7

The Eclipse of Care

Cora was eighteen and beautiful. Slim, with long blond hair, about average height. She had just finished high school, where she had been the homecoming queen. She was set to attend college, though she wasn't certain what direction to take there. She had a boyfriend but was worried he might want to leave her, while at the same time knowing her parents didn't approve of him.

At a rock festival with her boyfriend, she got lost and, trying to find him, had taken a fall and injured her arm. She was admitted to a local hospital for treatment and sent home from there. Several days later, in a state of perplexity she was brought to the psychiatric unit where I have inpatient beds.

Had she been traumatized or abused in some way? Had she been taking drugs and had a trip gone awry? Had her boyfriend left her? Her mental state was quite unstable, but despite having input from the many people involved in looking after someone in hospital I couldn't make a diagnosis I was happy with. Cora was not hearing voices, did not have delusional beliefs, and was not consistently depressed, elated, or anxious. But she was volatile. At times in the ensuing weeks, apparently improved, I gave her leave to go out with her parents, but she was typically brought back severely confused again—sometimes only minutes after having walked out through the hospital doors. At other times she was almost completely unresponsive and inaccessible. I could see no reason to give her an antidepressant or an antipsychotic. On occasion when she seemed particularly agitated I wrote her up for a minor tranquilizer—a benzodiazepine.

Finally after about six weeks she went on weekend leave with her parents, held steady on her own, and did not come back. I was happy to file her case as diagnosis unknown. I heard she was doing well at college and was still dating the same boyfriend.

I saw her again a year later—8 1/2 months pregnant. She was clearly too unwell to be managed at home. But where she had been mute and inaccessible previously, now she was overactive, manipulative, and attention-seeking while still seeming confused; her actions did not seem fully under her own control. She looked as though she might go into labor at any moment, so I held off medication.

After the birth, I sent her to a hospital that had a facility for new mothers and their babies. The psychiatric team that took over her care there, I learned, thought she had schizophrenia. She was put on regular antipsychotics but apparently was not making much progress and the baby was taken from her. Some months later, I heard she had been given weekend leave; one evening, having told her parents she was going out for a walk, she laid her neck on the track in the face of an oncoming express train.

Looking back at Cora's confusion, emotional lability, and switches between immobility and overactivity, I came to see that she had a textbook case of uncomplicated catatonia. Few readers of this book will know what catatonia is, as it has supposedly vanished, even though fifty years ago up to 15 percent of patients in asylums were estimated to suffer from it, and it was one of the most horrifying mental illnesses. While mental health professionals are aware catatonia is still listed in the Diagnostic and Statistical Manual, few would spot a case if faced with it.

If Cora had a rare condition that doctors do not now need to recognize, if she was the exception that proves the rule of medical progress, she would have been unfortunate. But in fact up to 10 percent of patients going through mental health units in the United States and indeed worldwide still have the features of catatonia—if these were looked for.¹ Sometimes the only condition they have is catatonia; other times catatonic features complicate another disorder and resolving the catatonia may make it easier to clear whatever other problem is present. But almost no one thinks of catatonia and so, like me, they miss the diagnosis. Cora was given antipsychotics, which are liable to make a catatonic syndrome worse. She died when a few days' of consistent treatment with a benzodiazepine would almost certainly have restored her to normal, making her death scandalous rather than accidental.

The benzodiazepines are a group of drugs that are no longer on patent, and thus no company has any incentive to help doctors see what might be in front of their eyes when it comes to a disease like catatonia. Instead, company exhortations are to attend to diseases for which on-patent drugs are designed, even if this means diseases conjured out of thin air—disease mongering—such as fibromyalgia to market on-patent medications such as Pfizer's Lyrica (pregabalin) or restless legs syndrome, a disorder conjured up as a target for GlaxoSmithKline's Requip (ropinirole).²

No one has any idea how many versions of Cora's story play out in daily clinical practice, the opportunity cost of disease mongering. These deaths are lost in the chatter about disorders that match up with onpatent drugs, invisible to doctors pleased with themselves for making a fashionable diagnosis like fibromyalgia and who, even in the face of treatment failure, will add ever more on-patent drugs to a patient's treatment regimen rather than go back to the drawing board and look more closely at the patient in front of them. Once upon a time the height of medical art lay in being able to go back and look at cases afresh and match the profile of symptoms against less fashionable or apparently uncommon disorders—no longer.³

But the outlines of an even more disturbing series of scandalous deaths have emerged from the pages of this book. Unlike catatonia, deaths in this case are from a disorder with no name. It is never recorded on death certificates as a cause of death even though we suspect that in hospital settings nearly two decades ago it was the fourth leading cause of death,⁴ and reports to regulators of deaths from this cause increase annually.⁵ It is almost certainly commoner now as drug prescriptions have escalated, and even commoner in community settings than in the hospital, but yet in an evidence-based medicine and guideline-driven era, there is no evidence base or guideline for its management.

If we are to cure this disease, we clearly need to name it. This is a disorder that was formerly sidelined as iatrogenic but if missing or corrupted data lie at the root of the problem, getting doctors to shoulder the blame—as iatrogenic suggests—no longer seems correct. A possible name is pharmakosis—a name that hints at some loss of insight.⁶

Here is where the interplay between cure and care, outlined in earlier chapters, comes into clearest focus. The critical test of medical care lies in how a doctor or medical system deals with the possibility that the latest treatment might be responsible for part of a patient's current problems—that the poison may actually be poisoning. Rather than care, for over a century we have had a default option to regard cures as an excellent form of care. But here is a disorder to cure, which offers no choice between cure and care—they are one and the same.

Thalidomide is the drug disaster that is classically seen as inaugurating our modern medical era but retrospectively it looks more like a bookend for an older style of medicine than representative of the problems that drug treatments and pharmaceutical companies now pose. Most of us, whether doctors or patients, likely think we can link problems as obvious as those caused by thalidomide to treatments we might have taken. But we are not faced with such obvious problems any more. The difficulties in grappling with what happens when treatments go wrong now come instead from the mechanisms put in place to ensure thalidomide could not happen again—these include controlled trials, a prescription-only status for drug treatments, and efforts to restrict drug use to traditional medical diseases. The story of tolbutamide, whose development paralleled that of thalidomide, brings out far better the difficulties we now face.

WHEN TREATMENT GOES WRONG

The discovery of insulin in 1922 is one of the most celebrated breakthroughs in medicine. A disease that came with sweet-smelling and frequent urine had been recognized in antiquity and had been named diabetes. It was occasionally possible to manage the disorder, which we now know is caused by a lack of insulin leading to raised blood sugars, for a time in older people by restricting sugar intake, but sweetsmelling urine heralded the end for most people. In childhood, the disease was even more malignant; for juvenile-onset diabetes, the discovery of insulin was the difference between life and death.

After 1924, the availability of insulin meant that a growing number of people could survive for decades longer than had been possible before that. Still, as those treated aged, a series of complications of living with diabetes became clear. The blood vessels of the eye might deteriorate, leading to blindness. Damage to the blood vessels or nerves to the pelvis brought about impotence. The nerves or bloods vessels to the legs might be compromised, leading to ulcers, possibly gangrene, and potentially amputation. The risk of heart attacks and strokes was increased.

It seemed reasonable at the time to think that these problems in part stemmed from the mismatch between naturally and artificially controlled blood sugars and in part from the fact that the insulin initially used was bovine or porcine rather than human insulin. But whatever the cause of such problems, people were alive who wouldn't have been alive and while they were alive the problems could be worked on. Perhaps improved preparations of insulin would reduce the risk of complications.

In an effort to improve the care of patients with diabetes, researchers renewed their focus on blood sugars and discovered that these vary substantially during the day in everyone, but especially in patients with diabetes. Might a tight control of blood sugar variability improve outcomes? Controlling blood sugars requires gadgets to read sugar levels and a range of insulin preparations. It also requires close cooperation between the medical team delivering care and the patient with the disease.

The teamwork that grew up around monitoring the hazards of diabetes and its treatment during the 1940s and 1950s became a byword for good medical care. Giving injections of insulin sounds simple, but there is a technique to giving the treatment subcutaneously. You need to learn what the symptoms of an overdose might be—slightly too much insulin risks reducing blood sugars to the point where the taker becomes confused or slips into a coma. Strategies have to be worked out to manage this hazard. The effort to balance the risks that come from allowing blood sugar levels to ride at too high a level and the risks inherent in reducing them too aggressively has to be incorporated into lives that need to be lived. A workman doing heavy labor will have to juggle things in a different way than a ghostwriter working by computer from home. Some people need gadgets to check their blood sugar levels, others seem able to read their bodies.⁷

Getting this right requires teamwork. The nurses of the diabetes team learn from Mr. X how he manages to slot the need for blood testing and injections into his schedule, while Ms. Y tells them of how she manages. They hear about how people juggle social situations, from business meals to outdoor activities with friends. And they pass these lessons on to new patients who have to be helped to handle the ramifications of a diabetes diagnosis at the right pace for them. If medical care is key to how well patients use the technologies available, and therefore how well they do, advance in technology is key to inching forward in treatment possibilities. Teamwork, in turn, helps make the best possible use of those new technologies—such as the new pills that were discovered in the 1950s that could, in tandem with insulin, better manage blood sugars.

While insulin mobilized all the best in medicine, it did not provide a good basis for business. Lilly attempted to wrest the American patent for insulin away from its discoverers—the University of Toronto—but failed.⁸ But even had they succeeded, the commercial opportunities for an injection were limited, even though insulin was quickly used for all sorts of things other than just diabetes. The drug's appetite-increasing properties were put to use in rest cures aimed at “building people up” through a program of sleep and eating. In high doses insulin will induce a coma—the ultimate rest cure—and insulin-induced coma treatment was used to treat drug abuse and schizophrenia.⁹ Despite all these uses, a pill would be far better for business than any injection because a much greater number of people have raised blood sugars (prediabetes) than have frank diabetes and this market would be easier to develop with a pill than with an injection.

