Skip to main content

Our take —

This study, which was available as a preprint and thus had not yet been peer reviewed, uses county-level SARS-CoV-2 testing data to show that the Sturgis motorcycle rally likely led to substantial increases in cases in the local community where the rally took place. However, there is considerable uncertainty surrounding the broader, national impact of the rally and its associated costs given limitations in the methodological approaches used. Results from this study should be interpreted cautiously.

Study design

Ecological

Study population and setting

This study assessed the spread of SARS-CoV-2 and social behaviors related to a large annual motorcycle rally event in Sturgis, South Dakota in early to mid-August 2020. The population consisted of both local residents of Sturgis as well as attendees of this rally, who travelled to Sturgis from all over the United States. The study examined a number of outcomes, mainly COVID-19 cases and social distancing behaviors, using a combination of county-level aggregated cases from the CDC and personal movement information based on cell phone location data from SafeGraph, from July 6 to August 31, 2020. The main methods used to estimate the impact of the rally on COVID-19 cases were based on comparing changes in regional COVID-19 cases relative to a synthetic control (a method for constructing a “control” from weighted average of other locations without exposure to the rally) to estimate what would have happened had the rally not taken place. Comparator regions were selected across the US (excluding adjacent regions), based mainly by matching on case rates, urbanicity, and population-related characteristics. Additional analyses were performed to see if impacts were larger for places that had more of its residents attending the rally. Finally, the authors estimated the costs of infections attributed to this rally by multiplying a cost per-case estimate (from another study) with their estimate of the number of additional cases resulting from the rally.

Summary of Main Findings

The study makes a large number of claims regarding the impact of the Sturgis event on behaviors, the spread of SARS-CoV-2, and costs to the healthcare system. First, the study finds substantial increases in COVID-19 cases by an additional 6 to 7 cases per 1,000 population locally within Meade County (the county in which the event took place), 3.6 and 3.9 cases per 1,000 population in the state of South Dakota, and ultimately over 266,796 additional cases nationwide attributable to the rally. The study found that social movement, both from residents and non-residents, increased substantially due to the rally. Costs attributable to these additional infections were estimated to be $12.2 billion.

Study Strengths

The key strength of this study is in the aggregation and documentation of cases over time and place, in combination with movement data. At least for the local results in Sturgis, the data demonstrates some impact. The case data show relatively stable trends prior to the event and clear changes around the event, with little reason to believe that the changes in cases could have been caused by anything but the event. The overall conclusions that the Sturgis event caused a large increase in COVID-19 cases and infections are likely to be relatively robust to the specific statistical methodologies used.

Limitations

While large and substantial increases in the local cases are clear from the observed data, the construction of the synthetic counterfactuals and the associated data analyses used to obtain nationwide estimates were relatively weak. In the synthetic control construction, there appear to be very few actual regions matched, and they are regionally clustered (e.g., in Texas or the North East), exacerbating the risk of confounding by concurrent changes and unobserved characteristics. While the assumptions underlying the synthetic control are plausible for the analyses considering the local area around Sturgis, they become decreasingly plausible as the analyses move to larger geographic regions. In addition, the analyses do not take changes in testing behaviors into account. People at and surrounding the rally may have been more likely to seek testing for SARS-CoV-2 due to the rally and public health messaging. This increase in testing could lead to an overestimate of the impact of the rally on infections. Ranges of uncertainty on estimates of rally impact nationwide are not reported. These are likely to be large, given the relatively small numbers of regions that contributed to the overall analysis. The methods behind estimating the cost to the healthcare system due to SARS-CoV-2 infections are largely undiscussed, and rely entirely on a cost estimate from a separate paper. We note that cost estimates are provided only at the national level, which was the least reliable level of aggregation, and as such should not be considered informative.

Value added

This analysis provides evidence of how large gatherings such as the Sturgis rally can greatly increase the local spread of SARS-CoV-2. It achieves this through combining data from routinely collected aggregated case data over time, in combination with evidence gathered from personal location data over time from cell phones.

