Aim: Toxoplasma gondii is a ubiquitous zoonotic parasite that can infect warm-blooded vertebrates, including humans. Felids, the definitive hosts, drive T. gondii infections by shedding the environmentally resistant stage of the parasite (oocysts) in their feces. Risk factors for oocyst shedding are well-documented in domestic cats under controlled settings, but few studies characterize the role of climate and anthropogenic factors in oocyst shedding among free-ranging felids, which are responsible for the majority of environmental contamination. We aimed to determine how climate and anthropogenic factors influence the level of oocyst shedding in free-ranging domestic cats and wild felids. 

Location: Global

Time period: 1973-2021

Major taxa studied: Felidae, Toxoplasma gondii

Methods: We used generalized linear mixed models to determine the association between climatic, ecological, and anthropogenic factors and T. gondii oocyst shedding in free-ranging felids. T. gondii oocyst shedding data from 47 studies were compiled for domestic cats and 6 wild felid species, encompassing 256 positive samples out of 9,635 total fecal samples. 

Results: T. gondii shedding prevalence in domestic cats and wild felids was positively associated with human population density at the sampling location. Average annual temperature and total precipitation were not associated with increased shedding, however, temperature variables that reflected fluctuation on a smaller timescale were associated with oocyst shedding. Larger mean diurnal range was associated with higher T. gondii oocyst shedding prevalence in domestic cats, while higher temperatures in the driest quarter were associated with lower oocyst shedding in wild felids.

Main conclusions: Anthropogenic factors associated with increasing human population density and climate change in the form of temperature fluctuation can exacerbate environmental contamination with the protozoan parasite T. gondii. Direct or indirect management of free-ranging domestic cats could lower the burden of environmental oocysts due to their large population sizes and close affinity with human settlements. 

Dataset was collected using systematic review. For studies that met inclusion criteria, we extracted metadata including the year of publication, type of felid sampled (domestic/wild), country of study, continent, diagnostic method, number of positive fecal samples, total fecal samples tested, and approximate latitude and longitude of the sample site as reported by the authors for each study. We repeated this process in studies with multiple species of sampled felids. 

Our primary variables of interest were annual mean temperature, annual precipitation, and human population density. Additionally, we considered other climate variables, such as maximum temperature, mean diurnal range and precipitation seasonality, as well as human activity variables such as habitat type and species richness for inclusion in subsequent models (Table 1). For each study, we used the R package ‘raster’ with a World Geodetic System 1984 projection and a 2.5 km buffer to obtain location-specific data. Climate data (temperature and precipitation variables) was extracted at a 5 km resolution from the WorldClim 2.0 (1970-2000) dataset, while human population density and human footprint data at a 5 km resolution was extracted from the NASA Center for International Earth Science Information Network (CIESIN). Human population density was paired to each study by the closest time period (2000, 2005, 2010, 2015 and 2020). Habitat type was extracted at a resolution of 1 km, and species richness at 110 m from the International Union for Conservation of Nature (IUCN) (Jung et al. 2020). 

WorldClim 2.0 https://www.worldclim.org/data/worldclim21.html

NASA CIESIN https://sedac.ciesin.columbia.edu/data/collection/gpw-v4

IUCN Jung et al. 2020 https://zenodo.org/record/3666246#.YroIni9XZpR

After data are imported into statistical software such as R or Python, it should be transformed and cleaned according to the attached R script. Note that all temperature variables from WorldClim are °C * 10 and need to be transformed prior to analysis (https://worldclim.org/data/v1.4/formats.html). Updated dataset includes all climate data extracted from WorldClim, NASA, and IUCN such that the user does not have to perform this task on their own.