The research that led to a pill that seemed promising stemmed from the first of modern medicine's magic bullets, the sulfanilamide antibiotics that were discovered in Germany just prior to the outbreak of World War II.¹⁰ Building on these discoveries, French investigators soon developed a range of sulfa drugs, among which was one that lowered blood sugars. Despite the devastation of war, it was in fact a German company, Boehringer-Ingelheim, that in 1956 came up with the first oral blood-sugar-lowering drug (a hypoglycemic drug), carbutamide, and soon thereafter another—tolbutamide.

The Michigan-based Upjohn company bought the US rights for tolbutamide and began marketing it as Orinase in 1961. It effectively lowered blood sugar, but it turned out to be close to useless for the management of juvenile-onset diabetes. It would not replace insulin, then, but it might have a place in the treatment of patients who developed diabetes in their middle to older years who had been able to manage their condition by diet alone before they eventually graduated to insulin.

The availability of tolbutamide led to distinctions between type 1, or insulin-dependent diabetes, and type 2, or noninsulin-dependent diabetes. Where before there had been little emphasis on seeking out information on blood-sugar elevations, as few people wanted to know about diabetes until they had to, there was a new premium on detection. For people who had elevated levels of blood sugars but who were not overtly diabetic, dieting offered a means to help regulate their blood sugars, but this was hard work. Tolbutamide was an easier option. It even seemed possible that treating high blood sugars early might prevent people from developing diabetes. In addition, supplementing insulin with tolbutamide might minimize the requirement for insulin and offer better control of blood sugars, potentially reducing problems in the longer run.

In 1961 a long-term study by the National Institutes of Health (NIH) involving over a thousand patients began in which Upjohn's tolbutamide was compared to insulin, placebo, and an amphetamine derivative developed by Smith Kline & French, phentermine, which it was thought might help because it suppressed appetite and led to weight loss. In 1967, a problem came to light: more patients were dying on tolbutamide than on phentermine, insulin, or placebo. The result ran completely against expectation, as the trial protocol excluded anyone thought to be at risk of dying, and controlling blood sugars should have reduced the risk of complications.¹¹

Had some of the treatment centers failed to adhere to the protocol or was there some other explanation of the findings? The investigators could find no explanation aside from some possible side effect of tolbutamide, and in 1969, the study was terminated so as not to put any patients on tolbutamide at further risk. The investigators consulted with Upjohn and the FDA, and an FDA meeting to discuss the study was organized for May 1970. The results of the study were not expected to be made public until an American Diabetics Association meeting the following month.

However, the study results appeared first in the business section of the Washington Post, just prior to the FDA meeting. The study, it seemed, had implications for the health of Upjohn that were of concern to many in the commercial sector. The first that many physicians knew of the issue was when anxious patients faced them in the clinic that day with the news. Close to a million people were on tolbutamide in the United States—a lot of anxious patients. This was not the way anyone was used to things happening in medicine.

Patients in many cases found that their doctors seemed personally offended, as if their judgments in putting the patient on this treatment were being questioned. But the doctors were stymied. Their patients didn't know what was going on, but neither did the doctors.¹² These doctors flooded the FDA with letters suggesting the bureaucrats had no idea how much distress this was causing patients. The FDA scrambled to respond but they too were faced with a novel situation. It had never been part of the FDA's brief to tell doctors how to practice medicine. They left the NIH academics to fight it out with academics recruited by or otherwise taking Upjohn's side.

Patients also sought out the FDA—asking the agency whether it was still safe to take tolbutamide. Some had worked out from the figures given in the media that over a decade tolbutamide may have killed more Americans than had then been killed in the Vietnam War. How could such a drug still be on the market? The FDA directed patients back to their doctor, still, the agency said, the person best placed to decide the right course of action for them.

The FDA was at pains to insist it was not involved in the practice of medicine and did not wish to subvert clinical judgment. As Senator Kefauver put it in the debate surrounding the 1962 Kefauver-Harris amendments to the FDA bill, introduced after the thalidomide crisis and aimed at controlling the pharmaceutical industry's inappropriate marketing of drugs, “It should be made very clear to Senators and to the country; this is not a Federal control bill. This is a Federal information bill…. I am not talking about regulating medicine between the physician and the patient.”¹³

Many clinicians simply refused to believe the findings on tolbutamide. They were not used to having data trump their common sense. Perhaps there were more dead bodies in a trial of the drug, but these doctors had given tolbutamide to hundreds of patients, few if any of whom had dropped dead. Besides, how could trials like this take into account all the people whose lives must have been saved by having their blood sugars better controlled? It didn't make sense that a drug doing something so obviously right could be causing problems, no matter what a fancy controlled trial cooked up by some academics might have shown.

Upjohn moved quickly to mobilize a coalition of experts to cast doubt on the NIH research. These company-recruited academics pointed to a number of tolbutamide trials in which, they claimed, there had been no hint of a problem. While these trials were smaller, there were many trials on one side, with only one, albeit larger, trial pointing to a serious problem on the other. The insults began to fly between the academics, with one side accusing the other of being hysterical publicity seekers, attempting only to advance their careers, and the other pointing to the conflicts of interest stemming from participation in Upjohn-convened panels.¹⁴ Tolbutamide remained on the market without warnings until the 1980s.

In the meantime, the availability of tolbutamide had put a new premium on detecting elevations of blood sugar. Because of this, when any of us have the most basic of blood screens, the results will include blood sugar levels. Studies of those results concluded that, as of 2008, 30 percent of Western populations have either diabetes or prediabetes. In the wake of such findings, majority medical opinion supported treatment for such patients as early as possible with the hypoglycemic successors to tolbutamide.¹⁵

A further National Institutes of Health trial, however, reported in 2008 that tight control of blood sugars with tolbutamide's hypoglycemic successors led in fact to a higher death rate than was found in patients whose blood sugar control was allowed to vary more.¹⁶ By 2008 a series of blood-sugar-lowering drugs from Rezulin to Avandia either had to be withdrawn from the market or were required to carry warnings after evidence emerged that they too were linked to excess mortality. Nevertheless, the hypoglycemic market remains one of the blockbuster markets, in 2010, worth over $25 billion per annum and showing growth of 10 percent per annum. This has been a market in which the best teamwork and care in medicine has been harnessed to company sales, but if the data are to be believed, this teamwork is in fact now delivering at least some patients to an earlier death than they might have otherwise had.

The story of tolbutamide prefigures a series of crises that developed around Prozac in the 1990s and Vioxx in 2005. But in 1970, the issues were new for the regulators. Before the amendments to the US Food and Drugs Act triggered by thalidomide in 1962, such conflicts would have been straight contests between doctors and industry. It was in fact the number of doctors who spoke out, at a risk to their careers, that brought the problems with thalidomide originally to light. The public, however, attributed the discovery to the actions of the FDA and as a result for the first time saw in the FDA a third party whom they might turn to for help.

THE WIZARD OF OZ

When asked what a regulatory agency does, many involved in healthcare respond that agencies like the FDA, either in a commissioning capacity or in conjunction with universities, even if supported by pharmaceutical companies, are responsible for the clinical trials that bring drugs to market. In this view, the agency, directly or through independent investigators cooperating with them, designs the protocols for trials, arranges for the conduct of the research, collects, tabulates, and analyzes the data that academics then write up in articles that faithfully represent the data. These articles are peer-reviewed, and especially if outlining notable advances, they are likely to appear in the most distinguished journals, offering further assurance, so it is thought, of the validity of the data and the conclusions presented. It is also commonly assumed that the FDA in some way houses the raw data so that should a problem develop with a drug after marketing, someone independent of the pharmaceutical industry can consult the data and determine whether there is cause for concern. None of this is the case.

What most people think the FDA now does is very much what the American Medical Association used to do with all new drugs from the 1940s to the mid- 1950s but does not do now. Once the AMA stopped doing this, except in the increasingly rare instances, such as the NIH tolbutamide studies, in which the government or an independent institution decides to do a study itself, there is no independent research done on new drugs. When it comes to the study of drug treatments, the medical profession has been eclipsed. The public think the FDA has stepped into the breach but in fact pharmaceutical companies now control everything.

Companies decide which trials to conduct, which ones to make public, and what data to release. The trials are designed not only to gain regulatory approval but also to test for characteristics that fit company marketing requirements and may bear little relationship to what some of the drug's most telling effects are. Thus the “benefit” of SSRIs for premature ejaculation was so clear-cut that trials were almost unnecessary. In contrast, the effects of the drugs on mood were so weak that in the clinical trials of SSRIs as antidepressants, hundreds of patients were needed to come up with statistically significant results. Yet, the drugs were licensed for depression and have since been called antidepressants, while the effects on sexual functioning, if acknowledged at all, are labeled side effects. It is not the job of the FDA, or any regulator worldwide, to tell a company what its business should be.

Drug trials are increasingly conducted in settings and by notional investigators that suit company interests. When it came to testing antidepressants in children, the first wave of trials took place in North America, recruiting heavily among children in foster care, before a second wave moved to South Africa, Brazil, and elsewhere. It is not the job of the FDA, or any other regulator, to note this and wonder if trials done in circumstances such as these will produce results that should be allowed to dictate the treatment of the child across the road—or at least the regulators have not read their safety brief in this way.

In the case of tolbutamide, the study was run by the NIH, but now private companies known as clinical research organizations (CROs) run almost all trials. These trials may only have had ethics clearance from the CROs' own panels, and they may include nonexistent patients (one convenient aspect to these bogus patients, unlike those taking tolbutamide, is that they can't die). Also in contrast to the patients dying on tolbutamide, patients suffering adverse effects in company trials may be coded as dropouts for noncompliance rather than injured by treatment. Finally, in contrast to the tolbutamide trials, the publications stemming from company trials will likely be ghostwritten.