Our take —

This study sought to examine workplace-associated COVID-19 outbreaks in Utah comparing across industry sectors. The most represented industries were manufacturing, construction, and wholesale trade. Hispanic and Black, Indigenous, and other people of color were at increased risk of infection compared to the overall population likely due to overrepresentation in these industries. The study likely did not include smaller workplaces in its analysis, and could not account for workplace changes due to stay-at-home orders or restrictions, thus likely underestimating attack rates amongst those who were at shared work locations.

Study design

Ecological, Other

Study population and setting

The study objective was to report the number of COVID-19 cases traced to workplace settings in Utah as determined by the state’s COVID-19 surveillance system. From March 6 to June 5, 2020, 277 COVID-19 outbreaks were reported, representing 1,389 COVID-19 cases out of 11,448 across the state (12%). Workplace outbreaks were defined as having two or more laboratory-confirmed cases within the same 14-day window among coworkers at the same facility. Utah Department of Health (UDOH) investigators collected the addresses and/or business names for all outbreaks and classified them into 20 industry sectors as determined by the North American Industry Classification System (NAICS), as obtained by the Division of Corporations and Commercial Code business registry the Utah state government maintains. Cases per 100,000 workers were calculated using estimates from the 2019 Census Quarterly Workforce Indicators. Race/ethnicity information, hospitalization status, and number of severe outcomes were also collected.

Summary of Main Findings

Of the 277 outbreaks in the state, 210 were linked to workplaces (75.8%), representing 1,389 cases. The most represented industries were manufacturing (20%), construction (15%) and wholesale trade (14%), which made up the majority of cases (806 total cases in these three sectors). The workplace outbreak attack rate was 106.4 cases per 100,000 workers overall, and was highest among manufacturing (339.4 per 100,000 workers) and wholesale trade (377.0 cases per 100,000 workers). Of the 1335 cases with race/ethnic data available, 73% were among Hispanic or Black, Indigenous and people of color (N=970). 85 were admitted to the hospital (6%) and 40 had severe outcomes (3%). The median cases per outbreak was 4, with ranges from 2 to 79 cases. Compared with people in the state 15 years or older, people with workplace-associated COVID-19 tended to be older (41 years on average, compared to 38 years), more likely to identify as Hispanic (56.4% vs. 39.8%), and be male (61.4% vs. 50.6%).

Study Strengths

The study used the Utah surveillance system to identify all workplace outbreaks, and were able to report on a number of important sociodemographic disparities with their individual-level information. Additionally, using the population of working age in the state, they were able to examine whether those with workplace-related COVID-19 were getting sick at a higher rate or not, which has important insights for the type of workers being placed most at risk. Additionally, they used lab-confirmed diagnosis of COVID-19, which reduced misclassification in their results (though they do not state what type of lab test was used).

Limitations

Grouping by industries shows important trends, but further disaggregation for the main categories (e.g., what type of wholesale trade or manufacturing) would given further insight into what conditions may be leading to the increased risk of outbreaks. Additionally, they do not collect temporal trends that may have altered the probability of an outbreak, such as mandatory closures, and may have affected certain industries more than others. If a handful of industries remained closed throughout the study period, then they would have fewer outbreaks regardless of the actual working conditions, and would not be a valid comparison to industries who did not face these closures. Additionally, worker-to-worker transmission could not be confirmed, and outbreaks that had two individuals who were independently infected would still be included in the analysis. Outbreaks in smaller workplaces would also likely be left out of this analysis.

Value added

This is one of the largest by-sector workplace analyses in the US that also collected race/ethnicity and age data to further identify populations most at risk for infection.

