The Real-Life Costs of Bad Regulation
Early administrative failings of the FDA and the CDC greatly exacerbated the Covid-19 crisis in the United States.
Entrepreneur Elon Musk, President Donald Trump, and New York’s Governor Andrew Cuomo have each touted chloroquine, a drug used to treat malaria, as a promising treatment option for those infected with Covid-19. Some media quickly pounced on the president’s statement. The commissioner of the Food and Drug Administration, Stephen Hahn, quickly clarified that the agency had not in fact approved the drug as a safe and effective treatment for the new disease, shortly after the president claimed that the drug was “approved very, very quickly and it’s now approved by prescription.”
Chloroquine is in fact available for prescription in the United States. It’s already being tried as a treatment for the new virus in U.S. hospitals. And multiple manufacturers are rushing to produce more and get it to doctors.
The confusion over chloroquine—along with the broader performance of U.S. regulatory agencies during this epidemic—highlights how our federal process for reviewing and approving drugs and medical devices still leaves much to be desired. Our regulatory regime is costing lives. The early administrative failings of the FDA and Centers for Disease Control, which greatly worsened the crisis in the United States, show how ugly that can be.
Getting a new pharmaceutical compound to market in the U.S. is an extraordinarily complex process. Development time is usually more than a decade. Costs add up to hundreds of millions, if not billions, of dollars. After an innovator submits an Investigational New Drug application, the FDA requires a three-stage testing process, then the submission of a formal New Drug Application that typically includes hundreds of thousands of pages of documentation.
However, once the FDA has approved a pharmaceutical compound as safe and effective for a particular use, doctors can prescribe the drug for an alternative use that has not been run through the full FDA review—commonly called “off-label” drug prescription. A study in the Archives of Internal Medicine estimated that 21 percent of commonly used drugs are prescribed for off-label uses. Off-label drug prescription is particularly common in emergency settings: a 2011 study in the Journal of Critical Care found that 36 percent of all drugs used in intensive-care units are for off-label indications.
That’s the story of chloroquine. In various formulations and analogs (chloroquine phosphate, chloroquine hydrochloride, hydroxychloroquine sulfate), chloroquine has been available for prescription in the United States for decades. (Discovered in 1934, the drug was approved by the FDA in the 1940s.) It has long been used as a common drug to prevent and treat malaria. More recently, scientists have discovered that it has antiviral effects as well—and doctors can prescribe it off-label as an antiviral treatment without further FDA review.
And they are doing so, in the fight against Covid-19. I can offer family reports from a law school friend of mine currently sick with Covid-19 on a ventilator in a New York hospital. Because the drug’s side-effect profile is well-known, it makes sense for doctors to try chloroquine formulations when handling critical cases, given anecdotal but promising reports from other countries further along in handling the epidemic, and especially when no other proven remedies currently exist. The drug might also be a promising prophylactic for health-care workers at serious risk of exposure, and it is currently being tested for that purpose. The anticipated demand for various chloroquine compounds has created a drug shortage, but manufacturers are scrambling to ramp up production—and donating supplies.
Chloroquine is not the only pharmaceutical product being explored as a Covid-19 treatment. Scientists have been testing scores of drugs approved for other uses for efficacy in treating the new disease. And the FDA has been allowing access to not-yet-approved drugs through its “expanded-access” or “compassionate-use” process, among them Gilead Science’s remdesivir, which has proved successful against the Ebola virus. (Alas, the company has been so swamped with requests that its ability to supply the drug has been sharply curtailed.)
The search for drug-treatment options is intense, given the high likelihood that severe cases of Covid-19 require ventilator-assisted breathing. Any treatment or prophylactic that might reduce demand for ventilators or speed up recovery times is extremely important: severe outbreaks that have swamped health-care systems have necessarily led to rationing of ventilator machines, which has contributed significantly to the high death rates in Northern Italy and in China’s Hubei province. The United States has significantly more intensive-care capacity per capita than most other countries—but still not nearly enough ventilators to meet expected peak demand if the virus continues to spread at current rates. Manufacturers, including Musk, are converting existing production lines to produce new ventilators, as well as other necessary protective equipment in short supply. Others are developing alternative processes—including through 3-D printing—to create substitute supply lines.
With cases and deaths growing exponentially, federal regulatory authorities can be expected to fast-track new approaches. The agencies were much less willing to afford latitude to the private sector just weeks ago, though—and the United States is now much more vulnerable as a result. Chinese authorities uploaded the SARS-CoV-2 genome onto the Internet on January 10. The CDC developed its own testing protocol by January 21; international scientists developed a different test by the same date, which was soon disseminated en masse by the World Health Organization.
