Study: Nonpharmaceutical interventions can control or eliminate COVID-19
Multi-university modeling shows that early termination of strict social distancing could trigger devastating second wave
Editor's note: As of May 11, America’s reopening has begun, just weeks after the coronavirus had the country on lockdown. Already, more than half the states have started to reopen their economies.
Since the novel coronavirus is transmitted among people who come in close contact with each other, the implementation of strict social distancing measures has been the primary tool for curbing the spread of the pandemic. As of April 7, 2020, mandatory lockdowns or stay-at-home orders have been imposed in more than 42 states in the U.S. This represents over 95% of the U.S. population, involving approximately 316 million Americans.
Predicting the course or severity of a pandemic, such as COVID-19, as well as the realistic assessment of proposed public health intervention strategies for combating them in real time, is a major challenge to both the public health and the scientific community.
Even as government leaders tout protecting public health as a top priority, some are looking for ways to start “opening up” and loosening stay-at-home restrictions in the hopes of providing some economic stimulation. Some states are proposing guidelines starting in early May that will allow people to get haircuts or eat at dine-in restaurants if businesses limit occupancy, physically distance patrons and check employees for COVID-19 symptoms before their shifts.
The duration and timing of the relaxation of strict social distancing measures are crucially important in determining the future trajectory of the COVID-19 pandemic.
A new collaborative research study — from Arizona State University, the University of Florida Gainesville, the University of New South Wales and Harvard Medical School — shows early termination of strict social distancing measures could trigger a devastating second wave of COVID-19 with results similar to those projected before they were implemented.
The simulations show that terminating the current strict social distancing by the end of April will result in a significant rebound of COVID-19 burden in as early as July 2020.
With no vaccine yet available, efforts at containing COVID-19 are focused on what nonpharmaceutical measures will have the greatest impact on curbing the spread of the disease.
The team of researchers recently developed a new mathematical model, which may be the first of its kind, that incorporates five nonpharmaceutical interventions: social distancing, quarantine of suspected cases, isolation of confirmed cases, contact tracing and testing, and use of face masks in public.
The scientists found that use of face masks in public could lead to the effective control, or elimination, of COVID-19 in the state of New York and the entire U.S. if the coverage levels are high enough. The compliance level needed decreases if wearing face masks in public is combined with a strict social distancing strategy.
Their results also showed quarantine of suspected cases and contact tracing have only marginal impact in minimizing COVID-19 burden, measured in terms of minimizing COVID-19 hospitalizations.
The study, Mathematical assessment of the impact of nonpharmaceutical interventions on curtailing the 2019 novel Coronavirus, was published this week in the journal Mathematical Biosciences. The study was supported in part by funding from the Simons Foundation and the National Science Foundation.
The project is a collaboration with members of the PLuS Alliance, which combines the strengths of leading research universities on different continents, including Arizona State University and University of New South Wales Sydney, to develop sustainable solutions to society’s global challenges, such as global health.
One of the lead authors and a graduate student in ASU’s School of Mathematical and Statistical Sciences, Enahoro Iboi, shared his concern after hearing about the state of Arizona’s plans to reopen in May.
“Lifting the current stay-at-home orders too soon in Arizona may likely lead to a second wave in both COVID-19-related daily mortality and cumulative cases,” said Iboi.
“This is probably the first most detailed modeling study to give a robust assessment of the various lockdown relaxation scenarios in terms of number of additional mortalities we can expect,” said Arizona State Foundation Professor Abba Gumel.
“We currently have about 10,000 cases and 400 mortality, or about 4% case-fatality ratio, in the state of Arizona. I really hope they do not do what they are planning to do. It is going to cause a lot more deaths,” said Gumel.
The researchers developed a new mathematical model for studying the transmission dynamics and control of the COVID-19 pandemic in the U.S. and in particular the state of New York, the epicenter of COVID-19.
“This study is so important because we need thorough, well-designed, informative data-driven research to guide public health action,” said Matthew Scotch, associate professor of biomedical informatics in the College of Health Solutions and assistant director of the Biodesign Center for Environmental Health Engineering at Arizona State University.