Our take —

The study compared the all-cause mortality in New York City during the H1N1 pandemic in 1918, and the COVID-19 pandemic, primarily in 2020. The study found that the H1N1 pandemic had an increased mortality ratio of 2.80 compared to prior years, and that there was a 4.15 rate ratio for the COVID-19 pandemic compared to prior years. The study was limited in the actual comparability between the H1N1 pandemic, given that the century of medical and technological changes that occurred that may also have impacted the mortality rates. Therefore, this study offers important context to the current pandemic compared to that of the past, demonstrating the COVID-19 pandemic had greater relative mortality than the 1918 H1N1 pandemic, though it remains a crude ecological estimate overall.

Study design

Ecological

Study population and setting

The study objective was to investigate whether the excess deaths in New York City during the 1918 H1N1 pandemic were comparable to those during the initial months of the COVID-19 pandemic. The analysis used public mortality data from the CDC from 1914 – 1918, the NYC Department of Health from 2020, and the US Census from 2017 – 2020. It calculated the all-cause mortality rates per 100,000 person-weeks in these years, and compared them to one another via incident rate ratio. It used mortality from the 61-day period corresponding to the peak of the respective pandemics. In other words, for the 1918 pandemic, it used October and November, 1918 for 61 days total, and calculated the person-month mortality rate for October to November in 1914 to 1917; for the 2020 pandemic, data from March 11 to May 11,2020 for 61 days total as well were used, and calculated the mortality rate based on comparisons between this period and March 11 to May 11 for 2017 to 2019.

Summary of Main Findings

The study found an incidence rate of 287.17 (95% CI: 282.71 – 291.69) deaths per 100,000 person-months during the influenza pandemic, with a 2.80 incident rate ratio (95% CI: 2.74 – 2.86) comparing when the pandemic began in 1918 to the prior period’s average mortality before from 1914 to 1917. For COVID-19, there were 202.08 deaths per 100,000 person-months (95% CI: 199.03 – 205.17), with a 4.15 rate ratio (95% CI: 4.05 – 4.24) comparing the mortality rate in 2020 to the prior period’s average mortality from 2017 to 2019. Comparing the H1N1 pandemic and the COVID-19 pandemic, there was a rate ratio of 0.70 (95% CI: 0.69 – 0.72).

Study Strengths

The study drew on publicly available datasets using an outcome that is often well-measured (all-cause mortality) even in 1918. By also examining the years before each pandemic, the study sets up an important baseline of mortality. Given there is a 100-year difference between each pandemic, using more recent years as a baseline is important to better account for the technological and medical advances that have occurred in the interim. The study used the 61-day period corresponding to the peak of the pandemics in order to focus on deaths likely due to these pandemics, as opposed to taking the entire year which is subject to more fluctuations in mortality due to other causes.

Limitations

The study compares the 1918 pandemic to the 2020 pandemic when there was a number of health and safety measures and technological/scientific limitations during 1918 that may inflate the mortality rate greater than what would have occurred based on the 1918 pandemic’s virulence alone, leading to a reduced rate ratio when comparing 2020 to 1918 due to this larger denominator. The study compared the all-cause mortality rate, which is assumed to be related specifically to the pandemic, and groups individuals who died directly due to the pandemic because of infection with individuals who died from other causes, potentially due to disruptions in receiving hospital care, reduced employment resulting in other disparate and social factors increasing the risk for death, etc. During this 61-day period, also, there may be different degrees of social distancing or other mitigation measures that impact the indirect mortality estimates for the pandemic overall, but are missed by this 61-day window.

Value added

This study directly compares the all-cause mortality changes from the 1918 H1N1 pandemic and the COVID-19 pandemic to better contextualize the changes in death rates seen in New York City.

Our take —

This analysis of smartphone geolocation data associated with 6,644 nursing homes across 23 U.S. states, published as a working paper and not yet peer reviewed, showed that facilities are highly connected, with 7% of phones appearing in a given facility appearing in at least one other facility.  This high connectivity is likely due to shared staff, and persisted even after the implementation of national guidelines designed to limit mobility between nursing homes. While these data highlight vulnerabilities to COVID-19 transmission, rather than recreate actual COVID-19 transmission pathways, the observed associations of nursing home centrality and connectivity with higher COVID-19 cases suggests additional regulations, like single-facility staffing, may be needed to curb COVID-19 outbreaks in nursing homes.

