Effects of weather-related social distancing on city-scale transmission of respiratory viruses

Michael L Jackson; Gregory R Hart; Denise J McCulloch; Amanda Adler; Elisabeth Brandstetter; Kairsten Fay; Peter Han; Kirsten Lacombe; Jover Lee; Thomas Sibley; Deborah A Nickerson; Mark Rieder; Lea Starita; Janet A Englund; Trevor Bedford; Helen Chu; Michael Famulare

We estimated the percent of infections averted by weather-related city-wide disruption by simulating transmission of each of these nine viruses with and without the presence of decreased effective contact rates and compared incidence of infection under both scenarios. The percent of infections averted ranged from 3.0% (95% CI, 2.0-3.7%) for human metapneumovirus to 9.2% (95% CI, 6.2-10.3%) for RSV B (Table 1) . Comparing the pathogens with the highest incidence, influenza A/H1N1 and influenza A/H3N2, we observed that the weather-related disruption occurred shortly before the predicted peak of the influenza A/H1N1 epidemic but early in the course of the influenza A/H3N2 epidemic (Figure 3) . The estimated impact of the disruption was significantly greater for A/H1N1 (7.6% of infections averted, 95% CI, 5.2-8.7%) than for A/H3N2 (3.1% averted, 95% CI, 2.5-3.2%). For A/H1N1, the weather-related transmission disruption appears to have reduced incidence without significant rebound, while the main effect on A/H3N2 was to delay the peak in incidence by an estimated 18 days (95% CI, 17-21 days).