Health Care

Most States Can Safely Relax Some Coronavirus Restrictions

A sign informing visitors that Mulholland Scenic Overlook is closed during the coronavirus outbreak, Los Angeles, Calif., April 5, 2020. (Lisa Girion/Reuters)
Continue more modest restrictions, keep watching the data, and focus on hospitals and protecting the most vulnerable.

One of the main questions on the minds of Americans is when government restrictions aimed at controlling the coronavirus epidemic can be safely lifted. In most states, we now have enough information to say that the strictest measures can be relaxed.

California offers a good example. A review of the data shows that in mid March, when the state instituted severe restrictions on social contact and personal movement, the pace of SARS-CoV-2 infections had already begun to decline. The CDC’s monitoring of data tracking COVID-19 hospital admissions in the San Francisco Bay Area since early March demonstrates that the highest number of hospital admissions occurred between March 21 and 28. In cases requiring hospitalization, that occurs about twelve to 14 days after infection with SARS-CoV-2. Thus infections may well have peaked before the most restrictive of the Bay Area orders issued March 16.

Another CDC analysis of how quickly reported cases of infection were increasing in four U.S. metropolitan areas — Seattle, San Francisco, New York City, and New Orleans — examined the timing of various emergency measures, including limits on mass gatherings, school closures, and stay-at-home orders. In the four metropolitan areas, the authors found a significant decline in relative increase in the number of reported cases after local and state emergency declarations and a continued decline after bans on mass gatherings. There was little further decline in case growth, however, in the two weeks after stay-at-home orders were issued in mid March.

Another report looked at the timing of various interventions in Wuhan, China, and their association with the reproduction number — the number of new cases, on average, caused by a primary case over time. When that number falls below one, the epidemic stops growing. Remarkably, the reproduction number for SARS-CoV-2 began to drop dramatically in mid to late January, before the implementation of the most stringent home-confinement policy in early February. The interventions before that time included the restriction of air and train travel, local traffic bans, increased mask-wearing, limits on social gatherings, isolation of those who were ill, and home quarantine for those who were exposed. By the date of the universal home-confinement order, the reproduction number had declined to 1.2 and only four days later to less than 1.0.

For our part, we have looked at the relationship between the timing of stay-at-home orders and the peaks in the number of reported cases in 31 U.S. states. We could not find a clear pattern relating the timing of such orders to the peaks in case numbers, suggesting no association between the orders and the limitation of cases. One would have expected, if there had been an effect of such state-at-home orders on the number of new cases, a consistent time-dependent effect —that is, a clear, observable average time, say ten to 14 days from the date of the order to the peak. That was not seen.

Closing down large sectors of society is a blunt tool, and its use comes with huge costs—the disruption of education, economic activity, and social interactions. The ongoing, and possibly future, use of such stay-at-home orders deserves timely and rigorous evaluation. As much as possible, we want to know how necessary, and how costly, each intervention (e.g., limiting gatherings, closing restaurants and retail businesses) is on its own. In the realm of public health, first principles of disease control dictate the use of the least costly set of interventions sufficient to prevent and mitigate illness.

Overall, several pieces of evidence suggest that the pace of new SARS-CoV-2 infections had declined significantly before government officials imposed the most severe restrictions. Those findings, along with the fact that most hospitals, outside of certain severely affected areas like New York City, had adequate capacity throughout the month of April, should give us confidence that self-protective changes in personal behavior along with testing, isolation, and quarantine can adequately mitigate the epidemic.

It is time to step back from restrictions like prohibition of access to outdoor spaces, universal orders for home confinement, and closure of small businesses. In stages, and watching the data on test-positivity in key populations and COVID-19–related emergency-room visits and hospital admissions, we can get back to work and school while limiting person-to-person contact with measures such as occupancy limits, promoting hand-washing, and adopting policies for paid sick leave to enable ill workers to avoid others and reduce the spread of infection. Public health should focus on protecting the most vulnerable with screening and rapid response to outbreaks in nursing homes and shelters.

Finally, ensuring that we’re using the least restrictive means necessary also requires that we focus on keeping hospitals above water. That can be done by adjusting policies specifically to hospital-admission rates of COVID-19 patients and local bed-capacity for their care (e.g., dedicating 50 percent of acute-care beds). Leading indicators like daily COVID-19–related outpatient and emergency-room visits can now be monitored to provide an early warning system.

The findings described above, and increasing experience in other settings, lead us to believe that relaxing the strictest interventions while continuing more modest ones will not cause an unmanageable resurgence in cases.

Jeffrey D. Klausner, a former CDC medical officer, is a professor of medicine and public health at UCLA. Rajiv Bhatia, a former San Francisco City and County deputy health officer, is an assistant clinical professor of medicine (affiliated) at Stanford School of Medicine.


The Latest