Study design

Ecological, Modeling/Simulation

Study population and setting

Investigators aggregated nursing home resident COVID-19 case data reported to 23 state health departments in the United States through May 31, 2020. Geolocation smartphone data from March 13 to April 23, 2020 were mapped onto COVID-19 case reports to identify mobility patterns among staff, contractors, and residents across 6,644 nursing homes. Adjusting for compositional characteristics of nursing homes (e.g., demographic factors, quality measures), the investigators assessed the relationship between COVID-19 cases at nursing homes and measures of nursing home connectivity (to other homes).

Summary of Main Findings

Even after guidelines restricted social visitors at nursing homes beginning March 13, 2020, approximately 7% of mobile phones that were identified in a given nursing home during the study period were found in at least one other home. Nursing homes were also highly connected (mean number of connections: 15), though these estimates varied widely by state. In regression analysis, nursing home connectivity measures – including the number of other homes to which a nursing home was connected, the number of mobile phones identified in multiple nursing homes, and the number of shared contacts between a nursing home and other homes with high connectivity – were significantly associated with higher cumulative COVID-19 case counts. Higher cumulative COVID-19 cases were also associated with nursing homes in urban areas, more beds, and higher proportions of Black residents (>25%) and Medicaid recipients (>50%).

Study Strengths

Investigators leveraged available geolocation data to identify networks of epidemiologically linked nursing homes vulnerable to COVID-19 transmission. Investigators performed sensitivity analyses to confirm that their findings were robust to COVID-19 prevalence and reporting differences across jurisdictions.

Limitations

The study could not determine whether individuals whose phones appeared in multiple nursing homes were the actual source of SARS-CoV-2 transmission, since the outcome measure was COVID-19 cases aggregated to the nursing home level. Additionally, while geolocation services were useful in constructing nursing home network profiles, the data presented do not offer insights into the duration or frequency of individuals’ exposure to multiple facilities. Lastly, analyses were cross-sectional and did not account for temporal shifts in COVID-19 cases at nursing homes, which could have driven variability in mobility during the observation period that was unexamined in the study.

Value added

This study provides valuable evidence that long-term care facilities are highly connected via shared staff, highlighting potential transmission pathways that threaten vulnerable residents.

Our take —

As other studies have shown, wealthier counties and census tracts were markedly more likely to be able to socially distance compared to less wealthy counties and census tracts.

Study design

Ecological

Study population and setting

The authors compared daily mobility before and after state-level emergency declarations were in place at the county and census tract level in the US. Mobility data were collected between January and April 2020. Four different mobility measures were used: being completely at home (SafeGraph); median distance traveled (SafeGraph); device exposure (the number of unique devices visiting a particular venue, Place IQ); and visitors to retail and recreation places (Google Mobility). Census tracts and counties were classified based on median income quintile to see if county or census tract mobility varied by average income levels. Additionally, the authors controlled for the cumulative number of COVID-19 infected cases in each county at each time point.

Summary of Main Findings

Wealthier census tracts and counties had markedly reduced mobility following statewide declarations. This suggests that wealthier areas are more likely to respond or have the ability to respond to social distancing measures. These results were broadly consistent across the four mobility measures (completely at home, median distance traveled, device exposure, and retail and recreation).

Study Strengths

The study uses multiple measures and databases to compare mobility at the county or census tract level before and after state-wide declarations are in place across measures of county or census tract level income levels, providing a more comprehensive view of changes in mobility.

