So when Schildkraut suggested that even if the catecholamine hypothesis was simplistic, it was still a worthy frame of reference, it didn’t occur to his colleagues to complain that he was assuming his conclusions and making vast claims not only about depression but about humanity in the bargain, or that they should do anything at all but rush headlong to the next beachhead. Indeed, “The Catecholamine Hypothesis” quickly became one of the most-cited papers in the medical literature. And even as the catecholamine hypothesis, which had replaced the serotonin hypothesis, gave way in the 1980s to a new serotonin hypothesis, which itself was replaced a couple of decades later by other hypotheses about other neurotransmitters, even as drug company scientists have packed their tents and rushed to successive new fronts, even as the depression doctors have confidently told their patients about the molecule, whatever it is, that is the source of their woes, the idea itself—that depression is caused by chemical imbalances—has only gathered strength.
Because that was the point all along: to find something inside your head on which doctors could work their magic and relieve you of your unhappiness. By the time Schildkraut wrote his paper, that wish had been building for nearly a century, since Paul Ehrlich first set medicine on the course of manufacturing bullets. Kraepelin’s notion that mental illnesses would be found in nature, Adolf Meyer’s idea that we all could have them, the shock therapists’ discovery that they could relieve mental illnesses by doing something to the body, Gaddum’s speculation that the problem was in the chemicals of the brain, Kuhn’s assertion that imipramine had shown those chemicals at work in one of those diseases—all of these ideas came together in Schildkraut’s neat and irresistible package. The fact that even he knew that it wasn’t exactly true—and that scientists would soon prove it false—couldn’t stop his idea from taking hold. It was a juggernaut, the culmination of decades of yearning for a way to set people free from their psychic afflictions, a way to comfort the Jobs of the world without accusing them of sin or forcing them to reckon with a whirlwind.
Of course the depression doctors were exuberant. Anyone would be. But science and enthusiasm are an uneasy pair. No scientist labors at his bench without the wish that his work will pay off, and no good scientist allows his excitement to run roughshod over the facts. That’s what the scientific method is for—to rain on the researcher’s parade—and that’s why we believe its results: because they are the facts pure and simple, not subject to the distortions of credulity or ambition.
Or so we hope. But even as the drug makers were gearing up to treat depression, the method for determining whether drugs worked was still under construction. And the solution that the Food and Drug Administration finally settled on could not have been better for the industry. Soon enough, the United States government had not only given its imprimatur to the drugs, it had also created a huge incentive to invent a disease for them to treat—a disease that would also receive the government’s stamp of approval.
CHAPTER 10
DOUBLE BLIND
Every two weeks, the research assistant at Mass General, a tall young woman fresh out of college, handed me a new supply of pills. They came in a brown paper bag, the kind you carry your lunch in, its top neatly folded over. Inside were two plastic bottles stuffed to the brim with fat golden capsules. The bottles had childproof caps and two labels: the standard prescription label telling me to take four from one bottle and two from the other every day and another, smaller one which read DRUG LIMITED BY FEDERAL LAW TO INVESTIGATIONAL USE.
That wasn’t exactly true. In fact, I could have bought my “drug” at the Whole Foods Market right next to the offices of the Depression and Clinical Research Group, and I could have taken it, as many people do for many reasons, as a supplement. (For that matter, so long as I didn’t mind a big dose of mercury with my omega-3s, I could have just eaten a lot of salmon.) Taking fish oil without all the medical folderol would not have felt anywhere near so momentous, and it certainly couldn’t have contributed to science, but going the commercial route would have had at least one major advantage: I would have known what I was getting.
Specifically, I would have known whether my pills actually contained omega-3s. They might have been placebos—in this case, soybean oil. No one involved in the experiment knew which was the case, except the pharmacist who filled the bottles according to the condition I’d been assigned by a random number generator. This ignorance was intentional and crucial to the business at hand. Indeed, the double-blind, placebo-controlled, randomized clinical trial (RCT), the kind of experiment I participated in, is known in the industry as the gold standard for investigating the efficacy of drugs. RCTs are the tool that the FDA uses to determine the validity of most drug treatments, including antidepressants.
The RCT is a simple method, or so it seems at first. My first day of testing established the baseline of my depression—eighteen points on the HAM-D, which is considered the threshold for major depression. On subsequent visits, I took the same tests. The score from my last day was then compared to the baseline. (Intermediate scores help to verify that the final score is not an outlier, a score whose divergence from the others makes it suspect.) After two hundred people are recruited and run through this procedure—which investigators reckon will take two to three years—the study will be unblinded, the subjects sorted into a treatment group and a placebo (or control) group, and the aggregate differences between their baseline and endpoint scores will be computed. If the treatment group shows a greater decrease in symptom severity than the placebo group, then that will indicate that omega-3s are effective antidepressants. The mutual ignorance of researcher and subject about who is in which group will increase confidence that the difference between groups reflects not persuasion or gullibility or hope or some other artifact of human credulity but something real: the drug and not just the pill.
