As April turned to May, more than a dozen states opted to relax coronavirus shutdowns and at least partly reopen their economies. This is not a public health decision, but an economic and social one. Still, they can contribute to public health by providing data all 50 states can use.

Nobody knows exactly what level of restrictions we need to keep COVID-19 from overwhelming hospitals. Georgia, Colorado, Texas and others are conducting a natural experiment. While some scientists fret that reopening too soon could increase infections nationwide, these early birds can still do us a favor: A promise to be our guinea pigs, regularly testing random samples of their populations and providing other kinds of data that will help us understand how the disease spreads. Even without an organized effort, they will be under the microscope.

On a podcast for the New England Journal of Medicine, editor-in-chief Eric Rubin and deputy editor Lindsey Baden discussed how little we still know about the efficacy of the patchwork of stay-at-home and lockdown rules. In the United States, we haven't even agreed on the goal of these shutdowns: Is it keeping case numbers as low as possible? Or just preventing hospitals from exceeding capacity? The answer may differ from one community to another, and could determine how strict the restrictions should be.

Even if COVID-19 cases do rise in these reopening states, it could have a benefit: A better way to test vaccines. Experts on vaccine development say there are no shortage of people asking to volunteer for human studies, but it's hard to test vaccine candidates on locked down populations. To know if a vaccine works, a certain number of people must be exposed to infection.

In mid-April, public health experts had argued that states that want to reopen should wait until new infections rates have fallen for 14 straight days, and should also have the testing and contact tracing capacity to quarantine those likely to be infected. That could entail Americans getting several million tests every day.

But testing won't ramp up to the level needed any time soon, says Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota.

Beyond the current inadequacies with tests, which he wrote about in the New York Times, he wants to address a misunderstanding about the benefit of lockdown-type policies. They don't prevent infections — only delay them. The main benefit for society is that by slowing down the rate of serious illness, we can keep our hospitals from being overwhelmed. Beyond that, as I've said in previous columns, there's also an upward learning curve so getting sick months from now rather than now means a higher standard of care.

Osterholm is right that it's unrealistic for entire populations to remain locked up for 16-20 months. The reality is that this pandemic will only end when some 70% of the population develops immunity — thus creating the so-called herd immunity that can starve the virus of hosts. "The bottom line is we're going to get there," he said. We might get help from a vaccine, or we might not. But that's the end result.

After various regions allow more activity, he said it's possible we'll see a steady stream of cases, or small peaks and valleys, or we'll see a large second peak.

A second peak was predicted by epidemiologist Marc Lipsitch and colleagues, in a model published in Science in mid-April. One strategy outlined in the paper was to let some herd immunity build up over the summer while using whatever limited testing capacity we have to do random sampling and monitor for the start of the next wave. Under that plan, we'd all have to prepare to go back into lockdown before that wave got too high.

Others, including Osterholm, have been contemplating a way to protect those who are likely to get seriously ill, while allowing more freedom for those who are at very low risk — though not no risk.

This has been Sweden's strategy, and while it has its detractors, they have decided on a long-term, sustainable strategy which may win in the long haul in reducing overall pain and suffering. As Osterholm has been saying, we're currently only in about the second inning of a nine-inning ballgame.

The fact that some countries and states are behaving differently isn't all bad — just as long as they keep tracking critical data, and offering it to scientists who can do a systematic evaluation of their results.

Faye Flam is a Bloomberg Opinion columnist. This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.