Of even greater importance is that in the case of tolbutamide, the investigators had the data, but today's investigators for statins, asthma drugs, hypoglycemics for diabetes, or antidepressants don't. The file of data from company trials of these drugs notionally goes to the FDA but the regulators in fact work from tables of the raw data already made up for them and company reports as to what the trials have shown. Furthermore, as of the 1990s, the FDA has been encouraged by both Congress and the executive branch of both Republican and Democratic administrations to see itself as a partner of industry; in this climate a zealous commitment to safety is not likely to be viewed with favor by their new partners.

There are no academics now who can speak out about a problem in the way the academics from NIH did in the case of tolbutamide. This shifts the burden to the FDA—but this agency is not equipped to be a scientific arbiter. The FDA acts essentially as an auditor for drug company data and no more. When the raw data are submitted to them, the agency typically samples the clinical records to determine whether they correspond with figures in various tables that have been constructed by the company.

FDA officials may analyze some of these tables themselves but usually will simply comment on the methods used by the company. If further analyses are deemed necessary, the company is asked to perform these. Should a problem arise with the drug after it has been approved for marketing, the FDA will typically write to the company and inquire what the company's database might reveal about the issue. They will ordinarily trust the report that is prepared by the company, despite a series of cases in which company reports have borne little relationship to the underlying raw data, as we shall see below.

The companies, not the regulators, write the labels and any warnings for the drug, though the FDA must approve them. There are a number of technical issues that are considered, such as the precise meanings the FDA has given to words like “rare” and “frequent.” These have been tied down to rates that may be, for example, greater than 1 in 100 (frequent) or less than1in 10,000 (rare), and the FDA will ensure that such words are used in their technically correct fashion.

The FDA's role in a drug's regulation can be brought out by considering the agency's role in regulating a food like butter. While one officer in one part of the FDA may have to decide whether a drug can be labeled as a treatment for asthma or osteoporosis, another may be faced with a slab of yellow material a company wishes to label as butter. Is it butter or lard colored to look like butter? The regulator has a set of criteria for butter, dealing with what it should contain and what it cannot contain. If the yellow material meets the criteria, the regulator has no option but to let it on the market labeled as butter. In the case of butter or other foods words such as “organic,” “fresh,” and “local” are defined by the regulator, and the challenge for companies is often to find a way around the definitions in order to make use of these premium descriptive terms. It is not the job of the regulator to decide if this is good butter or not. Nor is it considered the job of the FDA to decide if butter is good for our health. Consumer agencies survey available butters to answer the first point, while medical associations or others in healthcare will issue guidance from time to time on whether we should be eating as much butter as we do.

But unlike butter, in the case of drugs FDA approval is now taken to mean that these treatments are good treatments and that treatment is good for you—with no consumer groups scrutinizing just how good these products are, and above all, no groups of physicians reviewing the issue of whether it is appropriate to be on as much treatment as we are.

There might be less cause for concern if the FDA were the type of investigative agency most people imagine must have been put in place following the thalidomide scandal or if it were the kind of agency that was in the business of ensuring our medicines get progressively better. When we go to sleep at night with a child in the next room on some new medication, we expect the treatment will be at least marginally better and safer than treatments we may have been prescribed when we were children. After all, the music system is so much more advanced than the one around when we were young and there is a computer now where there was once a typewriter. But this is not the case for many drug treatments. There is every chance our family members will be taking a drug, whether an antibiotic or antidepressant, that is less effective than the one we were treated with or a new drug whose hazards remain to be discovered. All that is in place between our children and possible disaster is a set of auditors whose job it is to tick the boxes.

It is in fact a lot harder for the FDA to audit the books of Pfizer, Lilly, or GlaxoSmithKline than it was for the Arthur Andersen firm to audit Enron. In financial circles it is recognized that after a period of time auditors come to see the world the way their clients do. As a result it is regarded as good practice for major corporations to change their auditors every five years. But there are no changes of auditors when it comes to pharmaceuticals.

When a plane falls out of the sky, the Aviation Safety Board audits what has gone wrong rather than the Federal Aviation Board, the agency responsible for letting a plane fly in the first instance. But when a problem blows up with a drug, it is often the same FDA officials who have made the decision to let the drug on the market who are now called upon to make a further set of judgments that may reflect badly on their original decision.

When the crisis blew up around drugs like tolbutamide in the 1970s, one of the few safeguards in place was that regulators worldwide were paid from the public purse. But things have changed. Britain was one of the first countries in the 1980s to switch to a system where industry paid to be regulated, and the regulator was encouraged to become a business selling “regulatory services” to the industry and promoting itself as one of the fastest licensing authorities in the world for new drugs.¹⁷ The United States moved to a similar system in 1992, with the passage of a Prescription Drug User Fee Act (PDUFA) that allowed the FDA to collect a fee from industry based on marketing applications reviewed. The intention was, by employing more reviewers in order to speed up processing of industry applications, to get rid of the bureaucratic delay that, ironically, had been the FDA's only contribution to averting a thalidomide disaster.

A new emphasis on partnership has blurred the boundary between the FDA and industry. Since drugs began to be regulated in the United States in 1906 there has always been a revolving door between the regulatory agency and industry; senior figures in the regulatory apparatus found jobs in pharmaceutical companies or consulted for them after they left government, or employees from industry moved into senior regulatory positions. This was later true of the regulatory agencies that followed in Europe. This revolving door makes it difficult to know how independent our regulators are. If the regulator knew he could be scrutinized because the data from the clinical trials used to get drugs on the market were publicly available, this might not be a big issue, but the data are not available so none of us can assess how much the lure of a future job might bias a regulator to give a company favorable treatment.

Whatever shuttling between industry and the regulatory apparatus there has been, until the new emphasis on partnership between industry and regulator was put in place it would have been considered scandalous to find senior regulators sitting down with industry representatives, academics enmeshed in industry, and patient groups that had been sponsored by industry to issue a ghostwritten consensus statement advocating the detection and treatment of disorders in a manner that can only increase the sales of drugs.

Thus in 2005, Tom Laughren, head of the FDA section responsible for licensing drugs that act on the central nervous system (known as the CNS section), was a co-author on an article advocating the detection and treatment of mood disorders in those with other medical illnesses, and indeed giving consideration to prophylactic antidepressant treatment in those who might be put on other medical drugs at risk of triggering depression.¹⁸ The article was in fact written by Scientific Therapeutics Information.

In 2003, Laughren was a participant in a “consensus conference” convened under the auspices of the American Academy of Child and Adolescent Psychiatry but organized by Best Practices, a marketing firm whose website states that “we bring together opinion leadership and direct services to the pharmaceutical and biotechnology industries.”¹⁹ Its services include “Consensus Development Conferences…in areas of clinical controversy.” In this case the meeting was aimed at increasing the recognition and treatment of childhood bipolar disorder, a condition that few psychiatrists outside the United States believed existed.²⁰

In 2007, Laughren was named as an “author” of an article promoting antipsychotic drug studies in children labeled with “Impulsive Aggression.”²¹ This article, like the others above, came out of a sponsored meeting; in this case by Abbott Pharmaceuticals, Bristol-Myers Squibb, GlaxoSmithKline, INC Research, Janssen, Johnson & Johnson, Eli Lilly, Novartis, Pfizer, Solvay, Annie E. Casey Foundation, Forest Research Institute, Jazz, Otsuka, and Sanofi Synthelabo.

In these cases, Laughren endorsed a treatment option when, in his FDA capacity, he might later be called on to adjudicate whether companies should have a license for marketing the use of antipsychotics for children who ostensibly have bipolar disorder or impulsive aggression disorder. There is no reason to believe that this quite extraordinary situation is anything other than standard behavior within the FDA now.

When we are ill, we are at our most vulnerable and need someone to look after us. Someone who ensures we get treatments that work, treatments that are as safe as they can be. For a century this figure, in the shape of doctors like Alfred Worcester and Richard Cabot, used to be our doctor. After the thalidomide tragedy in 1962 many of us also looked to the FDA as a guardian of our well-being, but if the regulator ever was a truly independent safeguard, the reality now is that he is more like a Wizard of Oz figure, pulling the strings of publicity rather than capable of making any meaningful changes—a figure who has made his accommodations with the witches of the East and North.

THE CUTTING EDGE OF CARING

In marked contrast to the increasingly cozy partnerships between regulators and industry in other countries of the world, there have been individuals within the FDA who in recent years have spoken out when they felt that critical health information was being suppressed or unnecessarily delayed. In 2004, David Graham from the FDA's safety department went public with news that as early as 2000 the scientific data had pointed to a seven-fold increase in the risk of a heart attack among people on Vioxx though the FDA were still refusing to require Merck to warn doctors and patients about this risk.²² He estimated there might already have been thirty thousand unnecessary heart attacks that could be linked to Vioxx. Graham also pointed to the risks of death from asthma inhalers. Andy Mosholder, another FDA staffer, in 2004 similarly presented an analysis suggesting that the available company trials pointed to a risk of suicide from antidepressants. Both Graham and Mosholder were threatened and both needed support from the Whistleblowers Act. Terrible though such efforts to gag them are, it is unimaginable that anyone within the regulatory apparatus in Europe would step out of line in the way Graham or Mosholder did.

But why should a regulator have to lead the way? Drugs are available by prescription only in part so that doctors will sing out should there appear to be a problem. It's not the job of a regulator to go to the wall for patients—but it is close to the definition of good medical care for a doctor to do so. And even if all the doctors in a particular specialty have been cajoled by a pharmaceutical company's line or its perks, or fall short in their duty to provide safe care, still in the case of drugs such as Vioxx and Prozac the published data in major journals like the Journal of the American Medical Association and the British Medical Journal point to clear increases in risk that should have alerted the medical profession to look for problems. David Graham and Andy Mosholder were not dealing with confidential data that only the FDA had access to—they appealed to data that was within reach for thousands of doctors—but almost none spoke up.²³ Why not?

