Coronavirus crisis: Mr. President, governors, don’t pick a date. Look at the data.

President Donald Trump declared that he wants the country “opened up” by Easter Sunday, April 12, a date cited without data-driven projections but thought to be “beautiful time.” Policy experts, such as National Institute of Allergy and Infectious Diseases Director Anthony Fauci, reacted to the statement with alarm, calling it an “aspirational projection.” Happily, the president modified his pronouncement, now suggesting physical distancing through the month of April.

Still, how will the country know when coronavirus is no longer a threat? The answer: Lawmakers must make policy decisions in the same way they drive cars.

Think of coronavirus like a car, while a policy change, such as a countrywide or regional shutdown, acts like a brake. In the past few months, world governments have implemented unprecedented measures to combat COVID-19. Wuhan, China, was placed under quarantine on Jan. 23, a dozen days after China recorded its first death.

In the United States, Trump banned all travel from Europe on March 11 and Gov. Andrew Cuomo locked down New York on March 22.

Britain underwent a lockdown on March 23, a few days before Prime Minister Boris Johnson announced he was infected.

When is it time to let off?

Knowing how long to hold the brake is the issue. The moment the car stops is when the threat is over and the country can be opened up, but we also need to be mindful of holding the brake for too long because, if left unmonitored, the policy would have a more devastating impact on the economy than necessary. The Brookings Institution predicts that the United States will lose $420 billion in real GDP this year due to the effects of coronavirus.

When we learn to drive, we expect the car to immediately stop when the brake is pressed, but with experience, we know that this is not the case — serious lag time exists between the moment the brake is pressed and the moment the car stops. Awareness of this lag time is critical. We can avoid deaths and salvage gross domestic product.

So far, cross-country evidence shows an approximate 11-14 day lag time between the enactment of a lockdown and the peak of the daily new reported cases. The peak in daily new coronavirus cases in China occurred a dozen days after the government quarantined Wuhan, and the same happened in South Korea, through a de facto lockdown that occurred as a result of fear instilled by Kwon Young-jin, the mayor of South Korea’s fourth largest city, Daegu. The streets of South Korea were deserted on Feb. 20, but it wasn’t until March 3 that new cases started to trend downward.

Former CDC head on coronavirus testing: What went wrong and how we proceed

Multiple factors, including testing and progression of illness, account for this policy lag. An individual typically begins to show symptoms five days after infection, and it takes eight to nine days for a severe coronavirus case to progress from initial symptoms to the most severe outcome of the case (usually hospitalization, less often death). Testing can occur at any time between the initial onset of symptoms and the severe outcome of the case, and an unfortunate reality is that many labs still take three to five days to process test results. In sum, the lag from first onset of infection to appearance as a daily new case can be anywhere from eight to 19 days.

What’s important to note is that this lag — when we see a change in the rate of new infections as a result of behavioral change — works in reverse as well. If we relax our social distancing measures, it will be two weeks before the world knows whether or not the decision was too early.

Decisions must be made regionally

Based on what we have learned in China, Italy and South Korea, we know that Trump's April 12 estimate is too early, but we may be able to begin the opening process in May or June.

And it will vary by region. We cannot even consider relaxing measures in any one region until case growth is slowing in number and hospital ICU capacity for new cases is no longer constrained. New cases combined with long hospital stays means that our hospitals will continue to get overwhelmed long after the peak in new cases occurs.

HOTLINE: Share your coronavirus story

The experiences of South Korea and China have demonstrated how vigilant a country must be to keep cases declining, or at least stable at a low level. In South Korea, for instance, a large percent of new cases continues to be “imported,” and airport testing and quarantine may need to be in place for a while.

Keep in mind that deaths tend to lag case finding. We could witness, as in South Korea, that when you are aggressive with early case finding, you will ultimately see the overall case fatality rate trend upward over time. This should not be a cause for alarm or an indication that the actions have failed.

The Editorial Board: The right way forward on coronavirus

A universal truth that we can glean from successful countries is that certain policies, particularly lockdowns and mass testing, are effective in containing COVID-19. Nobody benefits from a sputtering economy. Even so, all of our efforts will be for naught if we release the brake too soon, and our coronavirus car plows through a crowd.

Evie Cai is a MPH student at the Yale School of Public Health. Sten Vermund is dean and Anna M.R. Lauder Professor of Public Health at Yale. Howard P. Forman is professor of Public Health, Management and Radiology at Yale. Follow them on Twitter: @caievie, @Svermund, @thehowie

You can read diverse opinions from our Board of Contributors and other writers on the Opinion front page, on Twitter @usatodayopinion and in our daily Opinion newsletter. To respond to a column, submit a comment to letters@usatoday.com.

This article originally appeared on USA TODAY: We can't simply pick a date to end coronavirus social distancing