Limitations

Several other factors beyond income may influence the tendency or ability to socially distance once mandates are in place. For example, one’s occupation will influence the ability to telework, with higher incomes associated with jobs more likely to be able to be done from home. Individual willingness to socially distance may depend on political views, culture, and beliefs and these may cluster at the county or census tract level. Housing also varies by county and census tract, and may influence the ability to distance. As the authors acknowledge, even though we observe differences in social distancing by income level, it is unclear what drives this difference. A second key limitation is that analyses were conducted at the aggregate level, and cannot directly examine whether individual’s income levels relate to mobility.

Value added

Using multiple mobility datasets, the authors show that wealthier areas had markedly greater reductions in mobility compared to poorer areas.

Our take —

In a time-series analysis accounting for variation in population composition and timing of other non-pharmaceutical interventions (e.g., shelter-in-place orders, suspension of mass gatherings), substantial COVID-19 incidence and mortality reductions were attributed to school closures. These were most effective when implemented when incidence rates were lower.

Study design

Ecological

Study population and setting

Investigators aggregated confirmed and probable COVID-19 cases and deaths between March 9 and May 7, 2020, at the state level. Changes in daily cumulative COVID-19 incidence (total cases per 1,000 persons) and mortality (COVID-19 deaths per 1,000 persons) were compared at the state level, before and after the closure of primary and secondary schools (March 13 to 23, 2020) in each state. The timing of other non-pharmaceutical interventions (e.g., stay-at-home orders, business closures) were adjusted for in statistical models.

Summary of Main Findings

Following school closures, significant reductions in both COVID-19 incidence (–62% per week) and mortality (–58% per week) were observed. After collapsing daily COVID-19 incidence estimates before school closings into four groups, reductions in COVID-19 incidence and mortality were greatest in states with the lowest COVID-19 incidence (median: 0.48 cases/100,000 persons) at the time of school closure (incidence: –72% per week; deaths: –63% per week) compared to states in the highest COVID-19 incidence quartile (median: 3.30 cases/100,000 persons), where weekly reductions in incidence and mortality relative to the prior week were –49% and –53%, respectively.

Study Strengths

Investigators modeled a policy interruption (i.e., school closures) in time-series data available for all states in the country to examine the associations between school closure and COVID-19 transmission/mortality. Inclusion of other variables potentially associated with COVID-19 incidence and mortality (i.e., timing of other non-pharmaceutical interventions, COVID-19 testing volume per 1,000 residents, nursing home residents per 1,000 residents, demographic composition of the state population) helped facilitate attribution of observed outcomes to the primary exposure of interest, school closures.

Limitations

The ecological nature of the analysis, which captures data aggregated to the state level, raises questions about unaddressed factors, like travel and mobility, or other non-pharmaceutical interventions driving incidence and mortality differences. Attribution of COVID-19 incidence and mortality reductions to school closures alone is challenging given the assumptions required regarding the timing lags (such as time from exposure to symptom onset) and the multitude of other interventions enacted during similar time frames. There was also little variation in the timing of school closures across states. Lastly, heterogeneities in States’ testing capacities and variability in testing accessibility between March to May could bias incidence and mortality estimates derived from the analysis.

Value added

This is among the first studies to quantify the independent effect of school closures on COVID-19 incidence and mortality in the United States.

Our take —

This is a contact tracing study based in Trento, Italy, published as a preprint and thus not yet peer reviewed. There were 2,812 cases and 6,690 contacts. There was an overall secondary attack rate of 13.3%, with the highest secondary attack rate occurring among contacts over the age of 75 years. However, index cases who were between the age of 0-14 years had the highest percentage of contacts who became infected (22%). There was no routine testing of contacts, so most of them were identified as a case through being symptomatic; thus, there is likely an underrepresentation of the secondary attack rates. However, the finding that the youngest age group of index cases had the highest transmission among their contacts is helpful to note when policy makers are making decisions on reopening schools.

