Stay-at-Home Orders Correlated With Slower COVID-19 Spread

Ken Terry

May 15, 2020

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

As many states begin to relax their social distancing measures and open up their economies, two studies suggest that government stay-at-home orders have had a significant impact on slowing the spread of COVID-19.

A study published today in the Journal of the American Medical Association found a significant difference in growth in the number of COVID-19 cases between border counties in Illinois, which has a stay-at-home order, and Iowa, which does not. The other study, published in Health Affairs on May 14, measured the degree to which shelter-in-place orders have decreased the spread of the virus, compared with other social distancing measures.

The JAMA study, which used a cross-sectional approach with a difference-in-differences design, compared daily changes in COVID-19 cases per 10,000 residents in eight Iowa counties that border Illinois with those in seven Illinois counties that border Iowa before and after March 21, when Illinois issued a stay-at-home order. Iowa is one of five states that do not have a stay-at-home order.

The case rates per 10,000 residents were similar in the border counties of both states before the Illinois order. After the order went into effect, the number of cases increased more quickly in Iowa and more slowly in Illinois.

Within 10, 20, and 30 days after the enactment of the stay-at-home order in Illinois, the difference in cases between the two areas, respectively, was −0.51 per 10,000 residents (P < .001), −1.15 per 10,000 residents (P = .02), and −4.71 per 10,000 residents (P = .02).

The researchers estimate that there might have been as many as 217 excess cases in the Iowa border counties after 1 month without a stay-at-home order. This estimate of excess cases, they say, represents 30.4% of the 716 total cases in the Iowa counties by that date.

This is the upper bound of the potential difference in cases had a stay-in-home order been in place in Iowa, George Wehby, PhD, a professor of health economics and health services research at the University of Iowa and one of the study's authors, told Medscape Medical News. He cautioned that there was "a lot of uncertainty in the evidence. Other factors could have come into play."

Wehby and Wei Lyu, MS, also from the Department of Health Management and Policy at the University of Iowa, in Iowa City, performed two sensitivity analyses of the data. The first accounted for the timing of the closing of schools and nonessential businesses in the two states. The second examined differential trends in COVID-19 cases by county population density and poverty rates.

After the Illinois stay-at-home order went into effect, the state's COVID-19 testing rate increased faster than that of Iowa. Assuming the county testing levels were similar to those of the states, however, this difference would not have explained the observed differences in cases per capita. "If it were, the difference [in infection rates] between Iowa and Illinois would have been even larger," Wehby noted.

Wehby stressed that this "is not a causal study. It's a descriptive study examining how these two areas compared when they started pretty similarly, and one of them had an early stay-at-home order that also closed essential businesses. It's an association suggesting there is a possible link and effect, but I wouldn't say this is a direct result of the stay-at-home order."

The correlation, however, is significant, Silvia Martins, MD, PhD, an associate professor of epidemiology at Columbia University, in New York City, told Medscape Medical News. The coauthors of the JAMA study, she said, "used a solid methodology, and they clearly showed that stay-at-home orders matter."

Wehby acknowledged that his study does not show how much stay-at-home orders decrease case growth beyond other restrictive measures, such as the closing of schools and nonessential businesses. The Health Affairs study, he said, "shows more restrictive policies are related to a greater decline in the daily growth rate of COVID."

Social Distancing Measures

The Health Affairs study examined the relative impact of four categories of social distancing measures that state and local governments adopted in March and April 2020 to contain the spread of COVID-19. These included bans on large events; school closures; closures of entertainment venues, gyms, bars, and restaurant dining areas; and shelter-in-place orders (SIPOs), which are the same as stay-at-home orders.

The researchers evaluated the impact of these measures on COVID-19 case growth between March 1 and April 27. "Adoption of government-imposed social distancing measures reduced the daily growth rate by 5.4 percentage points after 1–5 days, 6.8 after 6–10 days, 8.2 after 11–15 days, and 9.1 after 16–20 days," they concluded. "Holding the amount of voluntary social distancing constant, these results imply 10 times greater spread [than the actual number of cases] by April 27 without SIPOs (10 million cases) and more than 35 times greater spread without any of the four measures (35 million)."

Using an event-study design, the researchers measured the effects of these interventions at the county level, which had the same limitation as the other study in applying state-level testing data. Another challenge was that SIPOs were usually the final intervention that states and localities adopted after they'd put in place other interventions.

Neither the ban on large gatherings nor school closures had any impact on lowering COVID-19 case rates. However, closing restaurants, bars, entertainment centers, and gyms did lead to a statistically significant reduction in the spread of the virus. "The estimated effect was 4.4 percentage points after 1–5 days, 4.7 after 6–10 days, 6.1 after 11–15 days, 5.6 after 16–20 days, and 5.2 after 21 or more," the researchers explain.

SIPOs also led to statistically significant (P < .01) reductions in the COVID-19 case growth rate of 3.0 percentage points after 6–10 days, 4.5 after 11–15 days, 5.9 after 16–20 days, and 8.6 from day 21 onward. Because 95% of the population was covered by all four types of restrictions by the time the last SIPO went into effect, the researchers interpreted these estimates to signify the additional effect of SIPOs beyond shutting down schools, large gatherings, and entertainment-related businesses. They attributed the additional effect of SIPOs to "either the requirement/strong advisement to shelter-in-place aside from 'essential' activities or the accompanying closure of any 'non-essential' businesses that remained open."

If the SIPOs had been in place from the beginning of the study period, they would have contributed more than restaurant/bar/entertainment center closures to the reduction in COVID-19 case growth, study coauthor Charles Courtemanche, PhD, associate professor of economics at the University of Kentucky, told Medscape Medical News. But most states adopted the latter type of restrictions 2 weeks earlier than SIPOs.

Behavioral Effects

Although the majority of states have enacted penalties for disregarding SIPOs, Courtemanche doesn't think any states are enforcing those laws. "So what's the mechanism through which a SIPO has an effect? It's really a social pressure nudge," he says. "It's peer pressure. If norms start shifting, you get over a tipping point and people think it's overblown and they shouldn't be doing it anymore. It can unravel pretty quickly."

Martins agreed that the effectiveness of stay-at-home orders depends largely on how well they're communicated. "I think they send a message to the population. Even if you don't have fines, if you're consistent and you say to the population that you expect them to be home, you'll have greater compliance." The needs of essential workers, poor people, and those with chronic conditions must also be considered in this equation, she added.

How people behave when they leave home to, say, go food shopping also makes a big difference in the impact of stay-at-home orders, Wehby noted. "If a stay-at-home order is in effect and a person is outside, would they individually take more precautions to stay a larger distance from people? Would they be more likely to wear a mask? A stay-at-home order doesn't just have to be dramatically changing foot traffic outside."

As to whether the spread of COVID-19 will become more rapid as states across the country start to reopen and to relax or rescind stay-at-home orders, Wehby said, "A lot of national experts suggest that as social distancing criteria are relaxed, there's more of a chance that the infection will spread.

"At the same time," he continued, "it's unrealistic to keep these restrictions forever. But reopening has to be a calculated and gradual process. That's why the evidence from this study and others is useful — because it gives you some idea of what the results are of some of these measures."

The authors have disclosed no relevant financial relationships.

JAMA. Published online May 15, 2020. Full text

Health Aff. Published online May 14, 2020. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.