Fomites Could Determine Severity of SARS-CoV-2 Outbreaks in Low-Density White-Tailed Deer (Odocoileus virginianus) Populations

Abstract

The establishment of a reservoir species for zoonotic diseases is concerning for both animal and human health. Severe acute respiratory syndrome coronavirus (SARS-CoV)-2, the coronavirus responsible for the COVID-19 pandemic, has been detected in white-tailed deer (Odocoileus virginianus) in the United States. Since its initial detection, various studies have documented circulation and evolution of SARS-CoV-2 in deer, with human cases suspected of spill-back from infectious deer. A priority for mitigating SARS-CoV-2 outbreaks in deer populations is determining the contribution of direct (via aerosols and physical contact) and indirect (via contaminated objects and media) transmission pathways. We expanded existing epidemiological models founded on direct transmission pathways to include three indirect transmission pathways of infection for simulated deer populations, including contaminated water, food waste, and feed piles. Despite lower infection probabilities and transmission hazards (measured by force-of-infection (FOI)) posed solely by these indirect pathways compared to direct transmission pathways, the addition of indirect transmission pathways increased FOI, which had ramifications for the severity of SARS-CoV-2 outbreaks in simulated deer populations, particularly in populations with low degrees of spread between deer (measured by basic reproductive number; R ₀). We used contact rate models to estimate SARS-CoV-2 spread across deer range in the United States and identified widespread potential for indirect transmission to increase the severity of outbreaks in low-density deer populations. These results indicate that indirect transmission pathways need to be considered in the management of white-tailed deer as a reservoir species for SARS-CoV-2.

Keywords: SARS-CoV-2; fomite; indirect transmission; transmission risk; white-tailed deer.

Figure 1

A conceptual diagram of the epidemiological effects of indirect transmission in addition to direct transmission in SARS-CoV-2 spillover and spread in white-tailed deer (Odocoileus virginianus). Spillover occurs when infectious humans either infect deer through direct transmission via aerosols (solid blue arrow) or indirectly through contaminated fomites and media (dashed blue line). Once deer are infected, SARS-CoV-2 outbreaks are maintained by direct transmission between deer (solid black arrow) or through contaminated fomites and media (dashed black arrow). We quantified the per-capita rate that deer become infected (FOI) and the spread of SARS-CoV-2 within a deer population (basic reproductive numbers; R₀). We projected SARS-CoV-2 prevalence in a deer population and related increases in FOI in two scenarios, one with direct and indirect transmission pathways and one with only direct transmission and we quantified how each transmission scenario contributed to changes in outbreak metrics. Created in BioRender. Rosenblatt, E. (2025; https://BioRender.com/g60d995).

Figure 2

SARS-CoV-2 infection characteristics of indirect and direct transmission pathways of spillover to white-tailed deer (Odocoileus virginianus) (A). Distributions of FOI and probability of infection given exposure are summarized by quartiles along a log₁₀-transformed axis. Cumulative FOI increased when both direct and indirect transmission are considered (B). Shifts in cumulative probability distributions for four FOI categories are summarized as the proportion of simulations falling within each category. Break points for FOI categories are indicated on the x-axis (log₁₀-scale).

Figure 3

Sensitivities of average prevalence, persistence probability, and incidence proportion to categories of spillover (FOI) and spread amongst white-tailed deer (Odocoileus virginianus; R₀) in a wild population exposed to both direct and indirect transmission pathways. We defined R₀ categories of R₀ < 0.5, 0.5 ≤ R₀ ≤ 1, 1 < R₀ ≤ 2, 2 < R₀ ≤ 3, and 3 < R₀, spanning conditions where spread is unlikely, spread is characterized by stuttering chains, spread is likely sustained, and spread is widespread, respectively [23]. FOI categories include <1.096 × 10⁻⁵ (1), 1.096 × 10⁻⁵−4.571 × 10⁻⁵ (2), 4.571 × 10⁻⁵−1.047 × 10⁻⁴ (3), and >1.047 × 10⁻⁴ (4). Points indicate median values from simulations that fall within each combination of FOI and R₀ categories, with error bars indicating either 2.5 and 97.5 quantiles of simulation values (average prevalence and incidence proportion) or 95% confidence intervals (persistence probability).

Figure 4

(A) Median transmission of SARS-CoV-2 between white-tailed deer (Odocoileus virginianus; R₀) was dependent on deer density and the proportion of wooded habitat available. Using Habib et al.'s [42] contact rate model, we estimated R₀ for areas with 1−11.6 deer/km² and with 12%–63% wooded habitat. The vertical black line indicates the partition between two deer density categories identified by Hanberry and Hanberry's [44] county-level density estimates. Horizontal black lines indicate wooded habitat availability values included in Habib et al.'s [42] contact rate model. We estimated sensitive epidemiological conditions (0.5 < R₀ < 1) across a range of deer density and wooded habitat availability. (B) R₀ estimates covered 57% of low-density white-tailed deer range (between dashed white lines). Note that pixel bounds align with Hanberry and Hanberry's [44] county-level density estimates (vertical bounds) and wooded habitat availability considered by Habib et al. [42] (quantified using 42; horizontal bounds). (C) Forty-five states in the contiguous United States have areas that fall within the scope of inference from Habib et al.'s [42] contact rate model, with variable proportions of deer ranges in these states that are predicted to have epidemiological conditions sensitive to increases in FOI from indirect transmission pathways.

Figure 5

Median spread (R₀) of SARS-CoV-2 between white-tailed deer (Odocoileus virginianus) across their range in the contiguous United States. We estimated that outbreaks in areas where R₀ is predicted between 0.5 and 1 have the greatest sensitivity to increases in cumulative FOI, including those increases that may be caused by indirect transmission. Epidemiological models that only consider direct transmission pathways may underestimate the severity of outbreaks in these areas. Predictions were made at a scale of 100 km² pixels across wild deer range, using available county-level deer densities [43] and classifications from the National Land Cover Database classes [42]. Pixels with conditions outside the parameter space of Habib et al.'s [42] contact rate model are indicated by solid light gray coloration.