Diana Levine suffered from chronic migraine headaches for many years. So, in April 2000, when she went to a local clinic to get treatment, she knew what to expect. She’d received the same treatment several times before: an injection of Demerol for the pain, and an injection of an antihistamine called Phenergan to treat the nausea that accompanies both migraine headaches and Demerol itself. Everything was normal — except for how the drugs were actually administered. The physician’s assistant who gave her the drugs accidentally injected the Phenergan into an artery, instead of a vein, causing tissue death and gangrene. Ms. Levine, a professional musician, eventually had to have her arm amputated.
Levine naturally sued the physician’s assistant, the supervising physician, and the clinic for malpractice. After all, Phenergan’s FDA-approved label specifically warned against injection into an artery, precisely explaining the likely, tragic side-effects. But Levine also sued the pharmaceutical company that manufactures Phenergan, Wyeth, arguing that the warnings were not clear enough. Despite Wyeth’s defense that the drug had been on the market since 1955, FDA had known about the gangrene risk since 1967, and the agency had explicitly approved the warning that was included on the label, a Vermont jury awarded Levine $6.8 million dollars.
Wyeth appealed the jury verdict, arguing that, because it had complied with FDA regulations on both product safety and labeling, and because there was no new information about Phenergan’s risks that would supersede FDA’s judgment about the warning language, FDA regulation should preempt the state tort claim. And FDA agreed. Nevertheless, the Vermont Supreme Court affirmed the verdict, and Wyeth further appealed to the U.S. Supreme Court, which hears oral argument in the case today.
The usual argument against preemption in such cases is that immunizing drug firms against state tort claims would incentivize manufacturers to hide information from FDA — as is often claimed for some drugs, such as the Cox-2 inhibitor Vioxx, diabetes medication Rezulin, and weight-loss pill Meridia. Earlier this year, an equally divided Supreme Court affirmed a Second Circuit decision finding Warner-Lambert liable after plaintiffs claimed the company concealed safety and efficacy information from FDA. (Because Chief Justice John Roberts recused himself from the case, the 4-4 split leaves no high court precedent on the preemption issue.) But that’s decidedly not the case here. It seems that Wyeth attorney Bert Rein carefully selected this case for Supreme Court consideration precisely because it carries none of the same baggage that a drug like Vioxx would.
FDA has known for four decades about the considerable gangrene risk posed by arterial injection of Phenergan, and the agency struggled for a time with how to address that risk. After consulting with a panel of independent scientific experts, FDA decided to keep Phenergan on the market and to tailor the warning carefully with the language now used on the drug’s label. Unlike relatively new drugs, whose side effects often do not become apparent until after label language has been drafted and approved, FDA has known for a very long time about Phenergan’s risks. And the current label language was approved after those risks became well known and understood. With Phenergan, exposure to state tort claims would do nothing to incentivize the kind of additional vigilance and testing that would reveal new information about the drug’s risks.
Why does preemption matter? Because, as flawed as the system is, Congress has delegated to FDA the authority to carefully examine the risks and benefits of medical drugs and determine whether they are, on balance, safe enough for doctors and their patients to use. FDA also has authority to regulate the exact language that appears on every drug’s label — power that it uses vigorously. When, in a case like this, FDA has relied on its scientific expertise to carefully tailor the label language in such a way as to maximize the drug’s potential benefits, permitting a handful of lay jurors to decide whether that language provides sufficient safety information makes a mockery of the entire regulatory process. If laymen possessed the expertise necessary to make such careful risk balancing decisions on behalf of the entire society, FDA and other expert regulatory agencies wouldn’t be necessary.
Perhaps more important, once drugs are approved, so long as information about their risks is clearly stated on the labels, doctors and patients are free to choose whether the risks outweigh their benefits in individual cases. Risk averse doctors and patients can choose to avoid drugs that carry considerable, known risks — or even to forgo relatively new drugs about which some risk information may still be unknown. On the other hand, when drugs are not available at all — whether because the FDA has not approved them or because excessive tort liability incentivizes manufacturers to keep them off the market — all choice is removed from patients and their doctors.
Vigorous tort enforcement can, and often does, incentivize makers of all sorts of consumer products to be extra vigilant in the design, manufacture, and marketing of their products. But, as is the case here, where there is no evidence that the drug manufacturer was negligent in any meaningful way, and no evidence that the manufacturer hid or otherwise suppressed relevant information from the FDA, preemption is the only course that makes any sense at all.