May 20, 2020

Which social distancing policies are most effective? Here's what a new study found.

Daily Briefing

    A new study published in Health Affairs found that social distancing policies have significantly reduced the spread of the new coronavirus in America—and sheds light on which social distancing measures were most effective.

    10 takeaways: What health systems must consider as they mull reopening

    Study details

    For the study, researchers from the University of Kentucky, the University of Louisville, and Georgia State University looked at four social distancing policies that were implemented in various states throughout the country:

    • Bans on large gatherings;
    • Closing restaurants, bars, gyms, and entertainment venues;
    • Closing schools; and
    • Stay-at-home orders.

    The researchers used an event-design study to evaluate the policies' effects on the growth rate of confirmed cases of Covid-19, the disease caused by the new coronavirus, in the United States from March 1 through April 27, when Georgia became the first state to begin reopening nonessential businesses and scaling back the state's social distancing measures. They calculated their Covid-19 growth rate estimates based on data from the 2019 Novel Coronavirus COVID-19 Data Repository created by the Johns Hopkins Center for Systems Science and Engineering, which includes data on Covid-19 cases collected from government and independent health institutions.

    Social distancing measures lowered Covid-19 case growth rate, study finds

    The researchers estimated that the four social distancing policies together reduced the daily growth rate of confirmed Covid-19 cases by 5.4 percentage points after being in place for one to five days, by 6.8 percentage points after being in place for six to 10 days, by 8.2 percentage points after being in place for 11 to 15 days, and by 9.1 percentage points after being in place for 16 to 20 days. They wrote, "Holding the amount of voluntary social distancing constant, these results imply … more than 35 times greater spread without any of the four measures" in place.

    The most effective social distancing policies

    The researchers found that stay-at-home orders and closing entertainment-related businesses each had statistically significant effects on lowering the Covid-19 growth rate. They noted that closing entertainment-related businesses had a greater actual impact on the growth rate reduction than stay-at-home orders, but only because the closures typically were in place longer and in more areas of the country than stay-at-home orders. According to the researchers, stay-at-home orders had a greater estimated effect on the growth rate longer term.

    Specifically, the researchers estimated that stay-at-home orders were associated with a 3 percentage point decline in the Covid-19 growth rate after being in place for six to 10 days. That decline rose to 4.5 percentage points after being in place for 11 to 15 days, to 5.9 percentage points after being in place for 16 to 20 days, and to 8.6 percentage points from day 21 onward. The researchers wrote that those estimates "should be interpreted as the additional effect of [stay-at-home policies] beyond shutting down schools, large gatherings, and entertainment-related businesses."

    The researchers estimated that closing entertainment-related businesses was associated with a 4.4 percentage point decline in the Covid-19 growth rate after being in place for one to five days. That decline rose to 4.7 percentage points after being in place for six to 10 days and to 6.1 percentage points after being in place for 11 to 15 days. However, the policy's impact decreased to a 5.6 percentage-point reduction after being in place for 16 to 20 days and to 5.2 percentage points from day 21 onward.

    The researchers found closing schools and banning large gatherings did not result in statistically significant reductions in the Covid-19 growth rate, but they said those finding don't necessarily mean the policies had no effect. The researchers explained that those policies might have affected the new coronavirus' spread in a way their analysis was unable to measure. Further, they noted that the study's "95% confidence intervals meant that [they] could not rule out reductions of up to 4–5 percentage points" from those policies.

    Discussion

    The researchers accounted for a number of variables in their study, including the availability of testing kits for the new coronavirus and how susceptible different populations are to contracting the virus based on population density and age. The researchers also assumed a level of social distancing would occur voluntarily in areas where social distancing policies weren't in place.

    However, the researchers noted they couldn't account for every possible variable that could affect the new coronavirus' spread, and they didn't evaluate how wearing face masks, closing public parks and beaches, or limiting visits to nursing homes may have impacted the Covid-19 growth rate.

    Further, the researchers noted that numbers of reported Covid-19 cases in the United States are "known to understate the true prevalence of [Covid-19], as they do not include asymptomatic carriers, those who are not ill enough to seek medical care, and those who are unable to obtain a test due to supply constraints." In addition, they said more research will be "needed as gradual, untested steps toward reopening are taken across the country."

    Aaron Yelowitz, an economist at the University of Kentucky and a co-author of the study, said, "The highly visible, strong government-imposed measures achieved one key goal: dramatically reducing the spread of Covid-19 disease while allowing the health care system to not get overwhelmed. In principle, the measures also bought time to develop a more effective testing and contact tracing infrastructure, although we will see whether that actually has been achieved as states start to reopen" (Scott, Vox, 5/18; Owens, "Vitals," Axios, 5/15; Bella, "Morning Mix," Washington Post, 5/15; Moreno, The Hill, 5/15; Courtemanche at al., Health Affairs, 5/14).

    Have a Question?

    x

    Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.

    X
    Cookies help us improve your website experience. By using our website, you agree to our use of cookies.