Prior to the advent of scientific medicine in the mid-nineteenth century, credulity was the active ingredient in most cures. Aside from aspirin, quinine, morphine, digitalis, and a handful of other drugs, most of what doctors handed out was hokum, except for the stuff like mercury and calomel that was downright dangerous. This was the state of affairs that led Oliver Wendell Holmes to comment on how bad it would be for the fish if the nineteenth-century pharmacopoeia were thrown into the sea. According to Lewis Thomas, the great twentieth-century physician/essayist, Holmes wasn’t alone in this opinion. Doctors turned to science, Thomas wrote, after “it was discovered, sometime in the 1830s, that the greater part of medicine was nonsense.”
As nonsense goes, however, placebo effects are pretty impressive. Patients taking those ancient remedies—poisonous and inert alike—routinely got better. In part, that was because so many illnesses remit on their own, and the potion’s reputation was only coincidence trumped up by post hoc reasoning—superstition, in short. But after years of giving placebos in virtually every clinical trial, it is now a matter of scientific fact that there’s more to these cures than nature running its course. People given a pill, any pill, will do better than those for whom nothing is done. Researchers have figured out how to allow for this in their calculations: a drug’s effect is the treatment group’s response minus the placebo group’s. But despite the fact that placebos are without a doubt the most widely studied medical treatment in human history, and the hidden subject of every placebo-controlled trial, scientists haven’t figured out why they work. In part, that’s because science in general has a hard time grappling with irrationality, with cases that blur the bright line between sense and nonsense. But science, at least the variety of science bought and paid for by corporations like drug companies, also has a hard time getting interested in sugar pills—which, after all, can’t be patented.
So the placebo effect has been relegated to the role of stalking horse for drugs that can turn a profit. Sometimes it is entirely outmatched. Give a person with bacterial pneumonia a placebo and he is nowhere near as likely to survive as the person given streptomycin. But sometimes
the dummy pill gives the drug a run for its money. Sometimes it even wins. In fact, more often than not, that’s just what happens in trials for antidepressants. A total of seventy-four trials have been submitted to the Food and Drug Administration for the twelve leading antidepressants. Of those trials, only thirty-eight showed an advantage of drug over placebo. That advantage, when it is there at all, is small: another analysis of clinical trials showed that drugs improved HAM-D scores by an average of ten points, placebos by an average of eight, which means that 80 percent of the effect of the antidepressants is due to placebo effects. The drugs that are in the pills, in other words, don’t work very often or very well.
That’s not to say that this two-point difference is negligible. It doesn’t reach the threshold recommended by some governmental bodies—notably Great Britain’s National Institutes for Health and Clinical Excellence—and the FDA’s own director of clinical research in the psychiatric drug department once questioned whether antidepressants should continue to be approved based on such weak numbers (only to conclude that it would be unfair to change direction at this stage of the game). But the FDA data do show that the higher a subject’s initial HAM-D score, the greater the difference between drug and placebo response. And while some researchers think this only means that severe depressions blunt the placebo effect, rather than that drugs are more effective with the severely depressed, still it is clear that those two points may well have saved lives or at least made people feel better enough to get out of bed and go to work and rejoin their families. This is not a trivial effect.
It is, however, much less than what the drug companies claim. You wouldn’t know from a Prozac ad that the drugs have failed almost half of their tests, or that even their successes are well short of miraculous.* But then again when the FDA says a drug is scientifically proven to treat a disease, its manufacturer is well within its rights to take that ball and run with it; that is what the United States government has issued it a license to do. Especially if the company’s best marketing strategy is to sell not only the drug but the disease that it treats, and if its best proof for the existence of the disease is the effect of the drug, then getting this approval is an enormous boon.
That’s why the manufacturers of depression, as if they hadn’t had enough good luck already, must have pinched themselves to make sure they weren’t dreaming when, in 1962, at just about the time that the catecholamine hypothesis was catching on, the FDA suddenly got into a business that it had never been in before: assuring the American public that drugs did what their makers said they did. Government-sanctioned science: as marketing tools go, when it comes to convincing the citizenry that depression is as common as feeling not quite well, that you have the fever and they have the cure—well, it just doesn’t get much better than that.
Depression doctors didn’t invent the strategy of using science and government to sell their wares. Their professional ancestors got the ball rolling for them in 1847, when the American Medical Association, in its first code of ethics, forbade its member physicians to advertise. The ban was part of doctors’ attempt to distance themselves from apothecaries, barbers, medicine-show men, and all the motley assortment of healers to whom people traditionally turned with their suffering. The embargo covered the drugs sold by doctors, which came increasingly to be the drugs produced by the budding pharmaceutical industry and it helped both parties drape themselves in the mantle of modernity. Scientists, the AMA was telling the public, didn’t have to advertise; the results of scientific inquiry spoke for themselves, and it was up to real doctors to inform you of those results. That meant that when it came to drugs, there were two kinds in the world: patent medicines like Cuforhedake Brane-Fude (30 percent alcohol) whose ads you could read in any newspaper, and ethical drugs like heroin (Bayer’s cure for headache, which was mostly acetylated morphine), which were the only kind good doctors recommended.
