At the start of the COVID-19 outbreak, each additional day that states delayed declaring an emergency was associated with a 5% increase in mortality, a study has found.
All data and statistics are based on publicly available data at the time of publication. Some information may be out of date.
The imposition of physical distancing measures by state governments during the COVID-19 pandemic has been politically controversial in the United States.
Critics have questioned the necessity and effectiveness of such measures, particularly in light of the considerable economic costs.
In the absence of direct evidence early in the outbreak, as the number of local cases mounted, states based their decisions to close schools and declare an emergency on modeling studies and evidence from past epidemics.
A 2007 analysis of the influenza pandemic of 1918–1919, for example, found that earlier closure of schools and bans on public gatherings by U.S. city authorities was associated with lower subsequent mortality in those cities.
Researchers in Philadelphia, PA, have now published one of the first studies specifically assessing the effect of physical distancing on COVID-19 death rates.
They used the unambiguous dates when states declared emergencies and closed schools as proxies for when each began to implement significant physical distancing measures.
“Before this study, we assumed social distancing worked based on modeling and studies of prior pandemics, but we didn’t have substantial quantitative data to show its effectiveness for COVID-19,” says lead author Nadir Yehya, assistant professor in the Department of Anesthesiology and Critical Care Medicine at the Children’s Hospital of Philadelphia and the University of Pennsylvania.
“Our analyses demonstrate that states that issued emergency declarations earlier helped curb the spread of the disease,” he says. “These results confirm how important it is to implement social distancing measures early to reduce COVID-19 deaths.”
The study appears in the journal Clinical Infectious Diseases.
To test whether there was an association between physical distancing measures and mortality, the researchers analyzed 55,146 confirmed COVID-19 deaths across 37 states between January 21, 2020, and April 29, 2020.
They assigned the day when a state exceeded 10 COVID-19 deaths as “day one” and noted the number of deaths after 28 days.
After making adjustments for demographic variables, such as population size, density, and age distribution, every day that states delayed declaring an emergency was associated with a 5% increase in mortality.
Each day’s delay in closing schools was associated with a 6% increase in mortality.
By redefining the first confirmed death in a state as day one, the researchers were able to obtain 28-day mortality data for all 50 states.
The results were very similar across the states. In fact, all the associations held up after excluding the states of New York and New Jersey — which had exceptionally high death rates early on in the pandemic — from the analyses.
“The implementation of social distancing measures is fundamentally political, as the process is decided upon by elected officials,” says Dr. Yehya. “Real-time, scientific evidence of the efficacy of these measures will be helpful for informing future policy decisions.”
In their paper, the authors conclude:
“To our knowledge, this is the first demonstration of an association between statewide social distancing orders and mortality during COVID-19. Our results support early social distancing as a nonpharmaceutical intervention for reducing mortality.”
The researchers acknowledge several limitations of their analysis that may have skewed the results.
For example, some local school districts closed their schools independently of state orders. However, other research suggests that this only affected about 16% of the population.
In addition, the COVID-19 death rates that the researchers used in their analysis were based on inconsistent levels of testing and reporting across states. States also used different tests of varying reliability to confirm the number of infections.
The authors report that they were unable to adjust for some other factors that could potentially have biased their results, such as variation among states in terms of people’s access to healthcare and the availability of hospital and intensive care facilities.
For live updates on the latest developments regarding the novel coronavirus and COVID-19, click here.