
Study tracks effects of mobility on respiratory virus spread
Mobile device data show how population movements affected circulation of 18 respiratory viruses in Seattle from Nov. 2019 to June 2022.Media Contact: Leila Gray, leilag@uw.edu, 206-475-9809

Mobile device location data show how population movement patterns influenced the transmission of an array of respiratory viruses before, during and after the COVID-19 pandemic.
The findings were published May 16 in Nature Communications. The lead author is Amanda Perofsky, a research scientist for the Brotman Baty Institute for Precision Medicine at the University of Washington and the Fogarty International Center at the National Institutes of Health.
The researchers were interested in how mobility behavior — whether people stay at home or go out, the types of locations they frequent, and the rate of visitors coming into the city — might be connected to the circulation of respiratory viruses. During the pandemic , they considered other behavioral variables in their analysis, such as masking trends and public health interventions to curb the spread of SARS-CoV-2.
During the pandemic, many public health studies used cell phone location data to evaluate the effectiveness of travel bans, stay-at-home orders, gathering restrictions, and school and workplace closures in slowing the spread of SARS-CoV-2.
However, when it comes to common seasonal respiratory infections such as influenza and respiratory syncytial virus (RSV), the relationship between people’s mobility and virus transmissibility has been less explored.
For their modeling study, which covered Nov. 19, 2018, to June 30, 2022, the researchers obtained detailed respiratory virus surveillance data from the Seattle Flu Study. The Seattle Flu Study, launched in the fall of 2018, was an intensive citywide effort to test individuals in the community and in hospitals for over 20 respiratory pathogens. Its purpose was to improve the detection and control of epidemics and pandemics, by tracking the geographic spread of respiratory viruses and learning what contributes to transmission.
The study also incorporated data from the Seattle Coronavirus Assessment Network, launched in March 2020 to detect and understand the spread of SARS-CoV-2.
The scientists compared changes in the daily transmissibility of 18 common respiratory viruses to trends in people staying put, moving about the city, or entering different types of buildings, such as schools, daycares, grocery stores and restaurants.
They tracked three influenza viruses, two respiratory syncytial viruses, four seasonal coronaviruses, rhinovirus, enterovirus, adenovirus, metapneumonia virus, and four parainfluenza viruses. These viruses have different modes of transmission (respiratory droplets versus surfaces), seasonal cycles (winter outbreaks versus year-round circulation), and age groups affected.
A major snowstorm in February 2019, an unusual weather event for Seattle, enabled the researchers to witness a decline in respiratory virus infections that correlated with people being stranded at home due to school, daycare and workplace closures and transportation difficulties.
“This citywide shutdown,” the researchers reported, “led to a conspicuous dip in the incidence of several pathogens.” They mentioned that several viruses rebounded after schools and workplaces reopened, with transmission peaking in mid-March to early April.
The fall and winter before the pandemic emerged in March 2020 began as a typical respiratory virus season for Seattle. The autumn start of school and colleges, increased drop-offs at daycare centers, and the end-of-summer influx of congregants to religious services coincided with the first signs of an increase in flu viruses, RSV, seasonal coronaviruses and parainfluenza viruses.
“For most pathogens,” according to the research report, “the strongest relationship between transmission and mobility occurred at the beginning of the season in early autumn.”
After the first case of community acquired COVID was reported on Feb. 28 near Seattle, mobility levels and respiratory virus transmission declined substantially, and once businesses closed on March 16, fell even more precipitously. However, seasonal outbreaks of some viruses, such as RSV and influenza B, had already ended by late February.
The study also examined the effects of relaxing COVID-19 restrictions in early May as a limited re-opening plan was instituted. Mobility was strongly linked to increasing SARS-CoV-2 transmission when stay-at-home restrictions first relaxed, but this relationship faded after early June, when additional business re-opened.
Mobility was also initially predictive of increasing SARS-CoV-2 circulation at the beginning of subsequent COVID-19 waves, but relationships between mobility and transmission were less clear compared with the stay-at-home period.
As new variants of SARS-CoV-2 emerged, the researchers found that associations between mobility and transmission actually varied across different waves.
Certain endemic common cold viruses — rhinovirus, adenovirus, and enterovirus — came back quickly when stay-at-home orders were lifted in summer 2020, the study found. Their rise was preceded by or simultaneous with an increase in foot traffic to daycares, restaurants, and mass transit stations, and with larger-scale movements, such as flows of people between neighborhoods as well as out-of-state visitors traveling to Seattle.
