Novel respiratory viruses—such as SARS-CoV-2— arise periodically, causing global pandemics of varying severity. Therefore, early pandemic preparedness responses should be put in place for inevitable future outbreaks. Early, community-level mitigation strategies are crucial for decreasing viral transmission, infection, and death. However, the impact and dynamics of city-wide social distancing as an intervention to reduce influenza or respiratory virus spread are unknown. Most previous data has been collected from studying weather-related or routine school closures, which primarily affect children, precluding extrapolation of this data to estimate the effects of social distancing among an entire community. However, the unusually high snowfall in February, 2019 in western Washington State led to widespread school, workplace, and community closures, presenting a unique opportunity to study the effects of short-duration, high-intensity social distancing at the city-wide level.
The Seattle Flu Study began during the 2018/2019 influenza season and is a Seattle-area surveillance project aimed at evaluating city-level transmission of influenza and other respiratory pathogens. Under one arm of the study, respiratory swabs taken from community members who sought care at local hospitals for respiratory illness were tested for infection with common respiratory viruses such as respiratory syncytial virus (RSV), human corona virus (hCoV), rhinovirus (HRV), and influenza. Dr. Michael L. Jackson (Kaiser Permanente Washington Health Research Institute) and colleagues from the Seattle Flu Study and the Fred Hutch Vaccine and Infectious Disease Division used data collected under this arm of the Seattle Flu Study to determine how snowfall-induced social distancing over a short period during 2019 impacted respiratory virus transmission in the Seattle area. They recently published this work in BMC Infectious Diseases.
To model the effects of city-wide closures on virus transmission, the authors used the Susceptible-Exposed-Infectious-Recovered (SEIR) framework—which models virus transmission based on the proportions of community members who are susceptible, infected but not yet infectious, infectious, and recovered and not susceptible, respectively—the authors modeled transmission of nine respiratory viruses between February 3 and 15, 2019. To control for a possible decrease in diagnosis of respiratory infection due to an inability to travel to a hospital during the heavy snowfall, the authors compared traffic and regional hospital data during the snowfall to other dates between January and March 2019 and estimated decreases in traffic and hospital visits in their model. They found that the contact rate, or the rate at which susceptible people (S) become exposed (E), ranged from a 16.2% decrease for coronavirus to a 94.6% decrease for RSV, suggesting that city-wide closures and the resulting social distancing accounted for a large decrease in contacts between people infected with respiratory viruses and those susceptible to become infected, likely resulting in fewer infections. To determine the percent of infections prevented by weather-related, city-wide social distancing, the authors simulated transmission of the nine viruses with and without the decreased contact rates. They found that between 3 and 9.2% of infections, depending on the virus, were averted during this period, suggesting that even short periods of high-intensity social distancing can meaningfully prevent viral spread.
This study also presented a unique opportunity to study the timing of social distancing on viral transmission. During the 2019 snowfall, influenza A/H1N1 and influenza A/H3N2, the pathogens with the highest incidence in samples collected during the snowfall period, were circulating at different stages in their epidemic dynamics: the snowfall came just before the predicted peak of influenza A/H1N1 transmission, while influenza A/H3N2 was early in the course of the epidemic. Because the decrease in total influenza A/H1N1 infections was much greater than A/H3N2 infection, the authors concluded that city-wide disruption later, near the peak of transmission, has a greater impact on final incidence than does disruption early in an epidemic, which causes an initial decrease in infections that rebounds to an incidence comparable to no disruption. To further interrogate this finding, they next modeled the transmission of influenza A/H3N2 with a 14-day social distancing disruption at various points during the epidemic. Likewise, they found that an early disruption has little effect on final infection rates, while a disruption near the peak of transmission significantly reduces incidence of infection.
This work supports previous findings that have shown that social distancing disruptions initiated early in a pandemic have little effect on the final infection incidence, an important consideration for pandemic control measures. These results suggest that, instead, aggressive social distancing measure are best implemented slightly later, closer to peak transmission. Additionally, this study demonstrates that short-term, high-intensity social distancing measures have significant impacts, decreasing contact rates and total infections with respiratory viruses. These findings are important for pandemic response measures for SARS-CoV-2 and inevitable future pandemics, as short-term social distancing may be more feasible to enforce than pandemic-long social distancing, while still preventing a large number of infections. Going forward, intermittent social distancing events could be deployed as a tool to manage disease spread even after viral containment is not possible.
Jackson ML, Hart GR, McCulloch DJ, Adler A, Brandstetter E, Fay K, Han P, Lacombe K, Lee J, Sibley TR, Nickerson DA, Rieder MJ, Starita L, Englund JA, Bedford T, Chu H, Famulare M; Seattle Flu Study Investigators. Effects of weather-related social distancing on city-scale transmission of respiratory viruses: a retrospective cohort study. BMC Infect Dis. 2021 Apr 9;21(1):335. doi: 10.1186/s12879-021-06028-4. PMID: 33836685; PMCID: PMC8033554.
UW/Fred Hutch Cancer Consortium members Deborah Nickerson, Lea Starita, and Janet Englund contributed to this work.
The Seattle Flu Study was supported by the Brotman Baty Institute for Precision Medicine.