When doctors speak out about a treatment hazard, they often encounter antagonism from colleagues; the revelation is likely to be perceived as bad for medical business and to reflect badly on other doctors who have prescribed this treatment. This antagonism may be even more likely now as company marketing has become adept at ensuring that other doctors do not only see a product being attacked but see their own views being challenged. In addition, most doctors have accepted prescription privileges as a God-given right to a monopoly in drug treatments of people at their most desperate and most vulnerable and have divorced this right from a sense of duty to do something about the hazards of treatment. In this respect, doctors have become one more link in the drug distribution chain rather than the people who once took charge of medical treatments. Something quite extraordinary must have happened to produce such a change in the practice of medicine. The tolbutamide and, later, Prozac and Vioxx cases bring out what has changed.

Concerns that the blockbuster antidepressant Prozac might make people taking the drug suicidal broke in 1990, following the publication of a set of clinical cases in the American Journal of Psychiatry, as discussed in chapter 3.²⁴ As this article made clear, within days of starting Prozac some patients became more agitated and suicidal. The agitation got worse if the dose of the drug was increased. It cleared up if the treatment was stopped and reappeared if the treatment was restarted. While depression can cause suicidality, both the treating doctors and the affected patients identified the new state as quite different from their previous experience. In at least one case, the drug led to a completed suicide in a child being treated for anxiety. Finally, intervening with an antidote that blocked the effects of Prozac brought about an improvement. In this series of cases and in others published in the months that followed, according to the rules Robert Koch set out in the 1880s to determine when a bacterium or a drug might have brought about some effect (see chapter 3), there was a cast-iron argument that Prozac had caused the problem it stood accused of.

The main way hazards may be hidden right under the noses of doctors centers around Ronald Fisher's efforts in the 1930s to investigate whether fertilizers worked or not, as outlined in chapter 3. Fisher, quite arbitrarily, set up a standard—if on nineteen occasions out of twenty plants in a fertilized patch did better than those in an otherwise equal but nonfertilized patch, this was said to be statistically significant and the fertilizer could be said to be effective. If the result came out indicating that the fertilized patch did better only seventeen or eighteen times out of twenty, in Fisher's view, proper scientists should conclude they either had not designed the experiment correctly or they were dealing simply with the play of chance.

Soon after Fisher outlined his view of statistical significance, Jerzy Neyman, another early statistician, argued Fisher was throwing common sense out the window and offering a recipe for scientific sterility, not scientific progress. For most people, having a drug prove its worth in a study in which thousands of patients participate sounds more impressive than demonstrating a benefit in a handful of people. But as was discussed in chapter 3, the drug to go for is the one that consistently shows up as working in a small sample. Snake oil, which contains omega-3 and other fatty acids, could be shown to have statistically significant effects on depression rating scales or for pain relief purposes and possibly for some other conditions as well—provided several hundred patients are recruited to the trial. As Neyman was first to point out, we can be fooled by the fact that having hundreds of patients in a study can make a trivial finding statistically significant. Fooled to the extent that these hints of effectiveness can be sold by drug companies as convincing evidence their drug works and should all but be put in the drinking water.

Neyman's beef with Fisher becomes even clearer if we turn to drug hazards. To see this, let us move from fertilized fields to a hazard analogue, such as being faced with a loaded gun. If Fisher had said that it was only when there were bullets in nineteen chambers of a twentychambered gun that the gun could be said to be loaded this would make no sense to any of us. In the experiment that is real life, a lot of us wouldn't go near a gun with even one bullet in its twenty chambers. One bullet is significant enough for most of us. Fisher's approach makes sense in situations where skepticism is called for—as when faced with claims by a huckster or a company trying to make money out of a remedy for sick and vulnerable people. It might indeed be reasonable in this case to suspend skepticism if a treatment “worked” nineteen times out of twenty. But it makes no sense for doctors or patients to be skeptical when hazards are involved—only pharmaceutical companies have an interest in being skeptical about the existence of hazards.

But when it comes to drug hazards, pharmaceutical companies can now confidently bank on the medical community and the FDA following Fisher, not Neyman. The studies of Prozac, and later of other antidepressants, threw up double the number of suicidal acts than were found in patients on placebo, but the numbers involved were not statistically significant. Merck was happy to report over four times more heart attacks on Vioxx than on placebo, again because the numbers were not statistically significant. For Fisher, Merck, and Lilly, this essentially meant the suicides and heart attacks were not happening, whereas for Neyman the onus was on Merck and Lilly to show their gun wasn't loaded.

Following Fisher and the FDA, when it comes to a drug-induced injury, doctors now routinely line up to say the bullet through the head while on Zoloft or the heart attack while on Vioxx were just anecdotal events, and while anecdotes are regrettable, physicians have to deal in science, and if a finding is not statistically significant, the science doesn't support the existence of a risk. The doctors who line up in this way are not the rapidly vanishing doctors of yesteryear sticking to a regulatory line on things they know little or nothing about. Instead, they tend to be the most recently qualified doctors who, coached in the conventions of evidence-based medicine, are trained to follow Fisher and not Neyman when it comes to hazards.²⁵ In this respect, we are all at greater risk from tomorrow's doctors than from yesterday's.

These claims may seem incredible, so it is perhaps worth taking a little time to view the emperor from a few different angles before deciding whether he is really wearing any clothes or not. When the first concerns about Prozac made headlines in 1990 Lilly, understandably, rushed to defend their blockbuster. The company examined their clinical trials, which included over three thousand patients, for suicidal acts and published the results of their analysis in the British Medical Journal, claiming it showed no increase in risk on Prozac compared to placebo.²⁶ But smack in the middle of the article are the figures—six suicidal acts in 1,765 patients on Prozac versus one in 569 patients on placebo.

These figures indicate that while there is a chance there is no risk from Prozac, there is an equal chance that there is up to a sixteen-fold increase in the risk, and that the best guess is that there is roughly a doubling of risk on Prozac.²⁷ But according to Fisher's test there was no significant difference between placebo and Prozac on this score. The conclusion the company drew from this, fully endorsed by the British Medical Journal with no subsequent objections recorded from any of the 100,000 readers of the journal was that “data from these trials do not show that fluoxetine is associated with an increased risk of suicidal acts or emergence of substantial suicidal thoughts among depressed patients.”²⁸ The six suicidal acts simply vanished.

A twelve-year-old schoolchild could have told the British Medical Journal that a doubling of risk is an increase. Lots of the one hundred thousand readers of the British Medical Journal work in fields that have nothing to do with the pharmaceutical industry. Even in areas of medicine working closely with that industry, many of the brightest and the best have no conflicts of interest, so missing this doubling of risk cannot be put down to conflict of interest. Missing the problem in this instance cannot be put down to lack of access to the data either—the doubling of risk that the company denied was hidden like a boat in a harbor by being put in one of the most widely read journals in the world. We seem to have entered an Alice in Wonderland world, in which things are whatever the Red Queen says they are.

Emboldened by the complete lack of objections to their depression paper, Lilly went on to analyze the suicidal acts from a series of trials they had done with Prozac in eating disorders. In these cases, any increase in risk could not be attributed to depression. Again there was an excess of suicidal acts on Prozac—a 1.4-fold increase in risk but because this increase was not deemed statistically significant, it too apparently didn't exist.²⁹

When the FDA convened their public hearing in September 1991, there was room on the program for presentations from the public and for a presentation of the “science.” In the three-minute slots they were given, many wives and mothers offered convincing testimony of how husbands and children had been prescribed the drug for anxiety, weight loss, or smoking cessation—conditions not linked to suicide—and, having apparently never been suicidal before, had gone on to commit suicide. FDA officials and experts present acknowledged that the testimonies were striking but made it clear they had to go by what the scientific data showed.³⁰

The scientific data from their trials were presented by Lilly. None of the experts convened by the FDA, nor any of the regulatory officials speaking that day, nor any of those whose tales of horror were weighed on a scientific balance against the clinical trial data and found wanting appeared to notice that, actually, the trial data of increased suicides was entirely consistent with the personal tragedies.

But there is skullduggery to add to mystery here and a case for saying lack of access to the data also played a part in Lilly's getting away with their gamble. Unlike the readers of the British Medical Journal, the FDA had had a chance to see the real figures from the Prozac depression studies and plenty of opportunity to come to grips with the fact that there had in fact been NO suicidal acts on placebo—the real figures were six suicidal acts on Prozac versus zero on placebo. Technically, the risk on Prozac was infinitely greater than on placebo. Lilly had taken a suicide that had happened before the trial had started and filed it under the heading of placebo, in a manner that contravenes regulations and seems close to being fraudulent. These data are now in the public domain³¹—what is not public is any account from the FDA, or other regulators worldwide faced with the same data, as to why they chose to overlook this clearly inappropriate manipulation.

This single placebo suicide was very important to the company, and maybe to the FDA, because its addition to the calculation meant the increased risk on Prozac would not be statistically significant and if not statistically significant six suicidal acts on Prozac vanished. The company knew that medics across the board could be depended upon to agree with this view of significance.

Following Lilly's lead, GlaxoSmithKline in the case of Paxil and Pfizer in the case of Zoloft also took prestudy suicides and suicidal acts from Paxil and Zoloft trials, respectively—seven in the case of Paxil and three in the case of Zoloft—and dumped them into the placebo group, against regulatory rules. The FDA noted what was happening but did nothing, and again has offered no explanation since for their oversight.

