Shelter-in-Place Orders Averted Much Higher COVID-19 Death Toll

Ken Terry

July 13, 2020

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

The enacting of shelter-in-place orders (SIPOs), also known as stay-at-home orders, was associated with significantly decreased growth in COVID-19–related hospitalizations and deaths during the first phase of the pandemic, a study finds.

Using data covering various dates in March through May 15, the researchers estimate that 42 days after most states enacted SIPOs, the daily mortality growth rate declined by up to 6.1 percentage points in those states.

"Projections suggest as many as 250,000 – 370,000 deaths possibly averted by May 15 in the 42 states plus the District of Columbia with statewide SIPOs," study coauthors Wei Lyu, MS, and George L. Wehby, PhD, MPH, write.

The findings were published online on July 9 in Health Affairs.

During the same period, the daily hospitalization growth rate declined by up to 8.4 percentage points on average after 42 days in the 19 states with SIPOs and in three states without SIPOs for which data on COVID-19–related hospitalizations are available. The researchers' model estimates that as many as 750,000 to 840,000 additional COVID-19 hospitalizations might have been averted in the 19 states with SIPOs.

Earlier studies showed that SIPOs slowed the growth of new cases. In this study, the authors conclude, "We find evidence that SIPOs played an important role in decelerating the growth curve for COVID-19 deaths and hospitalizations." In addition, SIPOs and other social distancing measures helped "flatten the curve" to prevent hospitals from being overwhelmed by patients with COVID-19, Wehby, a professor of health economics and health services research at the University of Iowa, told Medscape Medical News.

Study Methods

The researchers used an event study design to account for the various dates on which different states enacted SIPOs. By April 6, 42 states and the District of Columbia had statewide SIPOs in place. Three other states had more limited SIPOs and were not included in the study. Five states never enacted stay-at-home orders.

The mortality and hospitalization datasets were drawn from public reports. Because of gaps in the data, the researchers were limited to hospitalization data from 22 states.

The reference period for measuring the growth in the death rate was the first 7 days of the study period, which began when a state enacted an SIPO. For the hospitalization arm of the study, the authors used the first 7 days, plus whatever earlier hospital data were available, as a reference period.

Because states enacted their SIPOs at different times, the study period differed for each state, Wehby noted. For example, the 29th day for one state might have been in late April; for another, it might have been in early May. States that enacted SIPOs later were treated as controls before they enacted these regulations, and not all of them were under SIPOs for 42 days.

Within 22 to 28 days after a SIPO became effective, the daily COVID-19 mortality growth rate declined by an average of 2.9 percentage points — a significant change — in relation to the baseline growth rate of 20.5%. From 29 to 35, 36 to 42, and 43 or more days, the mortality growth rate dropped by 3.6, 4.7, and 6.1 percentage points, respectively.

By May 15, the average daily growth in deaths was 2.8% in the SIPO states. Without stay-at-home orders, the average daily growth rate would have been 8.6%.

The growth in the daily hospitalization growth rate declined significantly (P < .05) at 15 days after the enactment of SIPOs. The effect increased over time and remained statistically significant except for the period of 29 to 35 days. The decreases in the hospitalization growth rate were 3.2, 5.5, 5.4, 6.9, and 8.4 percentage points within 15 to 21, 22 to 28, 29 to 35, 36 to 42, and 43 or more days after enacting SIPOs, respectively.

To distinguish the effects of SIPOs from those of other social distancing measures, the study controlled for the impact of six statewide interventions, including major disaster declarations; school closures; bans on large gatherings; travel restrictions that mandated 14-day quarantines for out-of-state visitors; the banning of visitors to nursing homes; and closures of restaurants, gyms, and entertainment venues. In the final 2 weeks of the study, the results were also adjusted for states that had begun to reopen their economies.

What Can Be Done With the Data?

Anthony Fauci, MD, director of the National Institute of Allergy and Infectious Diseases and a member of the White House Coronavirus Task Force, has said that stay-at-home orders may have to be reimposed if other methods of containing the current outbreak fail. But Wehby stopped short of advocating that approach.

"I'm not advocating for SIPOs because we know they impose dramatic effects on economic outcomes and on other aspects of health and life," he said. "But if you are purely focused on just flattening growth rates for hospitalizations and deaths, they are one of the tool kits that policymakers are already talking about — they're not taking them off the table."

Preeti Malani, MD, chief health officer and professor of medicine, Division of Infectious Diseases, the University of Michigan, Ann Arbor, told Medscape Medical News that it's unclear what to do with the data in the Health Affairs study. "In places that are having sustained, out-of-control spread ― particularly, Florida, Texas, and Arizona and other southern states ― I don't know there's an appetite to do this [adopt SIPOs]. We are in a very different place with the pandemic, emotionally and weather-wise, than we were in March and April."

However, she said, if the states where the pandemic has gotten out of control had the political will to return to SIPOs, they could control the latest outbreak. "If people stopped moving, within 2 weeks, you'd see improvement in Florida, in Texas. But short of that, you're not going to get control. Eventually, they'll have to come up with that kind of plan, whatever they call it. Otherwise, K-12 students will not come back to school, colleges won't go back, businesses will end up shuttered or people just won't go. They won't send their kids to school or go out."

Validity of Data

Overall, Malani said, the study's estimates of the number of hospitalizations and deaths that were averted by SIPOs might be a bit on the high side. On the other hand, she pointed out, the 1-week lag between the SIPOs' impact on hospitalizations and deaths is about right. So is the ratio between the estimates of averted deaths and hospital admissions, which is usually between 25% and 33%.

It is possible, Malani added, that the estimates are on target, considering that only 1% to 2% of the population has been infected so far, and the number of deaths is quickly approaching 150,000. "Maybe the numbers in the study are reasonable. Models depend on the numbers you put in, and there's always a range. But these numbers don't seem out of range."

Because it's clear that the US population won't reach herd immunity anytime soon, she said, the rise in hospitalizations and deaths is very concerning. "When you're seeing numbers like 60,000 new cases a day, and the numbers in very populous states, it's clear that this is not sustainable. It's just a matter of time before it is a situation like Italy."

The authors have disclosed no relevant financial relationships.

Health Aff. Published online July 9, 2020. Full text

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


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.