The pharmaceutical company Moderna started the last, longest step in the process of testing its COVID-19 vaccine candidate at the end of July — a Phase 3 clinical trial. It’s an enormous undertaking: their goal is to recruit 30,000 people, inject some of them with an experimental vaccine and then follow each and every one of them to see how many contract the coronavirus and how many do not. This will take months, even with the federal government’s Operation Warp Speed compressing the timeline whenever possible. Those months may seem endless when over a thousand people are dying from COVID-19 each day in the United States.

The process is long and intensive for a reason, though. Just because a vaccine exists doesn’t mean it’s reasonable or ethical to just give it to people before there’s proof it works, and sticking to the process is why the vaccines on the market today are so safe.

“It’s just fundamentally wrong to think that because there’s an emergency, that we should somehow throw out aspects of scientific research,” says Alex John London, director of the Center for Ethics and Policy at Carnegie Mellon University.

Researchers already know a lot about a few of the vaccine candidates. Moderna’s vaccine, for example, has gone through both Phase 1 and Phase 2 clinical trials, and it has been tested in monkeys. The trials didn’t raise any major red flags, and they showed that people injected with the vaccine produced antibodies against the coronavirus. But none of them were trying to answer the question people want to know the answer to: can this vaccine stop people from getting COVID-19 in the real world?

To answer that question, researchers turn to a Phase 3 trial. They’ll dose thousands of people with the vaccine candidate, and thousands more with a placebo vaccine. Then, they’ll see if fewer people in the vaccine group get COVID-19 than in the placebo group. They’re also watching for any side effects.

The amount of data we have on COVID-19 vaccines right now is only a fraction of what scientists need before they’d recommend something get widely distributed. “The evidence that would convince me to get a COVID-19 vaccine, or to recommend that my loved ones get vaccinated, does not yet exist,” Natalie Dean, an assistant professor of biostatistics at the University of Florida, wrote in a New York Times op-ed.

If a vaccine is a car, the Phase 1 and Phase 2 trials happen while it’s still in the factory. In those trials, scientists are still trying to assemble a vaccine that might work — they’re figuring out the pieces that they might need and how they should be used. If that assembly process goes well, the vaccine candidate can move into Phase 3, where it’s taken for a heavily monitored ride in test tracks and the real world. “Phase 1 and 2 lets you say, ‘we have a lot of things we need to clarify and decide on,’” London says. “Phase 3 says, ‘great, now let’s test that.’”

Those real-world tests are often unsuccessful. Only a small percentage of pharmaceutical products — even those that look promising in early-stage trials — make it through Phase 3 and end up getting approved for use. The success rate is higher in vaccines than treatments, but many of the vaccine successes are for viruses that scientists already know a lot about. New viruses, like the new coronavirus, are much harder to develop vaccines for and have a lower likelihood of success.

Sticking with the car metaphor, once a vaccine hits the test track, there’s a chance it’ll stop running (not actually protect people from COVID-19) or, in the worst-case scenario, crash (have some serious side effects). Its first test drives have to be carefully watched, so that the people designing it can monitor exactly what’s happening.

If, as the Phase 3 trials roll on, scientists start to see clear signs that it is working, they’d stop the trial early and start working to get it approved by the Food and Drug Administration. “If whatever it is we’re testing is really showing that it’s working, scientists would be remiss to continue to test it in people,” says Karen Maschke, who studies human research ethics at the Hastings Center in New York. But that’s a rare exception to the rule. In almost all cases, you should carry a trial through to the end, she says.

It’s particularly important to take vaccines all the way through the process because they’re intended for people who are already healthy. If someone is already sick, the benefits of trying a treatment out (even if it’s unclear how well it actually works) might be worth the potential risks. It’s much harder ethically to justify something that’s still experimental for a healthy person. “A vaccine is the sort of thing that we’re going to administer to hundreds of millions of healthy people, or perhaps even this vaccine could even be given to a billion people,” London says. “We need to know that it’s safe.”

Even if it doesn’t come with side effects, giving people an experimental vaccine outside of a clinical trial and before researchers know if it actually works is risky. People who get an unproven vaccine may feel safer and stop taking as many precautions (like wearing masks or avoiding indoor gatherings) against COVID-19, London says. That could increase the spread of the disease if the vaccine doesn’t work. “We need to know if it works because people are going to change their behavior after they get it,” London says.

Releasing a vaccine to the public before it’s proven to be safe and effective could also erode public trust in vaccines. “With concerns about science, and anti-science sentiment, you have to be really, really careful that you get enough data,” Maschke says. Researchers need to be able to point to the number of people they tested the vaccine in and their statistical analysis as justification for a recommendation. “Trust in the process of developing new drugs and new medicines is fragile,” London says. “We can’t put it in jeopardy.”  

There isn’t usually such focused attention on the clinical trial process, London says. “The public has a window into a discussion that they’re not usually a part of,” he says. That process can be confusing, even at the best of times. Press releases, sound bites, and jargon can create the illusion that scientists know more about vaccine candidates than they do, or that they look more promising than they actually are. For example, when experts say that a vaccine performed well in a Phase 1 trial, they mean that they didn’t see anything that would keep it from moving on to the next stage of the process.

“It doesn’t mean the things the public is interested in — is it safe; if I take this, does it mean I won’t have any adverse events a year later; and is it going to protect me if I’m exposed to the coronavirus,” he says.

Answering those questions is possible, but it will take time. Going at warp speed can save some time, but solid, conclusive answers are worth the wait.