Study design

Cross-Sectional, Ecological

Study population and setting

The provincial agency for health services (APSS) in Trento, Italy conducted contact tracing from March to April, 2020 using a contact tracing website. Data on cases was provided by the central local health unit database while data on contacts of cases was collected by telephone interviews contact tracers from each local health district. A contact was defined as anyone who had contact with a confirmed or probable case within 48 hours prior or 14 days after symptom onset.

Summary of Main Findings

There were 2,812 reported cases, with almost half having up to three contacts each, for a total of 6,690 contacts (890 of whom developed symptoms). Prior to the lockdown on March 10, 2020, (consisting of shutting down schools, universities, and businesses except for grocery stores, pharmacies, and newsstands), the majority of contacts were non-cohabitating family or friends (~37%); however, after March 10, the majority of contacts became household contacts (67%). Ultimately, household contacts comprised 56% of all contacts and non-cohabitating family or friends comprised 27%. The secondary attack rate steadily increased with age (e.g. 18.9% among those 75 year and older vs. 8.4% among those 0-14 years). However, the youngest age group (0-14 years) were more likely to spread infection than any other age group, as 22% of their contacts became infected.

Study Strengths

This study has a large sample of cases and respective contacts. The contact tracing website also provides a centralized resource for data on cases and contacts that can be helpful for future analyses.

Limitations

Classifying a contact as a case was determinant on being symptomatic and having an epidemiological link as there was no routine testing conducted among contacts. Thus, the study is likely reporting an underrepresentation of how many contacts became infected (especially among younger age groups as these groups are more likely to exhibit mild to no symptoms).

Value added

As schools are opening up in the United States and other countries, the fact that secondary infection was more likely to occur in the youngest age group in this study suggests a potential for high levels of transmission both in schools and households if there are not protocols in place to reduce transmission while children are in school.

Our take —

In this peer-reviewed study, the authors analyzed genomes of SARS-CoV-2 and related viruses (from the Sarbecovirus subgenus) to assess the history of recombination in this group and to estimate the timing of SARS-CoV-2 divergence from its ancestors. The results indicate that while recombination is common in sarbecoviruses, the receptor binding region of SARS-CoV-2 does not appear to be a recent recombination with pangolin coronaviruses and likely derives from ancestral viruses in bats. SARS-CoV-2 was estimated to have diverged from its nearest ancestor in bats between 1948 and 1982, indicating that evolutionary ancestors of the virus have been circulating in bats for many years prior to spillover into humans.

Study design

Ecological, Modeling/Simulation, Other

Study population and setting

The study used 68 full coronavirus genomes from the subgenus Sarbecovirus (containing SARS-CoV and SARS-CoV-2) collected from human cases, bats, and other intermediate hosts in northern, central, and southern China since 2002. The goal of the study was to determine the evolutionary history of SARS-CoV-2, specifically to understand the likely source of SARS-CoV-2 in humans (bats, pangolins, or another species), and to identify how long the virus had been circulating in that animal host.

Summary of Main Findings

The authors found that recombination is common among sarbecoviruses, with 67/68 genomes showing evidence of genomic exchange. They find that SARS-CoV-2 and bat-associated RaTG13 are part of a single lineage separate from SARS-CoV and related sarbevoviruses, suggesting that SARS-CoV-2 is the result of a direct (or nearly-direct) zoonotic transmission from bats. Specifically, SARS-CoV-2 did not acquire its variable loop region of the spike protein (containing the receptor binding domain that interfaces with human ACE2) through a recent recombination event with related sarbecoviruses in pangolins. Rather, RaTG13 is the recombinant virus, having acquired its variable loop domain from an as yet unsampled SARS-related coronavirus. The authors also used three different methods to estimate that SARS-CoV-2 appears to have diverged from a common ancestor in bats between 1948 and 1982. The estimated divergence time between the closest pangolin coronavirus to SARS-CoV-2 and the lineage containing SARS-CoV-2 and RaTG13 was between 1851 and 1877, indicating that pangolins likely acquired coronaviruses independently from bats, and were probably not an intermediate host that facilitated adaptation of SARS-CoV-2 to humans. These results indicate that a direct progenitor for SARS-CoV-2 has been circulating in horseshoe bats for decades before spillover into humans.

