With more than 250,000 Americans dead from the COVID virus and far more hospitalized, our country is desperate for a vaccine that prevents serious cases. Billions of dollars in taxpayer money are being spent to put a stop to the virus, yet drug companies are using predominantly mild cases that produce cold-like symptoms, not infections or serious cases, to test the efficacy of their vaccines. Based on the reviews of trial protocols for seven leading vaccine candidates, including those from Moderna and Pfizer, the British Medical Journal (BMJ) wrote last month that “none of the [COVID] vaccine trials are designed to detect significant reduction in hospital admissions, admission to intensive care, or death.” Can these vaccines stop the pandemic?

Drug companies — not the FDA — choose what is called the “primary end point” or “measure of efficacy” for the trials they design to get vaccines or drugs approved for mass marketing (a dangerous conflict of interest in its own right). By focusing on mild symptoms, it is much easier for a company to claim a vaccine is effective and bring it to market, but it might not prove as effective in serious cases.

On Nov. 9, Pfizer and BioNTech jointly announced a vaccine that was more than 90% effective in preventing COVID-19 in their large trial where 38,995 participants received both the first and second shot. The question that remains is: What does “preventing COVID-19” mean here? It does not necessarily mean preventing infections.

As is standard in clinical trials, half of these participants received an injection of salt water and half received the real vaccine. Three weeks later, still blinded from knowing which they received, participants received the second injection. Seven days later, the data monitoring committee found that more than 90% of the 94 cases that developed a mild COVID symptom like a sore throat were among the approximately 19,500 participants in the placebo group, with fewer than 10% among the other half receiving the real vaccine. This is the basis of the announcement that the vaccine is more than 90% effective in “preventing COVID.” Although vague wording has led many to believe otherwise, Pfizer provided no evidence about preventing infections. As for serious cases, Pfizer announced Wednesday data showing promising preliminary results but saying nothing about hospitalizations or death.

This week, Moderna announced that its vaccine was 94.5% effective in preventing symptomatic cases (not infections) in more than 30,000 U.S. participants, based on its study in collaboration with Anthony Fauci’s National Institute of Allergy and Infectious Diseases. The study found 90 cases in the placebo half but only five cases in the half getting its vaccine. Moderna also analyzed a secondary end point of severe cases. Eleven occurred in the placebo group not taking the vaccine, and none among those taking it. While the numbers are small, the finding is important and bodes well for more information on how the vaccines prevent severe cases.

Once a vaccine is approved effective, however, it becomes more expensive and difficult to develop and test a more effective vaccine for serious COVID-19 cases. The trial may have to prove the new vaccine is more effective than the first authorized vaccine, rather than a placebo or inert substance, requiring a much larger sample and greater expense. Furthermore, new trials are likely to be restricted to groups not approved to receive the first vaccine, such as health-care workers, the elderly, or other high-risk groups. Recruitment also becomes more difficult.

Other concerns associated with vaccine trials include stopping trials prematurely or not monitoring patients for at least two months after they receive a vaccine. Both Pfizer and Moderna have followed up by a median of more than two months, but ending trials early may downplay or hide potential health risks, making a vaccine look safer and more effective than it is.

Because vaccines are administered to entire populations, they need to be tested using volunteers from the whole population. But the Food and Drug Administration allows companies to exclude or under-represent certain populations from the sampling, and companies commonly exclude or under-represent those at higher risk and people with preexisting conditions. Both companies have made efforts to include older and minority participants, but the BMJ thinks it’s not enough. We need to know more about how well these and other vaccine candidates protect the elderly, minorities, and other high-risk groups against serious COVID.

In theory, a safe and effective vaccine should protect the health of the population from serious illness, hospitalization, and deaths; it should enable the economy to revive and restore millions of jobs; and it could also restore some of the damaged reputation of the pharmaceutical industry. But based on my years of research on the industry’s policies and the FDA as its paid partner, I am concerned that COVID vaccines will generate billions in revenue without much reducing hospitalization or deaths.

The public’s wariness about the rush to count on a COVID vaccine to end the pandemic is understandable. Our best hope is to learn from other countries that have used social isolation, masking, testing, and monitoring to drive down infections, serious cases, and deaths. In the near term, we cannot rely on vaccines to stop this pandemic.

Donald Light is a visiting scholar at the Institute for Advanced Study and a professor of psychiatry and health policy at the Rowan School of Osteopathic Medicine. He is a faculty affiliate of the New York University Division of Medical Ethics.

*Editor’s note: This post has been updated to reflect a new report from Pfizer about its vaccine trials on the morning of Nov. 18.