Quite astonishingly, when the British regulator (the MHRA—Medicines and Healthcare products Regulatory Agency) caught up with this maneuver thirteen years later, in 2003, and asked GlaxoSmithKline for their Paxil suicide data, making it clear companies should not move prestudy suicides into the placebo group as they had been doing, GSK instead took suicides recorded after the trials of Paxil had concluded and coded them under placebo, even including in the placebo group a patient who had committed suicide after having started Prozac. The MHRA did not object.

When Vioxx ran into trouble with the FDA in the late 1990s, Merck did something similar. In a major trial comparing Vioxx to the older drug naproxen in the treatment of osteoarthritis (the VIGOR trial), the company reported seventeen heart attacks in 4,047 patients taking Vioxx versus four in 4,029 patients taking naproxen.³² This more than fourfold increase in heart attacks could be made to vanish by breaking the patients up into groups and then reporting that when compared to naproxen, Vioxx did not increase the risk of a heart attack for patients without a previous history of cardiovascular disease. Creating groups such as those with previous history of cardiovascular disease and those without reduces the size of any one group compared to the overall group, thereby reducing the risk that a finding will be statistically significant.

In fact there had been three further heart attacks on Vioxx, making twenty in total; this only came to light when plaintiffs later took legal actions against Merck. The company had to make these three heart attacks disappear or they would have had to report that Vioxx significantly increased the risk of heart attack for all patients in the VIGOR trial, including those without previous cardiac problems.³³

For most readers, the invented placebo suicides and suicidal acts in the Prozac, Paxil, and Zoloft trials and the three deleted heart attacks in the Vioxx trials are likely to seem to be the major issue. These manipulations are both a real problem and possibly a criminal offence, but they are a lesser problem than the invisibility of the seventeen Vioxx heart attacks and six Prozac suicidal acts. Lesser in the sense that companies only occasionally have to move a dead body or two from an inconvenient spot in the dataset to a less problematic one, but at the click of a statistical button they hide far more deaths in academic articles that remain permanently in full public view.

Explaining why Fisher's ideas have such traction within medicine is not easy. Regulators have followed this line because the definition of statistical significance offers them a rule of thumb, an almost mechanical procedure that takes the place of a judgment call. For pharmaceutical companies, the issue is simple; Fisher's ideas mean that positive effects in a minority of clinical trials can transform a weak and inessential drug into one apparently certified by science, while at the same time airbrushing its hazards out of existence.

But why do doctors follow this line? A number of medical academics have attempted to grapple with this, pointing out that current dependence on statistical significance testing has created a “junk epidemiology” in the domain of therapeutics. So Louis Lasagna, the first professor of clinical pharmacology in the United States, and later dean of medicine at Tufts University, who introduced randomized controlled trials into drug development, described the approach outlined above as “p-value madness” (to say something has a p-value less than 0.05 is another way to say a finding is statistically significant).³⁴ For Sandor Greenland, professor of epidemiology and statistics at UCLA, “statistical thinking [of this type] has produced a chronic psychosis”³⁵—by which he means that researchers relying on Fisher's ideas have lost touch with reality. Ezra Hauer, a professor of civil engineering from the University of Toronto and authority on analyzing road traffic accidents, explains that “in this manner good data are drained of real content. The direction of the empirical conclusions is reversed and ordinary human and scientific reasoning is turned on its head.”³⁶ For Charlie Poole, professor of statistics and epidemiology at the University of North Carolina, “Statistical significance should be abandoned immediately and universally.”³⁷ Ironically when faced with this issue in 2011, with investors' dollars at stake, the US Supreme Court argued that statistical significance cannot be the arbiter of what an investor might deem a significant risk.³⁸ But patients it seems do not have the same rights as investors.

The differences between much of clinical practice and other branches of science were starkly framed by Kenneth Rothman, professor of epidemiology at Harvard and editor of the journal Epidemiology, in a note about submissions to the journal:

When writing for Epidemiology, you can…enhance your prospects if you omit tests of statistical significance…. We would like to see the interpretation of a study based not on statistical significance, or lack of it, for one or more study variables, but rather a careful quantitative consideration of the data in the light of competing explanations…. Misleading signals occur when a trivial effect is found to be “significant,” as often happens in large studies, or when a strong relation is found “non-significant,” as often happens in small studies.³⁹

Is Poole's term “junk epidemiology” too strong a term to use? Consider an analysis published in 2002 of the trials of Prozac, Paxil, Zoloft, Efexor, and other antidepressants submitted to the FDA by companies seeking to market their treatments for anxiety disorders.⁴⁰ To get into these studies patients had to be anxious and not depressed. The perception at the time was that anxiety had a minimal effect on the risk of suicide. Having combined the data for all these drugs, the authors announced a surprising finding, which the American Journal of Psychiatry was glad to accept. “We found that suicide risk among patients with anxiety disorders is higher than in the general population by a factor of 10 or more. Such a finding was unexpected…. The sample of patients selected was considered at minimal risk of suicide.”⁴¹ Nobody reading the American Journal of Psychiatry registered an objection to this conclusion, which was odd, given that there were eleven suicides in 12,914 anxious patients on active treatment compared with zero suicides in 3,875 patients taking placebo. The conclusion should have been that in this group of patients at minimal risk of suicide, the increased risk that came from antidepressants was more clearly visible than it had been in depression trials.

The same group of investigators looked at suicides and suicidal acts as reported in depression studies on the effects of the post-Prozac group of antidepressants, involving almost fifty thousand patients. The percentage of suicidal acts on antidepressants ranged from 0.15 to 0.20 percent, while on placebo it was 0.10 percent. The findings were not considered statistically significant, leading the authors to make the astounding statement that “the only possible conclusion supported by the present data is that prescription of SSRI antidepressants is not associated with a greater risk of completed suicide.”⁴²

In fact the situation was much worse. These investigators were not aware of ten suicides and suicidal acts miscoded by Glaxo and Pfizer under the heading of placebo. Taking these into account made the difference between antidepressants and placebo statistically significant, so that the only possible conclusion that the present data did not support was the conclusion that there was no increase in risk.

Where one might expect that evidence of only bare hints that a drug might work would not be sufficient to get a drug approved for market, while a doubling or tripling of a hazard on treatment would lead to warnings, this is not what happens. Where one might expect the burden of proving that black was really white would face companies rather than patients, it instead faces patients and any doctors who wish to look closely. This is the point where the world is turned upside down, where the baby is thrown out while the bathwater is carefully preserved.

Paul Anthony, speaking on behalf of the Pharmaceutical Manufacturers Association of America, makes the issues very clear:

I want you to think about it [the issue of hazards] in terms of your reputation. It is really the reputation of a brand that is being signaled. Imagine…someone reporting that they had early information that you may be a child molester. I know that sounds extreme but it is that type of thing…. It is just an allegation…[however] that is what people will remember, and that is the reason there is a lot of concern about presenting early signal information—when you don't really have any proof. It is very different than the kind of rigorous process we had in the past, where you had to do a trial and it had to be statistically significant before you presented that.⁴³

One of the greatest ironies of our evidence-based medicine era is that, faced with evidence of an increased number of adverse events in clinical trial data for its drug, the ultimate company defense has been to fall back on anecdotes. The most astonishing example of this mode of defense followed concerns beginning in the mid- 1990s that the antipsychotic Zyprexa might cause diabetes. Faced with growing concerns, Lilly wheeled out a view from Henry Maudsley, who in 1879 had supposedly recognized an association between psychosis and diabetes.⁴⁴ Lilly authored or sponsored several series of articles that were liberally sprinkled with this quote as grounds to think it was schizophrenia after all, and not Zyprexa, that was causing the problem—with not a single published objection from the psychiatric profession in the United States or Europe.

Invoking Maudsley had an effect, perhaps because owing to an accident of history his name had ended up on Britain's most distinguished psychiatric research facility—the Maudsley Hospital. But in fact Maudsley disliked clinical practice and had very little experience with serious mental illness. He preferred writing. And the book in which a throwaway sentence about diabetes appears had been written when he was twenty-nine, so that even if he had been interested in patients he would not have had time to form a seasoned judgment about their condition, especially as patients with type-1 diabetes did not then live more than a few months from the onset of their illness, and type-2 diabetes was rare.

In North Wales, we had access to over three thousand consecutive admissions to the local asylum from 1875 to 1924, covering Maudsley's career. During this time over twelve hundred patients were admitted with a psychosis but not a single one had diabetes. We also identified every single one of the 396 patients with a psychosis who had presented for the first time to the modern services between 1994 and 2006—not a single one had type-2 diabetes. But the modern group, once on treatment with antipsychotic drugs, went on to develop diabetes at double the normally expected rate.⁴⁵ These figures show how ridiculous the Maudsley quote is, but for most doctors the anecdote, told repeatedly by a drug company, will continue to work.

FACTORY DOCTORS

In medicine up until the 1960s, if some doctor discovered a new disorder or described a body part, it was commonly named after him—so we have Bright's, Hashimoto's and Grave's diseases, Duchenne muscular dystrophy, Epstein-Barr virus, Nissl bodies, and Mullerian ducts.⁴⁶ It was the height of ambition for many male doctors to have their name linked in this way to a new observation or finding, and many of our most distinguished journals became well known because they published these new syndromes. Our brave new pharmaceutical era should be a golden one for doctors who have even a scrap of motivation in this direction.

Since the breakthrough with smoking and lung cancer, there is been a widespread recognition that many of our diseases are environmentally caused.⁴⁷ From chemicals in the air we breathe leading to respiratory disorders; to pesticides or other toxins leading to cancers, Parkinson's, and other neurological disorders; or aspects of the food we eat leading to cardiovascular disorders or gut cancers—across all these disorders lies the common factor of people ingesting some chemical or toxin.

