A Cure Worse than the Disease?

Revisiting the perils of Prozac.

You want not to believe this book. The story David Healy tells is nasty and frightening to contemplate. John Le Carré, whose novel, The Perfect Gardener, centres on wrongdoing in the pharmaceutical industry, is quoted on the back cover of Let Them Eat Prozac: “This very important book will demonstrate beyond your worst dreams that the commercial needs of BigPharma are the natural-born enemy of independent scientific research.”

Healy is a psychiatrist with appointments at both the University of Wales and, on a visiting basis, the University of Toronto. He has published more than a dozen books and 120 peer-reviewed articles. His past credentials include consulting to major pharmaceutical companies and prescribing Prozac and related antidepressants to his patients. A somewhat recent addition to his curriculum vitae is whistle-blower in relation to his deep concerns about these same drugs. It is those concerns that he documents in Let Them Eat Prozac.

It is not the first time that Healy has courted controversy by going public with his opinions about Prozac and similar drugs. In 1999–2000, Healy was ardently pursued by the Centre for Addiction and Mental Health, a teaching hospital affiliated with the University of Toronto. The Centre made a written offer of employment that included the promise of a full professorship. Healy then gave a lecture at the Centre, and discussed his already-published concerns about suicidal tendencies associated with Prozac and the failure of drug companies to research the risks of their drugs properly and to publish unfavourable scientific evidence. His job offer was promptly rescinded, although the Centre insisted that it was not because Eli Lilly (the manufacturer of Prozac) was its major donor. A heated debate on academic freedom and the real reasons for the Centre’s change of heart ensued in the press, and Healy and the Centre finally settled in 2002. His current visiting fellowship at the University of Toronto is part of that settlement.

In his book, Healy basically makes three grave and worrying allegations. First, he states “that the Prozac drug group could trigger suicide and violence and that companies producing these drugs knew of the problem” and suppressed it. Indeed, they undertook public relations campaigns to expand the market for Prozac. Second, he says that drug regulators failed to protect the public. And third, he points out that researchers of new drugs often find themselves in serious conflicts of interest.

Concealing the Problem

Healy writes that as Eli Lilly learned that its blockbuster antidepressant drug Prozac could elicit suicidal thoughts or even murder-suicide, the evidence was either denied, concealed or “massaged.” The company’s motivation was largely commercial, perhaps even to avoid a serious financial threat to its continued existence. A memo written by Leigh Thompson, Lilly’s chief scientist, shows the potential extent and gravity of such concerns. It warned that “Lilly could go down the tubes if we lose Prozac and just one event … can cost us that.”

In terms of hiding risks, Healy compares the psychopharmacology industry to the tobacco industry. We are, with good reason, cynical about truth-telling by the latter. Healy’s description of the Wesbecker case raises the question of whether we should be equally cynical about some drug companies. Joseph Wesbecker worked at a printing plant in Louisville, Kentucky. While on Prozac, he walked into the plant with an AK47, killing eight people and severely wounding twelve, before killing himself. Wesbecker had no previous known history of violence. Healy argues convincingly that Lilly knew—or ought to have known—that Prozac increases the risk of suicide. Yet the company’s lawyers consistently denied that Prozac could have had any role in causing Wesbecker’s murder-suicide, or in causing suicide in anyone else.

Rather, Lilly responded to allegations that Prozac could induce suicide with a concerted public relations campaign. It took the line that “first, Prozac was the most researched drug in history. Second, the problem was the disease the drug was meant to treat, not the drug itself. And third, the real tragedy of the Prozac story was all the people who would commit suicide because they were being denied access to an effective treatment.”

Healy describes how sophisticated legal advice and lawsuits became a weapon in keeping Prozac on the market, more than just being a shield against liability. Information shared with a lawyer in contemplation of litigation is subject to an absolute privilege against disclosure—a rule that would seem, from an ethical (although not technically legal) perspective, to have been abused. Documents demonstrating the risks of selective serotonin re-uptake inhibitors (SSRIs) including Prozac may have been kept from public scrutiny by routing them through the company’s lawyers to allow the documents to be withheld on the basis of attorney-client privilege.

Research trials are used to identify and assess harms and risks. How they are designed and how their results are interpreted and reported must also be governed by ethics. But “nobody knew Lilly had put patients in its clinical trials on some … other medications [that might have reduced any suicide-inducing effect of Prozac] to minimize Prozac-induced problems.” Two psychiatrists, Maurizio Fava and Jerrold Rosenbaum, analyzed data on suicide in people receiving drug treatment for depression and concluded that suicidal thoughts were no more likely to emerge on Prozac than on other antidepressants. In contrast, Healy claims that a re-analysis of the same Prozac trial data indicates that “Prozac was approximately three times more likely to be associated with the emergence of suicidality than other treatments” for depression. Healy says Lilly uses this information still, without any acknowledgement that it could quite possibly be interpreted as support for the argument that Prozac induces suicidality. In the same vein, trials of SSRIs other than Prozac, also suspected of augmenting the risk of suicide, were “re-analyzed”—that is, the findings were massaged to reduce the assessed risk of their being linked to suicide.

