Imagine a society that lets its automakers oversee crash tests on new models, allowing the industry to report results, as it sees fit, to government and consumers. Sometimes, an automaker might not reveal the outcome of a test that turned out badly, deciding that the dummies in the vehicle had been too short—no wonder their chests were crushed!
In other cases, a company might postpone reporting on crash test results for a year or two, hoping that later trials would turn out better. In these cases dozens of trials might be required in order to achieve the desired outcome. The car maker would, of course, pass the additional costs along, in the form of higher sticker prices.
In this society, crash tests are not run and paid for by an independent entity like our National Highway Traffic Safety Administration (funded by taxpayers) or the Insurance Institute for Highway Safety (funded by insurers). Instead, the auto industry itself finances and controls the trials. Automakers also provide most of the funding for the government agency that rules on car safety. Finally, under this system, head-to-head comparisons of cars in a similar weight class are frowned upon. Such trials would create winners and losers—and who wants to be a loser? Instead, each company tests its own cars, and when outcomes finally are published, they tend to be excellent.
Probably you already have guessed where I’m heading. The system I’ve sketched comes pretty close to describing how we try to assure the safety of the prescription drugs and medical devices sold in the U.S. We may be the only country in the developed world that allows the companies that manufacture and peddle medical drugs and devices to control what we know about them. The industry also provides much of the funding for the Food & Drug Administration, the agency responsible for weighing the risks and benefits of these products. No wonder the FDA doesn’t require manufacturers to test their products against similar, less expensive products already on the market. Instead, the FDA asks only that the sponsor to test its new entry against a placebo—demonstrating that it is “better than nothing.”
In the U.S., medical research is rife with conflicts of interest. As Merrill Goozner recently testified before the Institute of Medicine (IOM) panel on “Conflict of Interest in Medical Research, Education, and Practice,” the share of our clinical research funded by industry has doubled over the past 30 years, and now accounts for well over 60 percent of all clinical trials. As a result, Goozner testified, the medical literature is riddled with:
- reports of negative research results being suppressed
- delays in publication
- failure to report serious adverse events in clinical trials
- the slanting of systematic reviews of medical evidence
- and a systematic bias in research results that favors the outcomes desired by study sponsors.