But there is no other walk of life in which people ingest chemicals to the same extent, or in such concentrated forms, as happens among those treated in the modern practice of medicine. Unfortunate though it might be to put it in this way, medicine increasingly provides the perfect natural experiment to track the effects of poisons in the human body. Even if many of their colleagues disapproved, if the medical market were a free market working as Adam Smith described it, some doctors at least might be expected to hover around their patients like gulls around a trawler or robbers around a bank, because their career interests in reporting any new diseases and the concerns of patients at risk of suffering adverse reactions from treatment should coincide.

But this doesn't happen. It doesn't happen because companies rather than doctors run clinical trials. Moreover, companies sequester the data from those trials, so no one can use these data to determine what the hazards of treatment might be. Also, in their ghostwritten articles appearing in the very best journals, apparently authored by the most distinguished academics, companies have been able to interpret data pointing to marginal benefits from treatments as evidence that these treatments are remarkably effective, thus casting doubt on any remaining data that might point to the existence of hazards.

Locked into the distribution channel for prescription-only drugs, hemmed in by the science, doctors increasingly resemble the employees of the occupational health department of a factory that in the course of business exposes its workers to disability-inducing aerosols. These are doctors who are all too aware that their ongoing employment depends first on keeping mum about problems the workers may be having and second on being willing to recommend laying off workers at the first signs of any ill health—having persuaded them that they aren't fit for the job rather than conceding that job conditions might be the problem. The characteristic portrayal of a factory doctor like this, as opposed to the best of family doctors, lies precisely in the degree to which the one cares about patients and is prepared to speak out and the other can't.

The pressure on individual doctors in these situations can be forbidding, but we would still expect medical organizations to speak out about the hazards of treatment. Yet even after the FDA had finally conceded that antidepressants could cause suicide and required the drug companies involved to place warnings on their drugs to this effect, the American Psychiatric Association (APA) could still pose the question, “Do antidepressants increase the risk of suicide?” and answer it, “There is no evidence that antidepressants increase the risk of suicide,”⁴⁸ stating instead that “antidepressants save lives.”⁴⁹

At a time when healthcare is becoming standardized, when clinical prescribing is being constrained by guidelines, if drugs save lives, and hazards aren't worth mentioning, then why not let nurses—who are much less expensive—prescribe them? Nurses are much more likely to adhere to best practice, as embodied in guidelines. The APA statement amounts to a professional suicide note.

What does it take to back a profession into this kind of a corner? Some hints come from the kind of session held behind closed doors and without minutes that I was invited to attend in October 2005. It was convened by Britain's Royal College of Psychiatrists in the wake of the scandal over the suicide-inducing effects of antidepressants on children, to review whether the college needed more firewalls against pharmaceutical influence.⁵⁰ Most American medical associations have looked at their policies vis-à-vis industry in similar fashion in recent years.

The invited audience was small and included most of the officers of the college. The first half of the day was taken up by presentations about the pharmaceutical industry's view of its current position within the British economy—that half the day was handed over to industry was itself remarkable. These presentations claimed that the pharmaceutical industry was the largest funder of research in Britain, providing nearly 40 percent of all corporate monies in industrial research and development and 70 percent of Britain's medical research funding. Getting drug trials started in the UK was slow, however, we were told, and Britain was the most expensive place in Europe to do research. While the quality of research in Britain had been excellent, the rest of world was catching up, and clinical trials were moving to Russia, India, and China—a similar message to one delivered repeatedly to American academics for over a decade.⁵¹

There was a “we're all consenting adults” tone to the presentations. Company personnel made it clear that the pharmaceutical industry was a business whose bottom line was profit. The clinicians were told that if they had been to a sponsored meeting or seen a company representative they had worked with industry. And if they had read the scientific literature they would have been influenced by the pharmaceutical industry. The issue now, in the pharmaceutical industry's view, was how to put a framework in place to support the contribution industry could make to Britain's health and prosperity and enable medicine, the regulatory apparatus, and the pharmaceutical industry all to work in partnership.

The scandal of company suppression of clinical trial data showing that antidepressants made children suicidal had come to light the year before. Because of this, many of the clinicians present felt that however pharmaceutical companies dressed up their material, nothing about the claimed benefits to patients could be taken at face value. Some recognized the larger dilemma, namely, that we doctors were in the same business as the pharmaceutical industry. We never told patients when we felt industry's treatments weren't very good, because that would be saying that some of the treatments we were giving weren't very good. Were we professionals whose interests really diverge all that much from those of industry?

Industry was marketing drug solutions to medical problems—soon, how many among us physicians would know what a medical solution to a medical problem would look like or what our role vis-à-vis patients was other than to give drugs? Being told we should partner with the pharmaceutical industry was like telling a group of planners crafting a public transport policy that they had to partner with the automobile industry. Or someone charged with bringing peace that they had to partner with the arms industry.

Though the British rarely state anything plainly, these concerns among the attendees were expressed clearly enough. In response, industry took off its gloves. Padraig White, working for GlaxoSmithKline, told those assembled that current policies were in fact the result of the closest cooperation between industry and the most senior figures in British medicine. When it came to issues like our asking for the data from clinical trials, these notables would not be putting their knighthoods on the line to insist upon access to raw data from clinical trials. Within academic psychiatry alone, he went on, there were twenty-five senior figures in Britain each earning in excess of $200,000 a year out of links with industry, so academia was unlikely to side with their clinical colleagues. And finally, we clinicians all needed to note that 40 percent of British life insurance policies were invested in the pharmaceutical industry. Whether we liked it or not, we were shareholders in the industry, and rocking the boat would not be in our interest.

There was a time, not so long ago, when even senior figures from the pharmaceutical industry attending a meeting involving senior officers of a national medical association would have appeared grateful to be present and would have held their tongue when issues such as these were raised. But the balance of power and position had changed so much that a relatively junior official from the pharmaceutical industry felt able to tell a professional medical association to “Get real.”

In such a world, what scope is there for professionalism? How can clinicians “get real”? A start might be to shift the focus represented by statements like the American Psychiatric Association's “antidepressants save lives” to something more in the spirit of “doctors can save lives.” How would they do this? Well, if they have a good knowledge of the natural history of the conditions they treat and use agents of limited effectiveness and uncertain safety with care while enlisting their patients' support in monitoring the effects of any interventions, lives might then indeed be saved.

Industry monitors the prescribing habits of every doctor in the Western world. Data on who prescribes what is used by industry to shape tomorrow's sales pitch to produce an increase in prescriptions of their compounds or to determine why they are losing market share to one of their competitors.

In contrast, doctors record almost none of the problems that develop when they put patients on a drug. The FDA and other regulatory websites make it clear that doctors in the United States and Britain report only one in a hundred serious adverse events on drug treatment to the regulator. The postal service, in other words, does a vastly better job in tracking the fate of parcels than clinicians do in tracking their patients. Rather than becoming evidence-based, medicine is monitoring and quantifying exactly the wrong things. It quantifies hints that the latest snake oil has some benefits, helping drug companies smuggle their products through the regulations designed to prevent undue injuries to patients. But in the face of a growing series of premature deaths and other injuries induced by drug treatments, it does nothing.

THE RULE OF DOUBT

The question of whether Prozac was in fact causing serious problems unleashed on the FDA a set of public and media concerns to rival those that had broken around tolbutamide in 1970. In 1991 the agency organized public hearings on the issue, which gave rise to a dramatic clash between the personal histories of a range of patients and randomized trial data supplied by Lilly, the makers of Prozac. But in marked contrast to their handling of tolbutamide, the FDA introduced an extraordinary argument that would have brought howls of medical outrage in the 1970s but produced not a peep of protest in the 1990s.

If the FDA did the right thing and warned about risks, agency representatives argued, it might paradoxically increase the number of deaths. Even if some people became suicidal on Prozac, senior officials within the FDA and academics linked to Lilly claimed, their numbers would be smaller than the number of deaths likely to occur among people who because of the existence of a warning would be deterred from seeking treatment in the first place. Prominent warnings about risks of suicide, in other words, might lead to an increase in the number of deaths.⁵²

There is an astonishing ethical calculus here. In the case of the antidepressants, even the FDA had doubts about whether the drugs worked. There was absolutely no evidence they reduce suicides—all the clinical trial evidence pointed to an increase in risk, so much so that the major companies were engaged in various subterfuges to minimize public attention on the problem.

Given the completely predictable vigor of pharmaceutical company marketing that has led doctors to switch patients to the latest branded drugs, such as from ibuprofen or naproxyn to Vioxx or from hypoglycemics such as metformin to Rezulin and Avandia almost from the day they are launched, not warning patients appropriately in these cases amounts to a covert vaccination program—we want to see as many people on these new drugs as possible. But unlike vaccination, as the use of blockbuster drugs has spread, whether for osteoporosis, lowering blood sugar, or mood stabilization, the human cost that clinical trials of these drugs point to is increasingly paid by people who do not stand to benefit.

When it comes to the injuries produced by other industrial products, it is now clear that companies have been quite prepared to sponsor studies designed to bring in results that cast doubt on claims that chemicals from vinyl chloride to lead pose any health risks. Recent legal actions involving Paxil and birth defects have unearthed documents suggesting that lawyers for GlaxoSmithKline have liaised with academics to generate studies that could in similar fashion be used to counter any claims that Paxil might cause birth defects.⁵³ This suggests that tactics honed in the battles over industrial pollution in the 1950s are being deployed within medicine today.

