Epidemiological Risk Factors and Modelling Approaches for Risk Assessment of Lumpy Skin Disease Virus Introduction and Spread: Methodological Review and Implications for Risk-Based Surveillance in Australia.

Lumpy skin disease (LSD) is a vector-borne infection caused by the poxvirus lumpy skin disease virus (LSDV) and is a serious disease of cattle, water buffalo, and banteng. While the disease has never occurred in Australia, it is regarded as a growing threat to the Australian cattle industry as there is on-going spread of the disease throughout Asia. The development of geospatial decision support tools, such as spatial epidemiological modelling, may assist in assessing areas at greater risk of this threat. To guide the design of disease modelling approaches to support future risk-based surveillance, existing LSDV epidemiological models need to be evaluated. In this study, we performed a literature review to evaluate existing LSDV epidemiological models, identify key risk factors for introduction and spread of LSDV, and consider previously adopted control strategies. The PRISMA guidelines were used to establish the processes for article selection and information extraction, and the PICO process was used to formulate search terms. From studies that met our inclusion criteria, we extracted information on LSDV epidemiological model structure and parameterisation, risk factors for LSDV transmission and spread, and biosecurity control strategies. The literature search retrieved a total of 402 articles from four databases, of which 68 were identified for inclusion in this review following screening. Of the 68 articles reviewed, 47 explored risk factors associated with LSDV transmission and spread, four explored risk factors of LSDV introduction, four explored existing surveillance strategies in LSD-free countries, and 14 presented epidemiological models. Our findings indicate that there are various risk factors for LSDV transmission in LSD endemic countries, including long-distance airborne movement of infected vectors such as stable flies and cattle movement between countries over land borders. Key risk factors for LSDV spread in LSD endemic countries include physical environmental characteristics, weather conditions, and population distributions of livestock and vectors. Our results indicate that while a variety of modelling studies have been conducted, the majority of studies experimentally explored LSD transmission mechanisms in vectors and cattle. Spatial and spatio-temporal models have primarily been developed for LSD endemic countries and focus on the spread of the disease in terms of environmental factors in relation to previous LSD events. There were very few studies on LSD-free countries, and these only focussed on risk of LSD introduction through specific entry pathways. This review did not identify any literature exploring the risk of spread of LSDV following introduction in LSD-free countries or geospatial modelling of the suitability of LSD-free countries for LSDV incursions. In conjunction with the risk parameters and models described in the identified literature, there is need to consider a wide range of risk factors specific to Australia to inform the design of risk-based surveillance for LSD in Australia.