Spatio-temporal epidemiology of animal and human rabies in northern South Africa between 1998 and 2017

Abstract

Background: Rabies is a fatal zoonotic disease that is maintained in domestic dogs and wildlife populations in the Republic of South Africa. A retrospective study was conducted to improve understanding of the dynamics of rabies in humans, domestic dogs, and wildlife species, in relation to the ecology for three northern provinces of South Africa (Limpopo, Mpumalanga, and North-West) between 1998 and 2017.

Methods: A descriptive epidemiology study was conducted for human and animal rabies. Dog rabies cases were analyzed using spatio-temporal scan statistics. The reproductive number (Rt) was estimated for the identified disease clusters. A phylogenetic tree was constructed based on the genome sequences of rabies viruses isolated from dogs, jackals, and an African civet, and Bayesian evolutionary analysis using a strict time clock model. Several ecological and socio-economic variables associated with dog rabies were modeled using univariate analyses with zero-inflated negative binomial regression and multivariable spatial analyses using the integrated nested Laplace approximation for two time periods: 1998-2002 and 2008-2012.

Results: Human rabies cases increased in 2006 following an increase in dog rabies cases; however, the human cases declined in the next year while dog rabies cases fluctuated. Ten disease clusters of dog rabies were identified, and utilizing the phylogenetic tree, the dynamics of animal rabies over 20 years was elucidated. In 2006, a virus strain that re-emerged in eastern Limpopo Province caused the large and persistent dog rabies outbreaks in Limpopo and Mpumalanga Provinces. Several clusters included a rabies virus variant maintained in jackals in Limpopo Province, and the other variant in dogs widely distributed. The widely distributed variant maintained in jackal populations in North-West Province caused an outbreak in dogs in 2014. The Rt was high when the disease clusters were associated with either multiple virus strains or multiple animal species. High-risk areas included Limpopo and Mpumalanga Provinces characterized by woodlands and high temperatures and precipitation.

Conclusion: Canine rabies was maintained mainly in dog populations but was also associated with jackal species. Rural communities in Limpopo and Mpumalanga Provinces were at high risk of canine rabies originating from dogs.

Fig 1. Map showing the provinces of South Africa and study areas.

< https://dataportal-mdb-sa.opendata.arcgis.com/>.

Fig 2. Spatial distributions of positively and negatively diagnosed animals between 1998 and 2017, and changes in land use over time.

< https://www.dffe.gov.za>.

Fig 3. Temporal dynamics of human, dog, and wildlife rabies cases in northern South Africa between 1998 and 2017.

Fig 4. Choropleth maps of the distributions between 1998 and 2017 of dog rabies cases indicated per square kilometer by local municipality as a color gradient, and locations of human rabies victims shown by different colors and symbols indicating the source animals.

< https://dataportal-mdb-sa.opendata.arcgis.com/>.

Fig 5. Spatio-temporal clusters of dog rabies in each time period and spatial distributions of wildlife rabies cases indicated on choropleth maps of dog rabies in four time periods between 1998 and 2017.

Red dots indicate local municipalities affected by spatio-temporal clusters. MC denotes a local municipality of Gauteng Province, and was excluded from this study. < https://dataportal-mdb-sa.opendata.arcgis.com/>.

Fig 6. Phylogenetic tree of rabies viruses in dogs and wildlife in 10 spatio-temporal clusters of dog rabies between 1998 and 2017.

The X-axis shows the estimated years of rabies virus evolution prior to 2017. The first block of the virus designation indicates the animal species: Dog, black-backed jackal (Bbj), or civet cat (Civet). The second block is the cluster identification, the third the virus code, followed by the year of sample submission.

Fig 7. Predicted ecological risk of dog rabies.

Panel a) the risk between 1998 and 2002 with the Kruger National Park (KNP); b) between 2008 and 2012 with the KNP; c) between 1998 and 2002 without the KNP; and d) between 2008 and 2012 without the KNP. Red municipalities had the highest rabies risk, whereas green areas had the lowest risk. < https://dataportal-mdb-sa.opendata.arcgis.com/>.