Healy concludes, as well, that some medical journal articles are startlingly inconsistent with the raw data with respect to risks. He also explains that many articles, purportedly written by eminent researchers, are actually ghostwritten and the researchers may not have verified the raw data. And at least one article claiming that Prozac did not increase the risk of suicide was “authored wholly by the manufacturers’ employees.” Healy rhetorically claims “that some of the most eminent figures in the academic health science research community appear party to the suppression of data on lethal side effects of treatment and that such behaviour has become something of a norm.” If accurate, this is a profoundly worrying situation and requires immediate investigation into whether similar instances are still occurring. If so, urgent remedial measures must be put in place. If not, the public needs to be reassured.

Were a research ethics review committee to hear concerns about the safety of a drug such as those Healy articulates about Prozac, it would, at the very least, require in-depth exploration of them. The vast majority of research ethics committees do not, however, continue to monitor the research they approve. Usually they have no resources to do so. Such concerns would also require drug monitoring committees or agencies to act responsibly and consider whether specific warnings about a drug should be given to physicians and patients, or even whether it should be withdrawn from the market. No such safeguards were implemented in relation to Prozac. That leads to Healy’s second major point: that drug regulators failed to protect the public.

Drug Safety Regulation

The pharmaceutical industry and drug regulation, as we know them today, are mid 20th-century phenomena. Before 1960, the pharmaceutical industry was a small enterprise, consisting of minor divisions within large chemical companies. Today, it is a huge, transnational entity. In the 1950s, new disorders—drug-induced conditions—were recognized. It was because of these adverse effects that “prescription only” arrangements were applied to all new treatments by many countries, including the United States through the Food and Drug Administration and Canada through the Health Protection and Promotion Branch, which administers the Food and Drugs Act and Regulations. Prior to that, drugs could be marketed in the same way as mouse traps—companies were only liable ex post facto for manufacturing or selling unreasonably unsafe products.

Healy says drug regulators failed in their duty to protect citizens in relation to Prozac. Here are some of the reasons he gives for this failure. Many people “who advise regulators on the entry of drugs into the market also advise the major pharmaceutical companies,” he writes. “There are vanishingly few independent investigators left in psycho-pharmacology.” The vast majority are involved with drug companies. Healy believes that industry funding for research is another of the problems. Randomized controlled trials and epidemiology are regarded by regulatory authorities, such as the FDA, as the only way to prove cause and effect, but the companies are the only ones able to afford to conduct such studies. Healy refers to this as the power of the pharmaceutical industry to “buy” the scientific agenda.

In short, only industry funds the research required to obtain marketing approval for new drugs, and in the case of SSRIs it favoured researchers “who managed to produce the right results.” To obtain such results, some investigators reported on patients who did not exist or on so-called “professional” patients who might have been taking more than one investigational drug. And “a large number of trials with less favourable results for SSRIs were simply not reported.” Not only do companies fail to tell regulators about trials that are not published, but they also have the power to block the publication of articles that clash with marketing interests. When factors such as these mar the trials, the scientific validity of the results is cast into doubt. These factors also involve serious breaches of ethics, but these may not be identified because even the ethics watchdogs may be compromised by the increasing privatization of ethics review bodies.

One crucial question in deciding which drugs should be approved, and which rejected, is which risks count. Let’s look at that, as Healy has done, in relation to Prozac.

One of the advantages of Prozac, when it was initially released, was that it was not lethal in an overdose, as were almost all other antidepressants available at the time. When discussing the safety of Prozac with regulators such as the FDA, Lilly used this fact as a key argument. Since regulators face potential legal action if they let a drug with known hazards onto the market, they are risk-averse and practice “defensive regulation.” That means avoiding known risks—such as the possibility of lethal overdoses with some antidepressants other than Prozac—but taking unknown ones—Prozac’s potential to precipitate suicide in some people. Decision making in such situations calls for “virtues ethics,” that is, disinterested, but not uninterested, courageous moral actors without any conflict of interest, who decide which risks can justifiably be run.

In addition to deciding which risks can or cannot be run, careful decision making is needed to make sure that the right risks are being investigated. The precipitating condition for suicidal thoughts from the SSRIs is akathisia, literally an inability to sit still—severe agitation. No one thought this was depression, and to this day this phenomenon “remains almost uninvestigated.” Healy asks whether “a lack of research” on important questions that affect our health and wellbeing might be a direct effect of industry funding practices that led to an absence of independent research on drugs such as Prozac.

Healy notes that once a drug is on the market, regulators such as the FDA are fairly powerless. That need not be the case, however, and the public would be shocked to know that it is. After Prozac was approved for marketing in the United States, it was stated that further research on it was needed. However, no research of the kind required was ever undertaken. Ongoing research should be one of the ethical requirements of marketing a new drug, especially if there are any doubts about serious risks. Suicide equals death, which is the most serious of risks. There need be only a small probability of such a risk for it to be material; even a remote doubt should make ongoing, post-approval monitoring mandatory.