The most striking example of company willingness to put profits before all else came at the end of the 1960s, when in response to concerns about smoking and lung cancer, the tobacco companies sponsored a series of papers showing that from 1900 to 1960 respiratory and cardiac deaths fell and life expectancy rose in tandem with increased tobacco consumption. This use of apparently scientific data (but misleading analysis) was part of a campaign to sow doubt about a link between smoking and disease, caught exquisitely in company admissions that “doubt is our product.”⁵⁴

But the ultimate use of the “doubt is our product” strategy faces us now in the form of company arguments that doubt about a hazard means it can be disregarded. As long as the data do not point to a statistically significant increase in heart attacks on Vioxx or deaths on statins or biphosphonates, the companies argue that doubt remains, and while doubt remains the drug must be regarded as being, in effect, free of hazards. The FDA and most doctors join them in this argument—to add warnings would, according to the American Psychiatric Association, have “a chilling effect on appropriate prescribing for patients.”⁵⁵

A PREEMPTIVE STRIKE

When it comes to other chemicals and the problems they pose, no company can buy up all doctors and hence they have to cope with the abilities of doctors to spot new hazards and their preparedness to speak out about these hazards. But when it comes to drugs, the interests of doctors, regulators, and companies are now so closely allied that pharmaceutical companies have been able to deploy an extraordinary legal maneuver.

When in 1991 the FDA talked about the public health consequences of deterring patients from treatment with antidepressants, it was a view aired in the course of a public debate. It wasn't public policy. The regulators and drug companies were still notionally managing the labels on drugs rather than determining the treatment of patients. A few years later a case in Southern California, the Motus case, made it clear that efforts were underway to shift public policy in the direction of actively supporting treatment with drugs.

In early November 1998, Victor Motus, a prominent member of the Filipino community in Southern California, had a few nights of poor sleep. He owned an architectural firm, was president of the local school district board, and had just launched his campaign for a seat on the Cerritos city council. He was due to go to Washington to receive an award from President Clinton for work done for the local school. With his wife, Flora, he owned numerous rental properties, an antique store, and he held an 80 percent ownership interest in two restaurants. Even without all this, a few nights of poor sleep were hardly surprising in a fifty-one-year-old man.⁵⁶

The only health problem Motus had was type 2 diabetes that he managed by a combination of diet and medication. When he developed problems sleeping, Flora made an appointment with a general practitioner, Dr. Trostler, on November 6, 1998 for her husband to obtain sleeping pills. During the visit, mention was made of financial problems with one of his restaurants, and even though there was no prior medical history to point to depression and little wrong clinically, educated by companies like Pfizer to think that sleep difficulties commonly pointed to a depressive disorder, in which case antidepressants would be a more appropriate treatment than hypnotics, Trostler gave Victor a sample pack of the antidepressant Zoloft—something unlikely to help his sleep. The Motuses were told that Zoloft might take several weeks to work.

Victor took Zoloft as instructed over the next six days. His family later said that for the first day he thought the drug was helping him. By the third day, he told his sister the drug didn't suit him, and his wife noticed he was pacing and sleeping even less at night. On the fourth day, he had become shaky. On the fifth day, his birthday, he told Flora “I don't feel like myself,” “the drug is making me crazy,” and “I want to kill myself.” She urged him to continue taking the drug because the doctor had told them that it might not work for another week.

He was due to fly to Washington the next day, November 12, the sixth day of treatment. When his brother arrived to take him to the airport, Victor was not there. He was found in his car several blocks away, dead behind the wheel from a single gun shot to his head.

In 2001, Flora took legal action against the makers of Zoloft, alleging that Pfizer “negligently…failed to adequately warn the medical community, the general public and (her husband)…of the dangers, contraindications and side effects…of Zoloft.” Malcolm Wheeler, the lead attorney for Pfizer, contacted the new chief counsel of the FDA, Daniel Troy. Troy had joined the Department of Justice the year before, from a legal office that had done over $358,000 worth of business with Pfizer that year. Troy filed an amicus brief in the Motus case. In order to file such a brief, he had to have clearance from the solicitor general in the George W. Bush government, Paul Clements, a former partner in King and Spalding, the main law firm representing GlaxoSmithKline, who were also faced with a series of suicide cases involving an antidepressant, in this instance Paxil.

Troy's brief argued that the California court had no jurisdiction in the case. The FDA is responsible for the labeling of drugs and if Pfizer had warned that the drug could cause people to become suicidal it would have broken the law. Troy was firing a shot across the court's bow to ensure it did not do anything that would “undermine the agency's authority to protect the public health.”⁵⁷

When considering warnings, according to Troy, the FDA has to take into account the risks posed by the untreated illness—depression. “Under-utilization of a drug based on dissemination of scientifically unsubstantiated warnings, so as to deprive patients of beneficial, possibly lifesaving treatment could well frustrate the purpose of federal regulation.”⁵⁸

This was an extraordinary and unprecedented move. This was not the conventional company argument that juries cannot possibly decide on the science of an issue such as whether a drug might cause suicide. Troy was arguing that a jury had no place in considering whether the FDA should have asked for warnings or not—and therefore the company could not be prosecuted provided it had followed the directives the FDA had given it, regardless of whatever effect the drug had or had not had on Victor Motus. This was even though, as federal law stands, companies are obliged to warn if there are hints of a problem—not just on the basis of proof of a problem.

The idea of using a federal agency to preempt legal action in a state court was Malcolm Wheeler's brainchild. He had used it first when defending the Honda Motor Company in a 1980 case in which a plaintiff argued the company should have installed airbags even though federal regulations didn't at the time require them.⁵⁹ This case went all the way to the Supreme Court who, noting that the car contained many other safety features such as seatbelts, sided with Honda.

In the case of cars and butter as outlined earlier, it is not the job of the regulator to even begin to think about whether people should be driving cars or using butter. In the Motus case, however, the Pfizer argument was that the FDA should be thinking not just about whether drugs would be used but also attempt to ensure that they would be used, and that in this light, warnings might put a chill on this use. This opens up an extraordinary vista—and certainly not the vista Senator Kefauver had in sight when sponsoring the 1962 amendments to the Food and Drugs Act.

When I turn the ignition key to start a car, wires have to connect to a starter motor and there has to be a flow of both oxygen and gasoline for the act of driving to become possible. For a physician to practice medicine, having a drug that works is helpful but often not essential. From time immemorial, and certainly as embodied in Pinel's famous aphorism about the art of medicine, good practice has been held to be much more likely where there is detailed knowledge about the hazards of any drug used. Semiautomatic prescribing has never been regarded as good practice.

In arguing about the merits of safety warnings, Wheeler and Troy portrayed the issue as being the equivalent of having airbags in a car, when in fact they were arguing against a need to have the ignition wires properly connected to the starter motor or letting people know that, in some instances, the ignition wires might default to the gas tank (an instance of immediate-onset side effects), or, in a case of lateronset adverse effects, faulty brake linings that would give way after six months. Safe starting connections and functioning brakes have never been optional extras to a car in the way airbags once were; they are integral to its basic operation.

Before there was a ruling on preemption in the Motus case, Pfizer in fact got the case thrown out, but on the basis of another astonishing defense—the prescription-only status of Zoloft that Senator Kefauver unwittingly cemented in place in 1962. In a pretrial deposition, Dr. Trostler testified that no matter what the warnings were on Zoloft, he would have gone ahead and prescribed the drug anyway. If a doctor testifies as Trostler did that he would have used the drug in a similar fashion whether there were warnings or not on it, any case against a pharmaceutical company on the basis of failure to warn about the risks of a drug that is available only by prescription collapses. The reason to make them available by prescription only was precisely this—to interpose a professional judgment between the patient and the pharmaceutical company. As a majority of his colleagues at the time also held the view that SSRIs posed no risk, Trostler was not at risk of a malpractice suit. Victor Motus, like almost everyone who dies unnecessarily on a prescription-only medicine, was the victim of a perpetrator-less crime.

The service that pharmaceutical companies get from primary care physicians or other medical doctors, in other words, goes well beyond what most industries get from their onsite doctors. Not only can companies typically depend on doctors not to rock the boat, but they can palm off any legal responsibility for injuries caused by exposure to chemicals onto the doctors who do the prescribing. Complain about or investigate a problem and the doctor knows he is in for a rough ride. But faced with a problem that turns up later, the company gets off scot-free because doctors refuse to accept that the chemicals have caused the problem.

DESCENT INTO THE UNDERWORLD

Before his doctor abandoned him, Victor Motus's case led to an unearthing of a series of documents about how Pfizer had transformed clinical trials into marketing exercises, recoded patient details where the original data were inconvenient, and planted ghostwritten articles in all major journals. By focusing attention on the role of ghostwriting and company manipulations of their data it set a template for later legal actions surrounding Vioxx and other drugs.

The Motus case also led directly to evidence of children becoming suicidal on Zoloft, ultimately triggering, in 2004, two further sets of public hearings on antidepressants and suicide in children. At the second of these FDA hearings, Tom Laughren, head of the Central Nervous System division at the FDA, presented the clinical trial data and the FDA's views of the data.

Members of the public were then given three-minute slots to present views—seventy-three of them. Among these, a series of doctors made presentations—almost all male and all warning against the issuing of warnings. These were intermingled with a series of mothers, a modernday set of Demeters. Demeter was the Greek goddess of the Earth and of fertility whose daughter, Cora, was forcibly abducted and carried off to the underworld by Hades. Demeter protested to Zeus, who professed himself helpless, until Demeter threatened Earth with permanent winter. Zeus intervened and restored Cora to her mother as Persephone. Because Persephone had eaten some pomegranate seeds while in the underworld, however, she was obliged to return to Hades for the several months of winter each year.

These modern-day Demeters similarly aimed at shaming the heavens to take action. In a presentation with mythic resonance, in the 72^nd slot Mathy Downing confronted Tom Laughren:

On January 10th, 2004 our beautiful little girl, Candace, died by hanging four days after ingesting 100 mg of Zoloft. She was 12 years old. The autopsy report indicated that Zoloft was present in her system. We had no warning that this would happen. This was not a child who had ever been depressed or had suicidal ideation. She was a happy little girl and a friend to everyone. She had been prescribed Zoloft for generalized anxiety disorder, by a qualified child psychiatrist, which manifested in school anxiety…. She had the full support of a loving, caring, functional family and a nurturing school environment.