Study Strengths

The authors used a robust approach to deal with the issue of recombination in phylogenetic inference, which if unaddressed can lead to longer branch lengths and inflated divergence times. The authors also used a robust approach involving multiple prior distributions to estimate evolutionary divergence times, thereby capturing some of the uncertainty that is inherent in time-measured phylogenetic analysis and improving upon previously published results.

Limitations

As with other phylogenetic analyses of sarbecoviruses, inferences about the evolutionary origin of SARS-CoV-2 and the diversification of sarbecoviruses generally are limited by the current availability of genomes related to SARS-CoV-2 in bats and potential intermediate hosts. Additional sampling of bats and potential intermediate hosts around Wuhan and other areas of central China could reveal sarbecoviruses that represent a closer ancestor of SARS-CoV-2 that would provide more information on when and how the virus spilled over into humans.

Value added

This study provides a detailed explanation of the recombination history among sarbecoviruses and in the lineage containing SARS-CoV-2 and its closest relative in bats, RaTG13. The authors demonstrate that SARS-CoV-2 is not a recent recombinant of pangolin and bat viruses, and instead shares features with bat-associated sarbecoviruses. This suggests that spillover from bats to humans may have been direct or near-direct (i.e., a brief residence in an intermediate host). Pangolins do not appear to have been intermediate hosts based on the currently available data.

Our take —

A study, available as a preprint and thus not yet peer reviewed, of 540 dogs and 277 cats in northern Italy showed no evidence of current SARS-CoV-2 infection, but >3% of animals had antibodies suggesting that in areas with active transmission among humans, pets may be occasionally exposed to the virus. Although there is little evidence that dogs or cats develop serious symptoms following exposure or transmit the virus to other animals or humans, household pets should be isolated from owners with COVID-19 to prevent transmission.

Study design

Ecological

Study population and setting

The focal population included 817 pets (540 dogs, 277 cats) in northern Italy surveyed from March to May 2020. Samples mainly came from the region of Lombardy (476 dogs, 187 cats). Oropharyngeal, nasal, and rectal swabs were collected by their regular veterinarian during routine visits. Pets were from households with and without COVID-19 cases. A subset of animals (340 dogs, 188 cats) had full clinical histories available, including breed, sex, age, exposure to COVID-19 cases, and presence of respiratory symptoms. Another subset of animals had serum samples available to test for SARS-CoV-2 antibodies: 188 dogs and 63 cats with full histories and 200 dogs and 89 without historical information.

Summary of Main Findings

None of the 839 swab samples from 817 animals tested positive for SARS-CoV-2 based on a PCR test targeting the nucleoprotein and envelope protein genes; this includes 38 dogs and 38 cats with respiratory symptoms at the time of sampling, and 64 dogs and 57 cats that were living in homes with confirmed COVID-19 cases. Serological testing showed that 13/388 dogs (3.4%) and 6/152 cats (3.9%) had neutralizing antibodies to SARS-CoV-2. Dogs from households with COVID-19 cases and male dogs were significantly more likely to be seropositive. There was also a borderline positive (p = 0.051) rank correlation between the seropositivity of dogs and the human COVID-19 case density (cases per 10,000 people) in provinces with at least ten samples. Similar trends were observed in cats, but statistical tests were not significant.

Study Strengths

The study included a much larger sample size of cats than a previous serological survey in China, and was the first large survey of SARS-CoV-2 infection and serology in dogs. Animals came from a range of age groups and varied in their exposure to human COVID-19 cases, unlike other reports that tested animals from COVID-19 case households only.

Limitations

It is unclear where and when many animals were exposed to SARS-CoV-2, since some were not from households with COVID-19 cases. The small sample size meant that the study was not powered to detect differences in prevalence between animals from COVID-19 case households and non-case households or differences between male and female cats. The correlation between animal seropositivity and human COVID-19 case density across provinces should be regarded with caution because of the very limited number of provinces with sufficient data (n = 4 for cats, n = 6 for dogs).