As that last example might indicate, the drug industry’s claim to possess the facts about disease and cure was really no more justified than that of Emil Kraepelin, who at exactly this time was using the language of science to create the impression that he had discovered mental illnesses in nature. The United States Pharmacopeia, the official compendium of ethical drugs, consisted largely of the same drugs that Holmes hesitated to feed to the fishes. But the branding was brilliant—after all, if your drugs are ethical, then it’s pretty clear what the other guy’s are—and the U.S. government, just then on the cusp of its Progressive Era, ratified this distinction as it turned its attention to the wide-open medical marketplace.
Congress’s attempt to impose some order on that frontier began in the 1890s as part of an overall effort to ensure the safety of America’s supply of food and drugs. Congress was reluctant at first. Despite ongoing suspicions about the drug industry, despite impassioned oratory, despite even a highly publicized stunt—Harvey Wiley, a Department of Agriculture chemist, organized a Poison Squad, whose volunteer members dined for six months on nothing but foods containing suspect additives and got really sick—the bills died in committee. Until the winter of 1905–6, that is, when Samuel Hopkins Adams, in a series of Collier’s articles called “The Great American Fraud,” blew the whistle on the patent drug companies, revealing, for instance, that Radam’s Microbe Killer was 99 percent water with a little bit of sulfuric and hydrochloric acids thrown in, and that infant cough syrup was likely to contain cocaine or opium, or maybe even both. Also that winter, Upton Sinclair’s The Jungle grossed out Americans with its barely fictionalized account of the meatpacking industry. By June, the Pure Food and Drug Act had been passed and signed into law.
The ethical drug manufacturers were ecstatic. The new law, said Mahlon Kline, vice-president of Smith, Kline and French, would be “the greatest instrument for the moral uplifting of the average businessman that had ever been bestowed upon the American people.” The fact that the Bureau of Chemistry, the Agriculture Department agency charged with enforcing the act, chose his company’s chief chemist as its new head probably figured into Kline’s enthusiasm. But even without this favor, the regulations were helpful to his industry. They would give it something that the makers of Brane-Fude and Microbe Killer couldn’t have: the legitimacy conferred by the United States government.
The bureau obliged the industry by imposing standards for drug preparations with which only a large company possessing the resources to control its supply and manufacturing process precisely could comply. Even better, while it forbade manufacturers to place on the label “any statement…which shall be false or misleading in any particular,” the bureau didn’t establish the method by which those statements could be verified. It was left to the drug makers to engage in a war of rhetoric in which the products backed by the most convincing experts prevailed. A company like Smith, Kline, which could muster an army of respected, science-talking doctors to attest to the truth of their labels, had a huge advantage over patent drug makers and their hucksters. Especially after the Harrison Anti-Narcotics Law had limited the amount of opiates allowed in medicines and the Internal Revenue Service began to impose taxes based on alcohol content, the makers of Muco-Solvent, Humbug Oil, and Swamp Root were on the run.
The situation only got better for the ethical drug industry in 1911, when the bureau tried to shut down the company that made Dr. Johnson’s Mild Combination Treatment for Cancer, on the grounds that it didn’t really, as the label claimed, cure cancer at home. Johnson fought the law all the way to the Supreme Court, and the law lost. Oliver Wendell Holmes—the jurist son, not the physician father—delivered the bad news to the bureau. He wrote in his majority opinion that Congress had only meant to regulate drugs “with reference to plain matter of fact, so that food and drugs should be what they professed to be.” That was easy enough to establish, by comparing what the label said was in the bottle with what was actually there. But drug efficacy was a different story. Congress, Holmes said, surely did not mean “to distort the uses of its constitutional power to establi
shing criteria in regions where opinions are far apart.” And whether or not a drug actually did what the label said it did, whether or not Dr. Johnson’s Mild Combination Treatment actually cured people’s cancer—this, Justice Holmes said, was a matter of opinion.
The United States Congress was in a reforming mood in those days, and lawmakers were not content to leave public health to the invisible hand. They tried to give the Bureau of Chemistry the power to require that drugs be effective without running afoul of the Supreme Court, but after a few bruising floor battles, the best they could do was to outlaw “any statement…regarding the curative or therapeutic effect…which is false and fraudulent.” This law actually discouraged research by making ignorance a defense; falsehood is not fraud unless a company has taken the time to figure out the truth. The “greatest instrument” was only getting better for Mahlon Kline and his friends.
Manufacturing depression Page 22