Conflicts of Interest

Let Them Eat Prozac could well be subtitled “A Study of Conflicts of Interest.” Many people misunderstand the true nature of such a conflict. It does not require any wrongdoing. It is present merely if a person could be subject to conflicting obligations. For example, an advisor to a regulatory authority is in a conflict of interest if that person’s stake in the financial well-being of the manufacturer could conflict with the duty to warn regulatory authorities of the harms or risks of a drug.

That raises the increasingly ethically complex issue of the interface between big business and big science. More and more often, academic researchers find themselves in the position of serving both. The now well-known case of Nancy Olivieri and her struggle with the generic pharmaceutical company Apotex, which was funding her research, dramatically illustrates what can happen when a physician-researcher puts her ethical and legal obligations to her child-patient research participants above the financial interests of a company, as it perceives them. Olivieri’s case is documented in voluminous detail and reads like a medical research ethics horror story. It is a case that lends credence to the breaches of ethics described by Healy on the Prozac file.

Creating a Disease

Healy explains that before the development of antidepressants only about 50 to 100 people per million were thought to suffer from depression. Current estimates put that figure at 100,000 to 200,000 affected people per million. This is a thousand-fold increase, despite treatment supposed to cure this terrible affliction. If the treatment of depression is so good, why has the frequency of the disease apparently jumped so much since the introduction of the antidepressant? Moreover, there is no evidence that mass detection and treatment of depression have made a difference to national suicide or disability rates.

Creating a disease for which a treatment has been discovered is a marketing strategy. Paxil, another SSRI, was promoted as the pill for “social phobia,” later renamed social anxiety disorder, long recognized as the common human trait—or should we say affliction?—of shyness. This approach raises a host of ethical issues, for instance, the pathologization and medicalization of normal human conditions and the tendency to trade on people’s desire for a quick technological fix for all their problems. It also sounds a warning with respect to some drugs used today, such as Ritalin. A large number of boys have been placed on Ritalin or similar drugs to treat attention deficit hyperactivity disorder. But is that condition really just normal behaviour redefined as a disease? Are such treatments “chemical straitjackets” or justified medical interventions? Our answer to that question will radically alter our perception of the ethics of putting children on such drugs.

Healy says that “the Prozac story brings interlinked problems to light, among them a wholesale creation of depression on so extraordinary and unwarranted a scale as to raise grave questions whether pharmaceutical and other health care companies are more wedded to making profits from health than contributing to it.” People who would not previously have been diagnosed as being depressed now are, as a result of drug companies’ “education” of physicians.

What Can Be Done?

What we can do about the broad range of concerns that Healy’s book identifies is canvassed in great detail in another very recent book, Responsible Research: A Systems Approach to Protecting Research Participants, published by the Institute of Medicine in the United States and edited by Daniel Federman, Kathi Hanna and Laura Lyman Rodriguez. In a much less accusatory and sensational way than Healy, the Institute’s findings and recommendations for change reflect much the same realities and concerns as those he describes. Primary among their recommendations are the following:

• Open all data to public scrutiny. Healy also proposes that “we need a new contract between society and the pharmaceutical industry—a contract that will require access to the raw data that result when people serve as research subjects and in doing so take risks in studies for pharmaceutical companies.”

• Recognize the rights and contributions of research participants (people who serve as research subjects).

• Emphasize the overriding and ongoing responsibility of ethics review boards to protect participants.

• Understand that all people involved in research have ethical obligations that are nondelegable, that is, they are responsible for ensuring research ethics and are liable for any breach.

• Implement the principle that ensuring ethics in research is a complex undertaking that requires education, adequate support and research.

In many ways, Let Them Eat Prozac and Responsible Research are two sides of the same coin. The former details some of the very serious ethical failings that have recently afflicted medical research and the marketing of drugs. The latter sets out a blueprint for overcoming those failings, in particular by seeing the medical research enterprise as a complex system that needs ethics in relation to every level, institution and person engaged in it. Implementing this blueprint is imperative, because stopping medical research is not an option. It must continue, and we should all hope that it will allow us to enjoy even more of the sometimes extraordinary benefits it provides in terms of reducing suffering and maintaining health.

Conclusion

Pharmaceutical companies must be able to justify, in ethical terms, placing their drugs on the market and promoting their use. The pharmaceutical industry is not an industry like any other. Because its business involves the health and lives of people, it has a contract with society. Its responsibilities are far greater and more serious than those of companies who make, for instance, luxury goods or mousetraps. There is increasing and deep disquiet about the current state of affairs in medical research. BigPharma is a major player and will remain so; it should strive to be an ethical leader, by taking as its own the Hippocratic maxim primum non nocere—first do no harm.