The U.S. testing process failed. The day that the CDC announced it had developed its testing protocol, January 21, was the date of the first documented American case of coronavirus. South Korea documented its first case the same day. But by March 17, the United States had administered only 125 tests per 1 million people; South Korea had administered more than 5,000 tests per 1 million over the same time span. By aggressive testing, South Korea was able to trace viral spread and contain it. Without it, the U.S. was left with little choice but the draconian measures that have shut down much of American life.
As has been widely reported, the CDC’s in-house testing design was flawed, thus compromising early testing results. Mistakes happen, but the impact of the test-design flaw was much greater than it should have been—owing to the U.S. bureaucracy’s tightly controlled process. Even had the CDC test worked perfectly, not nearly enough tests would have been available for wide-scale testing on the South Korean model.
The reasons: the American regulatory system, cumbersome even in emergency settings; and the specific choices made by regulators that proved to be tragic misjudgments. As Alec Stapp of the left-leaning Progressive Policy Institute has documented, after Secretary of Health and Human Services Alex Azar declared a public-health emergency on January 31, private laboratories had to obtain an Emergency Use Authorization to conduct their own testing. On February 4, the FDA approved an authorization for the CDC—and only the CDC. This created a testing bottleneck, with all testing in the nation routed through the government agency. By February 28, the CDC had processed only 4,000 tests. The next day, the FDA finally invoked the Clinical Laboratory Improvement Amendments to permit testing at some 5,000 highly specialized virology labs (among more than a quarter-million laboratories nationwide with some testing capability). The first Emergency Use Authorization granted to any entity other than the CDC was issued on March 12, to Roche. Throughout this period, the rollout of mass testing was limited by privacy rules in the Health Insurance Portability and Accountability Act (HIPPA); they were not waived until March 15.
Some of the holdups in the critical early U.S. testing effort read like paradigmatic illustrations of bureaucratic bloat. In Emergency Use Authorization applications in the face of an epidemic, the government was actually requiring labs to mail in CD-ROMs for agency review, rather than permitting online submission, owing to outdated rules. (Thankfully, they’ve since dropped that particular rule.)
The botched regulatory response in the United States owed little to the choices of political actors. The individuals running the FDA and CDC are experts in their field, not hacks. FDA Commissioner Hahn is a distinguished oncologist who has authored hundreds of peer-reviewed articles and has executive experience running one of America’s largest hospitals. He’s not a virologist, but the FDA is staffed with them; and Robert Redfield, director of the CDC, is one of the nation’s foremost researchers and clinicians specializing in viruses.
Thus, if the current pandemic crisis highlights the need for experts, the American response also shows the limits of relying on centralized expert command. Doctors and scientists who are the best in their field don’t understand all fields; they are more than capable of designing a test that works (even if they sometimes err), but less capable of figuring out mass-production constraints and the necessary processes for large-scale manufacturing and distribution. We should always be careful not to succumb to hindsight bias, but it’s hard not to believe that the U.S. agencies’ decision to micromanage all national testing for Covid-19 was fatally flawed.
The current viral pandemic highlights the tradeoffs between speed and perfect safety. Delaying the introduction of new drugs to market costs lives, too. The 1992 Prescription Drug User Fee Act has helped facilitate getting some drugs to market more quickly. But it may not go far enough. Scholars studying the question have persuasively argued that we might be better served by jettisoning FDA efficacy review entirely for some or all new-drug approvals, and require only testing for safety. This would mean treating new drugs the same way we do old drugs prescribed off-label.
Finally, there’s little public-health cause for regulating speech about potential off-label drug uses—the easiest way to get information about new applications in the hands of prescribing doctors. When Bayer tweeted that it was donating 3 million tablets of its chloroquine phosphate drug to the United States, it observed “New data, while limited, shows potential for the use of [chloroquine phosphate] in treating patients with #COVID19 infection.” But exactly that tweet by a manufacturer selling its drug in ordinary circumstances could land a company in hot water—exposing it to criminal-law investigations and enforcement that would potentially prevent a manufacturer from getting reimbursed through Medicare for its pharmaceuticals. To avoid that outcome, pharmaceutical companies have entered into multibillion-dollar settlements with federal authorities for such off-label marketing—even though it’s highly likely that informing doctors about new drug uses saves lives.
It’s too soon to tell how effective chloroquine, or other drug therapies, might be as antiviral agents against Covid-19. What is safe to say is that decentralized, private-sector innovation is among the most promising hopes for curbing the mounting death toll from the coronavirus in the United States and other countries—along with necessary, government-led efforts to slow the viral spread. The problems that U.S. federal agencies have faced highlight some of the inherent problems with a regulatory structure that favors centralized action and prioritizes perfection over speed. When ordinary life returns, we need to think about how to create a more sensible and effective regulatory framework.
Photo by Jessica McGowan/Getty Images
City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).