacturers, who had not performed the necessary studies to demonstrate safety, were cavalier in their response to the FDA's new hard line. Who would back down? Meanwhile, the public pressure to do something about breast implants was intensifying. Panels convened by the FDA to consider the matter included vocal partisans on the issue. Hearings were public and included emotional testimony about the harm caused by implants, but little critical analysis. The FDA's action, then, was taken in a charged atmosphere of advocacy and in the virtual absence of evidence. The ban raises the question of the degree to which regulatory decisions should be influenced by political and social considerations. It also raises the issue of the proper balance between regulatory protection of consumers and individual choice. In this case, the letter of the law, buttressed by public opinion, dictated one course, but consideration of the likely social consequences engendered bythe widespread alarm might have supported another.

The second theme illuminated by the breast implant story is the impact of tort law on American life. A tort is a wrongful act or injury for which damages are sought in a civil (as opposed to criminal) court. Americans increasingly turn to the courts to remedy a whole variety of complaints, and the number of tort cases -- particularly product liability and malpractice cases -- is growing accordingly. The breast implant case illustrates in the extreme the influence of the tort system. Even before the ban, there had been a steady trickle of lawsuits alleging that breast implants caused connective tissue or autoimmune disease. Successful lawsuits bred more lawsuits, which in turn bred publicity. Not surprisingly, as more women with implants and their doctors became aware of the possible risks, there were more reports of disease caused by implants. As the chairman of one of the FDA panels said, "The lawyers were ahead of the doctors, and the public was ahead of the FDA."11

After the ban, the plaintiffs' bar moved quickly, and many thousands of lawsuits were filed within a matter of months. The result has been an immense transfer of money from the breast implant manufacturers to plaintiffs and their lawyers. There has been little attention to the fact that the money originates from consumers who purchase the manufacturers' other products. Nor has there been much attention to other consequences of the lawsuits, although they are likely to be far-reaching. In particular, manufacturers of other medical devices and the suppliers of their raw materials are threatening to pull out of the market because of the legal liabilities. What accounts for the torrent of successful litigation, with all its ripple effects, in the absence of solid scientific evidence about the risks of breast implants? Several peculiarities of the American tort system, which I will discuss in later chapters, bear on this question. Tort reform is now a contentious political issue, and the breast implant controversy throws the debate into sharp relief. It well exemplifies the tension between the pressure to curb the excesses of the tort system and the need to maintain reasonable access to the courts to seek redress for injury.

The third theme is the marginal and ambiguous role of scientific evidence in our society. To be sure, the United States has a large scientific enterprise, and its research findings are often widely publicized. But interest in scientific conclusions is not the same thing as interest in how they originated. The nature of evidence is simply not a front-burner item for most Americans, a fact illustrated by the breast implant story. As it happened, medical researchers did not systematically begin to collect evidence on breast implants until around the time of the FDA ban, when several large studies were initiated. These were epidemiologic studies that sought to determine whether the diseases said to be caused by breast implants are more common in women with implants than in women without implants. Unless this question is answered, it is impossible to say whether breast implants contribute to the diseases. Other types of studies, including animal studies, laboratory studies, and case reports, can raise interesting questions and answer some, but they cannot answer this one.

So far, none of the epidemiologic studies has been able to demonstrate a clear link between breast implants and connective tissue disease or suggestive symptoms. This does not mean that there cannot be a link, just that it is too small to have been detected by the studies that have been done. Possibly, much larger studies will show some risk. The answer to the question will come in incremental steps, one study at a time, and represent the accumulated weight of evidence from many sources. This is the way medical research works; evidence is accumulated slowly and the conclusion is inseparable from the evidence. The lack of understanding about how scientific conclusions are reached causes a great deal of confusion in American society, as I will discuss in a later chapter.

The fourth theme concerns science in the courtroom. When a trial involves a matter of scientific fact, such as whether breast implants cause disease, the approach to answering the question is very different in court than it is in scientific research. In the courtroom, expert witnesses, chosen and paid by the adversaries, are invited to give their opinion. The opinion may include some reference to research, published or unpublished, but it doesn't have to. If research is cited, there is no procedure for evaluating its reliability. Expert witnesses often simply refer vaguely to their "experience." What they are asked to provide, then, is essentially an educated guess. Although they are expected to have certain credentials, they needn't give the basis for their conclusions, except in the uncommon circumstance of a scientifically sophisticated cross-examination. This approach essentially reverses the scientific process. In science, the evidence leads to the conclusion; in the courtroom, the expert's conclusion comes first and becomes the legal evidence. Not surprisingly, the answers yielded by these two approaches may differ greatly.