Value added

This study adds to evidence in China and the United States that dogs and cats produce antibody responses following exposure to SARS-CoV-2 from their owners or within the community. However, research has yet to provide evidence that pets with SARS-CoV-2 infection can transmit the virus back to humans. Additionally, studies have not provided evidence of transmission between cats or dogs outside of the laboratory.

Our take —

In a study of 51 COVID-19 patients from Hong Kong, researchers detected SARS-CoV-2 variants with deletions in and around the spike S1/S2 cleavage site, a feature shared with related viruses in bats and pangolins. These results, combined with results showing S1/S2 site insertions in related bat coronaviruses, indicate that the PRRA motif of SARS-CoV-2 is genetically labile but has a selective advantage over variants in experimental models and infected humans. These results suggest that natural selection (occurring in animals prior to spillover, or in humans shortly after spillover) has fixed the PRRA insertion variant in SARS-CoV-2. Nevertheless, the exact origin of the PRRA motif in SARS-CoV-2 from an animal virus has not yet been determined and will require further sampling to clarify the selective processes that have shaped SARS-CoV-2.

Study design

Ecological, Other

Study population and setting

SARS-CoV-2 contains a unique furin cleavage site amino acid insertion (PRRA) in the S1/S2 region of the spike protein that is absent in related coronaviruses in bats and pangolins. The authors developed a PCR-based assay to detect viral variants with deletions of the PRRA site or upstream of the site and distinguish these variants from normal (wildtype) SARS-CoV-2. The infectivity of wildtype and S1/S2 cleavage site variants was tested in human intestinal cell clusters (organoids; n = 3) and hamsters (n = 2). The presence of viral variants in human COVID-19 patients was assessed in clinical specimens (saliva, nasopharyngeal secretions, throat swabs, and endotracheal aspirates) from 51 adult patients from Hong Kong, China.

Summary of Main Findings

The authors determined that their assay was capable of detecting S1/S2 cleavage site variants at low amounts, down to single viral copies. Inoculation of human intestinal organoids and hamsters with a wildtype virus containing a low copy number of viral variants with PRRA deletions and upstream deletions resulted in infections, although the copies of cleavage site variants were suppressed relative to the inoculum. In the samples from 51 COVID-19 patients, wildtype viral sequences were the most abundant (99% of viral molecules on average). A majority of samples (53%) had PRRA deletion variant molecules, but the mutant population was always a minority of molecules (0.3% on average). Upstream variants were observed in 82% of samples and the mutant represented 1% of detected viral copies. Upstream variants were more common than deletion variants in 93% of samples. These results suggest that cleavage site variants are less competitive than the wildtype virus with the PRRA insertion in cell culture and in natural human infections. The presence of variants had no relationship with symptom severity or sample type; upstream variants were always more common than PRRA deletion variants.

Study Strengths

The authors developed a sensitive assay to detect SARS-CoV-2 genetic variants in clinical samples. Cell culture and a hamster animal model were useful in determining the relative replication fitness of variants compared to wildtype SARS-CoV-2.

Limitations

The study size was very limited, including only 51 patients, three intestinal organoids, and two hamsters. Human samples only came from adult patients in Hong Kong, so the presence of S1/S2 cleavage site variants in other geographic areas cannot be determined from this study. The study also cannot determine how SARS-CoV-2 acquired its unique PRRA motif, or whether that event took place in humans or occurred in an animal host prior to spillover into humans.

Value added

The study provided evidence that natural spike S1/S2 cleavage site variants of SARS-CoV-2, similar to related viruses in bats and pangolins, are currently circulating in humans with COVID-19; however, the low frequency of these variants in natural infections indicates that they are less competitive than the wildtype with the PRRA insertion.