Her death not only affected us but rocked our community…. When Candace died her school was closed for the day of her memorial service, a service that had to be held in the school gym in order to seat the thousand or so people who attended. How ironic, Dr. Laughren, that your family attended Candace's memorial service. Our daughters had been in class together since kindergarten. How devastating to us that your daughter will graduate from the school that they both attended for the past eight years and that Candace will never have the opportunity to do so.

Candace's death was entirely avoidable, had we been given appropriate warnings and implications of the possible effects of Zoloft. It should have been our choice to make and not yours. We are not comforted by the insensitive comments of a corrupt and uncaring FDA or pharmaceutical benefactors such as Pfizer who sit in their ivory towers, passing judgments on the lives and deaths of so many innocent children. The blood of these children is on your hands. To continue to blame the victim rather than the drug is wrong. To make such blatant statements that depressed children run the risk of becoming suicidal does not fit the profile of our little girl.⁶⁰

A few minutes earlier, Mary Ellen Winter had also confronted Laughren and the FDA about her twenty-three-year-old daughter, Beth, and brought a new player into the frame:

Beth was looking forward to a career in communication and was experiencing some anxiety and having trouble sleeping when she consulted our family physician. He prescribed Paxil and said she would start feeling better in two weeks. Seven days later Beth took her own life.

We, like most of you in this room, grew up with confidence in the strides made in medicine and accepted with faith antibiotics and vaccinations prescribed. We believed the FDA would always act to protect our family's well being. When my daughter went to our family GP last year, we trusted that our doctor was well educated and informed. We were wrong. We now know that pharmaceutical sales are a high stake business, driven to increase shareholder wealth. The consolidation of pharmaceutical companies like GlaxoSmithKline has resulted in increased sophistication in the quest to market and distribute pharmaceutical products. Priority has moved from health to profit. Not all doctors are equipped to understand the marketing targets they have become. The FDA has allowed our daughter to be the victim of a highly commercial enterprise that selectively releases clinical data to maximize sales efforts and seeks only to gain corporate profits….

As residents of the State of New York, we thank our Attorney General, Elliot Spitzer, for addressing issues that the FDA has been unwilling to address…. ⁶¹

A few weeks before, the attorney general's office in New York had opened a fraud action against GlaxoSmithKline. The brief had been prepared by the unlikeliest of heroines, Rose Firestein. Firestein was blind and in poor health, but she had a passionate concern for the abuse of children. She charged GlaxoSmithKline with fraudulent interference with the practice of medicine. The company had engineered sales of Paxil for use in treating children who were depressed through the publication of ghostwritten papers that had concealed the drug's hazards and portrayed the outcome of studies, like Study 329, as positive even though this and other clinical trials had failed to show the drug worked. Her trump card was the internal GlaxoSmithKline document I had made public at an FDA hearing on pediatric suicidality on antidepressants six months earlier that showed the company deliberately setting out to present the good bits of the evidence and suppress the rest, and to publish papers in influential journals that would lead doctors to prescribe a drug that the company had already internally recognized did not work.

When faced by GSK's lawyers with the argument Troy made in the Motus case, that the FDA's views preempted any views a state might have, Firestein responded that New York was not attempting to argue the toss with the FDA on the labeling of the drug. When it came to the practice of medicine in the state of New York, however, the state and not the FDA had primacy. GlaxoSmithKline settled, and as part of the settlement posted on their website details of all of their clinical trials.

Ultimately, Mary Ellen Winter, Mathy Downing, and others who testified played a part in forcing the FDA to bow and issue warnings that antidepressants might lead to suicide in children, warnings that were extended to adults in 2006. These mothers did not win because their pleas were impassioned. They won because the FDA, although still giving every sign of willingness to ignore the nineteenth bullet in the twenty-bullet gun barrel, faced with nineteen bullets there could see no way out of issuing warnings. The FDA have since gone further and issued suicide warnings for anticonvulsants and antipsychotics, as well as Champix and Zyban for smoking cessation, Roaccutane for acne, Tamiflu for flu, and Singulair for asthma, and have withdrawn from the market Rimonabant for weight loss and Cymbalta/Yentreve for urinary incontinence for this reason.

Legal actions against pharmaceutical companies are almost nonexistent outside the United States. In part this is because patients in Europe and other countries have universal healthcare and there is much less or no need to recover the costs of treatment for an injury from a pharmaceutical company. A great deal of what the world knows therefore about how pharmaceutical companies do business—how they market drugs, ghostwrite articles, ensure trials recruit just the right number of patients so that the results for serious hazards cannot become statistically significant—stems from legal actions in the United States.

The attempted preemptive strike engineered by Malcolm Wheeler and Dan Troy in the Motus case—that of claiming that federal regulatory agency statutes preempted any action, state or federal, against a pharmaceutical company—aimed at removing one of the few remaining impediments in the world to company abilities to do business. Although Pfizer won the Motus case by other means, other companies began to use preemption with increasing frequency, and success. A series of SSRI-induced suicides, Vioxx-induced heart attacks, and Avandia- and Rezulin-induced deaths were thrown out on this basis.

As the argument raised was legally unprecedented, it was taken all the way to the Supreme Court.⁶² When preemption had first been raised in 1998, it had seemed inconceivable to most legal minds that such a defense might succeed as a matter of principle, but a decade of company arguments and lobbying had made a difference. In attempting to come to a resolution, the Supreme Court scheduled two cases. The first involved Pfizer's Rezulin, which had been approved in 1997 as a treatment to lower blood sugar and then withdrawn from the market in 2000 after being linked with an excess of deaths from liver failure. In this case in 2008, the nine justices tied 4 to 4, Chief Justice Rehnquist having recused himself on the basis that he held shares in Pfizer.⁶³

A second case heard in October 2008 involved phenergan, a Wyeth drug, given for nausea, which when given by injection caused an arterial spasm that led to Diana Levine losing her right arm. At the heart of the case was the issue of whether this loss stemmed from a lack of warnings regarding its proper administration.⁶⁴ The issue was not whether Wyeth knew about the hazard, but rather whether the FDA had or had not specifically instructed the company to warn about the hazard and if not, whether this exculpated Wyeth—even if the company did know about the hazard. On March 4, 2009, the Court came to a 6–3 verdict against Wyeth and against preemption.⁶⁵

For the moment therefore, the kind of documents on which Firestein based her case in New York may still occasionally come to light in actions taken by plaintiffs against drug companies. For the moment, companies remain somewhat limited in their abilities to interfere with the practice of medicine in whatever way they see fit. For the moment, the FDA has not unequivocally become the de facto regulator of medicine rather than just medicines.

As this was happening another drama was unfolding. As part of their settlement with New York State, GlaxoSmithKline agreed to post details of all their clinical trials. This sounds better than it is: the company only posted internal study reports, not the actual data. These differ little from ghostwritten articles. They do not let anyone establish, for instance, whether a dropout from a trial was really a dropout for noncompliance or whether the child had attempted to commit suicide.

But it is difficult to hide deaths. In 2007, Steve Nissen, a cardiologist at the Cleveland Clinic, combed through the data from trials on Avandia (rosiglitazone), GlaxoSmithKline's latest blockbuster, a successor to tolbutamide in the treatment of adult-onset diabetes, and found that, while no one trial of Avandia showed an excess of cardiac deaths, in all trials combined the drug was linked to a marked increase in such deaths.⁶⁶

The Avandia story repeats all the elements of drug disasters from tolbutamide to Paxil. When first licensed by the FDA in 1999, there was no evidence that it saved lives, but influential academics lobbied regulators and other decision makers to ensure favorable treatment for this must-have drug. The American Diabetes Association came out strongly in favor of the drug even before it was approved.⁶⁷ Over the next five years, Avandia moved rapidly to blockbuster status despite warnings by the World Health Organization (WHO) in 2004 of an increased risk of death. Internal GlaxoSmithKline analyses also found increased—but nonsignificant—rates of cardiac events on Avandia in their clinical trials, but in public the company continued to argue that the drug was absolutely safe and sales continued to boom. In the face of Nissen's analysis, the company fast-tracked a major study of cardiac safety on the drug, the RECORD study, and published it claiming that it demonstrated that Avandia was safe.⁶⁸ Both the FDA and European regulators agreed. But mounting concerns led to hearings at the FDA, and further investigations by FDA staff, one of whom, Thomas Marciniak, found a pattern of concealment of serious events on Avandia in the RECORD study that echoed the hiding of deaths and other serious events on SSRIs and Vioxx.⁶⁹

These events led a Senate committee to investigate GlaxoSmithKline and issue a damning report.⁷⁰ Legal cases have been instituted. These reveal extensive ghostwriting of the medical literature as part of the marketing of Avandia, comparable to that found in the marketing of Hormone Replacement Therapy (HRT)⁷¹ or oral contraceptives such as Yaz. These cases reveal that the pattern of using evidence-based medicine to drive prescribing is now all but universal. Apparently rational prescribing in such circumstances is close to certain irrationality.

Zeus responded to Demeter that he was helpless to do anything about her daughter. Once, if our daughters were threatened by Hades, doctors from Philippe Pinel two centuries ago, to Alfred Worcester and Richard Cabot a century ago, and Dr. Lapin who looked after my father a quarter of a century ago, might have intervened on our behalf, but for the moment all we have standing between us and Hades is a small group, comprised of mothers and wives like Mathy Downing and Mary Ellen Winter and a dwindling number of Lapins, Worcesters, and Pinels. It is these who do the job of alerting us to the hazards of treatment, a job that one might have thought lay at the heart of good medical care. How long they can hold on before the Heavens answer is not clear.