Skip to main content

Jumping back and forth: anthropozoonotic and zoonotic transmission of SARS-CoV-2 on mink farms

Our take —

In this preprint that has not yet been peer-reviewed, the authors detailed the occurrence of SARS-CoV-2 in animals and humans on 16 mink farms in the Netherlands. The timing of outbreaks in animals relative to human cases, along with viral genome sequencing, provides evidence of transmission from humans to mink and vice versa, and multiple independent introductions of viral phylogenetic groups across farms. Countries with mink farms (or other carnivore farms), including Denmark, Spain, the US, and China, should monitor animals and workers for SARS-CoV-2 to prevent large outbreaks and the possibility of viral persistence in carnivore populations.

Study design

Ecological, Other

Study population and setting

This paper summarizes epidemiological investigations of SARS-CoV-2 outbreaks on 16 mink (Neovison vison) farms in the Netherlands between April and June 2020. The first outbreaks occurred on April 23 and April 25, 2020, after which the Dutch Minister of Agriculture implemented a surveillance system for SARS-Cov-2, involving weekly testing of dead animals and full investigation of outbreaks, and ordered all animals on infected mink farms to be culled. Workers on infected farms were tested by PCR or serology within days of the first diagnosis of infection in animals. Full investigations included whole genome sequencing of SARS-CoV-2 from infected animals and farm workers and tracing contacts with other farms that could have led to virus introduction, including movement of vehicles, workers, equipment, manure, or food. A total of 97 farm workers were tested for evidence of infection with the virus via PCR or serological assays. To assess the potential sources of the virus on the infected farms, the authors included 1,775 reference SARS-CoV-2 genomes from a national database, and 34 from anonymized COVID-19 patients in the same postal codes as farms in March and April. Additional genomes (N = 65) from COVID-19 cases in Poland were included because many seasonal workers on the farms come from Poland.

Summary of Main Findings

Screening of farm workers at 16 farms showed that 66/97 (67%) of individuals had evidence of SARS-CoV-2 infection either via PCR or serology. In most cases, virus genomes from human workers clustered phylogenetically with genomes from mink on the same. SARS-CoV-2 viruses grouped into five different phylogenetic clusters with outbreaks on farms having infections of one, or in one case two, virus groups. Such patterns imply that outbreaks on farm were largely from separate introductions and not due to transmission between farms. On two farms (NB1 and NB2), the timing of infections suggest that infected workers first introduced the virus to animals. On farms NB3 and NB7, workers became infected with SARS-CoV-2 only after the virus was spreading in the mink populations on the farms, suggesting that the human cases resulted from transmission from mink. Comparison of sequences from mink and mink farm workers to SARS-CoV-2 genomes from patients across the Netherlands (N = 1,775), from cases nearby to the first four farms (N = 34) and sequences from Poland (N = 65) showed that none were related to the mink farm outbreaks. Additionally, the presence of viral groups across different farms could not be explained by common ownership, shared personnel, feed supplier, veterinary service provider, or geographic distance.

Study Strengths

The study benefited from the simultaneous or retrospective testing of both human workers and mink on each farm. Full genome sequencing increased the phylogenetic resolution for assessing relatedness between viruses infecting mink and humans. Including reference sequence from COVID-19 cases in the Netherlands and from elsewhere in Europe were important for understanding how outbreaks on farms began.


The authors did not provide details on the number of animals that became ill or died from SARS-CoV-2 on each farm, or the prevalence of infection by PCR testing. Thus, we cannot determine any epidemiological parameters for the virus in mink (e.g., R0, infection fatality rate). The limited number of viral genomes that were used for comparison with viruses on each farm may explain their inability to determine the broader origin of outbreaks on farms. Limited testing in the Netherlands also means that spillback into the wider community from infected mink farms cannot be ruled out, even though it was not observed in these data. Lastly, interviews identified that some temporary workers were not included in testing and were lost to follow-up, so it cannot be ruled out that these workers may have partly contributed to virus introduction on farms, along with the permanent employees.

Value added

This report provides additional information on the outbreaks of SARS-CoV-2 on mink farms in the Netherlands, particularly the first evidence of an outbreak of SARS-CoV-2 in non-human animals and of animal-to-human transmission of SARS-CoV-2.

This review was posted on: 9 October 2020