Much of the discrepancy is simply a matter of how the legal process has been structured. But I believe it is exacerbated by a genuine lack of understanding between the two disciplines. The law and science represent two strikingly different ways of thinking, which reflect their different methods. The law frames questions in adversarial terms, and lawyers see problems as best resolved by controlled argument. In contrast, the scientific method is (ideally) not adversarial, but cooperative, and scientists usually find answers in the slow accumulation of evidence from many sources. The different ways of thinking are so ingrained that they may be virtually unconscious. For example, a lawyer questioning an epidemiologist in a deposition asked him why he was undertaking a study of breast implants when one had already been done. To the lawyer, a second study clearly implied that there was something wrong with the first. The epidemiologist was initially confused by the line of questioning. When he explained that no single study was conclusive, that all studies yielded tentative answers, that he was looking for consistency among a number of differently designed studies, it was the lawyer's turn to be confused. Similarly, I am occasionally asked by lawyers why the New England Journal of Medicine does not publish studies "on the other side," a concept that has no meaning in medical research. An adversarial approach is very effective for settling many types of disputes, which is no doubt why the law is based on it, but it is not the way to reach scientific conclusions. Yet science in the courtroom, no matter how inadequate, has great impact on people's lives and fortunes. Later I will discuss how the system might be reformed to make it better able to deal with scientific questions.

The fifth theme is the pervasive effect of the profit motive on so much of our public life. Just as the struggle to determine the facts about breast implants is between science and the law, the struggle for financial gain is between the manufacturers and the plaintiffs' attorneys. In their haste to cash in on the rapidly expanding market for silicone breast implants, the manufacturers glossed over their responsibility to perform adequate studies of the biological effects of implants. Internal memos from one manufacturer, Dow Corning, showed that there were safety concerns within the company that were not investigated. In fact, one memo instructed salespeople to wash the oily slick off implants before showing them to plastic surgeons at trade shows, thereby removing evidence of leakage. The overriding concern was to win the race against a competing manufacturer.

After the FDA ban, when the manufacturers were clearly on the ropes, the plaintiffs' attorneys moved in to capture the profits of the manufacturers. Since the lawyers are paid on a contingency basis -- that is, a percentage of the plaintiff's award or out-of-court settlement -- they had every interest in filing as many lawsuits as possible. If they lost most of them, it didn't matter. With contingency fees customarily in the range of 30 to 40 percent, all they needed was to win a few cases from time to time and settle most of them out of court. The system resulted in relatively small gains for many women and enormous gains for a few lawyers. The stake of the lawyers in the breast implant dispute is now so great that some of them have embarked on a campaign to discredit and harass scientists who are conducting epidemiologic research on the subject. Dr. Sherine Gabriel of the Mayo Clinic, author of the first published epidemiologic study of implants, was served with subpoenas demanding that she produce large numbers of documents about the women in her study. Authors of two other epidemiologic studies were served with similar subpoenas. These researchers say that responding to the subpoenas has been extremely burdensome and time-consuming. They believe the harassment might well dissuade other investigators from conducting epidemiologic studies. The avarice of manufacturers is necessarily restrained to some extent; if they are too obviously heedless of safety, they will lose their customers and find themselves in trouble with the law. Plaintiffs' attorneys are not similarly restrained. Their clients gain right along with them, and they are not only acting within the law, but using it as an instrument.

Finally, the sixth theme is the way in which the media present medical issues to the public. Until about 1990 the media had little to say about breast implants, except to report a few large jury verdicts. But as the time approached for the manufacturers to produce evidence of safety, the intermittent media reports became more frequent, until finally the media coverage was nearly incessant. It also grew more sensational and uncritical with time. On television and in newspapers and magazines, women -- some of them celebrities -- began to recount stories of deformity and disease stemming from their implants. The subject became a staple of talk shows. The tone of the coverage was anything but analytical. Instead, the premise that breast implants were dangerous seemed to be tacitly accepted. In addition, it was often strongly implied that the FDA, as well as the manufacturers, knew the devices were dangerous and suppressed the fact. The story played to both the American fascination with newly discovered, mass health hazards and the penchant to assume a cover-up in any disaster. Instead of presenting a complicated health story, the media simply generated outrage.

Which came first -- the media spin or public opinion? As in all media-intensive stories, it is impossible to say because public opinion and media reports are so closely intertwined. What can be said is that the tone of the coverage almost certainly reflects a predisposition of the public to see it that way. The public is easily swept up in medical alarms, particularly when there is an element of wrongdoing involved. The power of the media and public opinion is immense, and that is by and large appropriate in a democracy. But not all endeavors are meant to be ruled by public opinion. Justice, for example, should not be, nor should science.

In subsequent chapters, I will consider each of the above themes using the breast implant controversy as a prism. But first, it is necessary to review the nature of the implants themselves -- how they were developed, what they are made of, how the body reacts to them, and who has them and why. Understanding these devices makes it easier to see how the controversy about them illuminates the themes of this book.

FRONTLINE / WGBH Educational Foundation /
web site copyright 1995-2014 WGBH educational foundation