Intervention Serology and Interaction Substitution: Modeling the Role of 'Shield Immunity' in Reducing COVID-19 Epidemic Spread

Joshua S Weitz; Stephen J Beckett; Ashley R Coenen; David Demory; Marian Dominguez-Mirazo; Jonathan Dushoff; Chung-Yin Leung; Guanlin Li; Andreea Magalie; Sang Woo Park; Rogelio Rodriguez-Gonzalez; Shashwat Shivam; Conan Zhao

We use the baseline epidemiological parameters and seed an outbreak with a single exposed individual until the outbreak reaches 0.1% total prevalence, e.g., 10,000 individuals infected out of a population of 10,000,000, at which point a shielding strategy is implemented. Outbreak scenarios differ in transmission rates, with R 0 = 1.57 and 2.33 in the low and high scenarios, respectively. Early estimates of R 0 from Wuhan are consistent with a 95% CI of between 2.1 and 4.5 [28] , putting our high scenario on the conservative end of estimated ranges. However, the R 0 of the high scenario we examine here is consistent with the range of 2.0 to 2.6 considered by the Imperial College London group [5] , and with the median of R ef f = 2.38 (95% CI: 2.04-2.77) as estimated via stochastic model fits to outbreak data in China that accounts for undocumented transmission [26] . Moreover, control measures reduce transmission, and our low scenario is consistent with estimates of R ef f = 1.36 (95% CI: 1.14-1.63) in China from Jan. 24 to Feb. 3 after travel restrictions and other control measures were imposed. Figure 3 shows the results of comparing interventions to the baseline case. As in the simple SIR model, shielding (on its own) could potentially decrease epidemic burden across multiple metrics, decreasing both the total impact and shortening the peak event. In a population of size 10,000,000 for the high scenario, the final epidemic predictions are 71,000 deaths in the baseline case vs. 58,000 deaths given intermediate shielding (α = 2), and 20,000 deaths given enhanced shielding (α = 20). In a population of size 10,000,000 for the low scenario, the final epi-. CC-BY 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity. demic predictions are 50,000 deaths in the baseline case vs. 34,000 deaths given intermediate shielding, and 8,300 deaths given enhanced shielding. The majority of deaths are in those ages 60 and above, despite the lower fraction of individuals in those ranges (see Figure 3 ), consistent with estimates in related COVID-19 models [5, 25, 26] and from outbreaks in Italy and China [29] . Note that our simulation results consider impacts based on shielding alone; whereas ongoing restrictions via social distancing and shelter in place orders will reduce interaction rates (a point we revisit later). The effectiveness of shielding depends on the product of the number of potential shields identified and their effective substitutability, i.e, αR, combining identification of and interaction rate by shields. The population-scale impacts of shielding depends on multiple factors, including demographic distributions, the fraction of asymptomatic transmission ( [24, 26] ), and the duration of immunity. The SI treats each of these items at length. First, we find that populations with a strongly right-shifted demographic distribution will receive more potential benefits from shielding. Even though there are fewer recovered individuals between the ages of 20-60 to draw from (in a relative sense), the impact of shield immunity is greater. We find that the relative reduction in deaths via shield immunity is proportional to the relative differences in the fraction of population over 60 (e.g., see SI for details on US-state level analysis, similar results hold for countries like Italy where more than 23% of the population is older than 65 and nearly 30% is older