The study researchers explained that these three pathogens are non-enveloped viruses, which can live on hands and surfaces for a long time and resist disinfectants. Also, these viruses spread year-round, and people infected with them tend to shed them for a long while. Non-enveloped viruses may have continued to spread during lockdown restrictions, and slight rises in mobility were enough to fuel their spread as soon as Seattle reopened.
The remaining viruses in the study, such as influenza, RSV and seasonal coronaviruses, are all enveloped viruses, which are less stable outside of the human body, easier to kill with handwashing and disinfection, and have defined seasonal waves. These viruses did not recirculate in Seattle until 2021.
In late 2021, the surge in Omicron cases was met by a decline in endemic virus cases. These reductions followed or occurred during a time when people were less mobile in response to the first Omicron wave.
The researchers built predictive models to see if they could use mobility metrics and other variables, such as the proportion of individuals masking in public and weather conditions, to forecast daily transmission rates up to a week in advance for three viruses that circulated throughout most of their study: rhinovirus, adenovirus, and SARS-CoV-2.
They discovered that knowledge of past virus circulation was most informative for predicting future transmission dynamics. Mobility data offered only modest improvements in making such predictions.
Some of the more notable findings of the overall study included the remarkable differences in the timing and extent of the rebound of each endemic pathogen the researchers tracked during the COVID pandemic. They also discovered that mobility was best able to predict transmission rates during times of significant changes in people’s behavior, such as when stay-at-home orders were instituted in March 2020 in response to the newly emerged coronavirus.
Distinct types of mobility, the researchers surmised, may promote different aspects of viral spread. Foot traffic to businesses, schools and religious activities may be a proxy for close contacts or crowded conditions that are ideal for passing along respiratory viruses directly to individuals, through the air or by droplets, while movement within and between neighborhoods or travelers visiting from outside the region might provide opportunities for viruses to enter a population and disperse within it.
The mobility study findings reinforced the association between the spread of viruses and attending school or daycare. Previous studies have shown that children have the highest rate of infection for endemic respiratory viruses and a major source of transmission into the community at large.
The researchers also concluded that, in Seattle, the links between population behavior and the transmission of endemic respiratory viruses were, “overall, stronger and longer lasting than those observed for SARS-CoV-2.”
Worldwide, non-pharmaceutical interventions to control the COVID pandemic disrupted the usual patterns of endemic respiratory infections. They seemed to disappear from the scene, come back later than expected or pop up off-season. In Seattle, when stay-at-home orders were instituted in March of 2020 and mobility was at a near standstill, the transmission of all endemic respiratory viruses swiftly diminished. When movement restrictions eased, the endemic viruses each returned in their own manner.
The return of in-person classes at schools was a key factor in the rebound of enveloped respiratory viruses. They also found that the rebound was associated with the renewed ability to travel into Seattle, because flights can bring in influenza strains from distant locations and spark new outbreaks. Moreover, prior studies suggest that the regional spread of influenza may match closely with people commuting to work again.
Some of the limitations of their study, according to the researchers, pertained to the type of cell phone mobility data they had available and its demographics and geographic scope. They also stressed that results from a single metropolitan region may not hold up in less populated rural areas. They added that, had they been able to collect data on which cell phones were near each other, they might have had better indicators of how in-person social contacts contributed to transmission rates. The length of their study, covering only a few respiratory virus seasons, also makes it difficult to determine if their findings would be the same for longer time spans.
Despite these limitations, the study did show that mobile phone data has potential in providing additional information for surveillance of endemic respiratory viruses, provided that it is clear that links between mobility and transmission depend on the type of virus, the degree of change in population mobility, and the phase of the epidemic for that virus.
Funding for the Seattle Flu Study was from Gates Ventures and the Howard Hughes Medical Institute, as well as from the Brotman Baty Institute for Precision Medicine. Samples collected in Pierce County were funded by the Tacoma-Pierce County Health Department. Support for the study was also provided from the Centers for Disease Control and Prevention contract 75D30122C14368 and from the in-house research division of the Fogarty International Center at the National Institutes of Health.
For details about UW Medicine, please visit http://uwmedicine.org/about.