The model is a Kermack-McKendrick, compartmental, deterministic system of nonlinear equations. It incorporates features pertinent to COVID-19 transmission dynamics and control, such as the quarantine of suspected cases, the isolation or hospitalization of confirmed, contact tracing, social distancing, and the use of face masks in public. The model uses available COVID-19 mortality data, which is more reliable than case data, and provides a realistic real-time assessment and estimate of the burden of the pandemic in New York.
The researchers first used the model to simulate the impact of social distancing, which for this study included not only individuals staying six feet apart, but also closures of schools and nonessential businesses, staying at home, avoiding crowded events and large gatherings, and moving in-person meetings online.
They found that the state of New York and the entire U.S. could have faced between 100,000 to 200,000 deaths had the strict social distancing measures not been implemented in March.
The effect of the timing of when to terminate the current strict social distancing protocols was also monitored. The simulations showed that terminating strict social distancing, face mask usage, contact tracing, quarantine and isolation by the end of April will result in 144,000 deaths in New York state, representing a 37% increase, while the nation will record up to 156,000 deaths.
The study shows that early termination of the current strict social distancing measures by the end of April will result in such a catastrophic outcome it would be as if all the gains made by social distancing and other mitigation measures will essentially be lost.
However, if the strict distancing measures were not terminated until the end of May, the cumulative mortality figures are projected to be 91,800 for New York state and 118,300 for the entire U.S.
If the social distancing measures are terminated at the end of June, the projection for the cumulative mortality figures are 33,200 for New York state and 50,300 for the entire US. This represents 68% and 69% reductions, respectively, in the baseline cumulative mortality.
This study clearly shows that the clamor to “reopen” and relax or terminate the social distancing measures that have proven to be hugely successful would undoubtedly trigger a devastating rebound of COVID-19 in both New York state and the entire U.S.
Contact tracing involves searching for individuals with whom a confirmed case has closely interacted within a certain time frame, such as two days prior to the onset of symptoms, then interviewing, testing and isolating or hospitalizing that contact if they have the disease.
In this study, contact tracing is very much interlinked with testing. Contact tracing is carried out after a confirmed case is diagnosed, following testing and diagnosis.
Results show that while contact tracing is important in reducing the size of the pandemic peak number of new COVID-19 cases, investing much resources toward contact tracing beyond the baseline rate might not be cost-effective.
Simulations were further carried out to assess the impact of widespread use of masks in public. The results show a marked decrease in the number of hospitalizations, for both New York and the entire U.S., with increasing use of more effective masks and with wider coverage.
Additional simulations assessed combining the strict social distancing strategy with using a moderately effective mask. If only 30% of the residents of the state wear masks with efficacy of about 50%, the study shows that COVID-19 can be eliminated from the state of New York.
Similar results were obtained for the entire U.S., showing a mere 10% of mask-wearing compliance with a 50% efficacy mask combined with strict social distancing will be needed for COVID-19 elimination.
“Community transmission can be controlled if lockdown measures are partially lifted, as long as individuals wear a mask in public and large-scale randomized testing is done to really give us a clear assessment of where we are,” said Gumel. “And the testing is complemented with the strict containment strategy of rapidly isolating the confirmed cases and isolating their contacts.”
“Wide-scale randomized testing will let us know if we need to stop the partial lifting of restrictions, or if we need to further relax the lockdowns.”
The study suggests that COVID-19 is a disease that appears to be controllable using basic nonpharmaceutical interventions, particularly social distancing and the use of face masks in public, especially when implemented in combinations.
The factors that are critically important to the success of the COVID-19 control efforts are the early implementation of these intervention measures and ensuring their high adherence in the community.
“The lockdown-lifting guidelines announced by the White House are actually very reasonable, but it is clear many states are rushing to reopen without even meeting the basic steps of the gating criteria,” added Gumel. “The first step is to ensure that you have a two-week downward trend in new cases. Most of the states that lifted or partially lifted the lockdown have not met this basic first step yet.”