The role of modeling in the epidemiology and control of lumpy skin disease: a systematic review

Abstract

Abstract Background Lumpy skin disease (LSD) is an economically important viral disease of cattle caused by lumpy disease virus (LSDV) and transmitted by blood-feeding insects, such as certain species of flies and mosquitoes, or ticks. Direct transmission can occur but at low rate and efficiency. Vaccination has been used as the major disease control method in cooperation with other methods, yet outbreaks recur and the disease still persists and is subsequently spreading into new territories. LSD has of late been spreading at an alarming rate to many countries in the world including Africa where it originated, Middle East, Asia and some member countries of the European Union except the Western Hemisphere, New Zealand and Australia. In order to take control of the disease, various research endeavors are going on different fronts including epidemiology, virology, social economics and modeling, just to mention a few. This systematic review aims at exploring models that have been formulated and/or adopted to study the disease, estimate the advancement in knowledge accrued from these studies and highlight more areas that can be further advanced using this important tool. Main body of the abstract Electronic databases of PubMed, Scopus and EMBASE were searched for published records on modeling of LSD in a period of ten (10) years from 2013 to 2022 written in English language only. Extracted information was the title, objectives of the study, type of formulated or adopted models and study findings. A total of 31 publications met the inclusion criteria in the systematic review. Most studies were conducted in Europe reflecting the concern for LSD outbreaks in Eastern Europe and also availability of research funding. Majority of modeling publications were focused on LSD transmission behavior, and the kernel-based modeling was more popular. The role of modeling was organized into four categories, namely risk factors, transmission behaviors, diagnosis and forecasting, and intervention strategies. The results on modeling outbreaks data identified various factors including breed type, weather, vegetation, topography, animal density, herd size, proximity to infected farms or countries and importation of animals and animal products. Using these modeling techniques, it should be possible to come up with LSD risk maps in many regions or countries particularly in Africa to advise cattle herders to avoid high risk areas. Indirect transmission by insect vectors was the major transmission route with Stomoxys calcitrans being more effective, indicating need to include insect control mechanisms in reducing the spread of LSD. However, as the disease spread further into cold climates of Russia, data show new emerging trends; in that transmission was still occurring at temperatures that preclude insect activities, probably by direct contact, and furthermore, some outbreaks were not caused by field viruses, instead, by vaccine-like viruses due to recombination of vaccine strains with field viruses. Machine learning methods have become a useful tool for diagnosing LSD, especially in resource limited countries such as in Africa. Modeling has also forecasted LSD outbreaks and trends in the foreseeable future indicating more outbreaks in Africa and stability in Europe and Asia. This brings African countries into attention to develop long-term plans to deal with LSD. Intervention methods represented by culling and vaccination are showing promising results in limiting the spread of LSD. However, culling was more successful when close to 100% of infected animals are removed. But this is complicated, firstly because the cost of its implementation is massive and secondly it needed application of diagnostic techniques in order to be able to rapidly identify the infected and/or asymptomatic animals. Vaccination was more successful when an effective vaccine, such as the homologous LSD vaccine, was used and complemented by a high coverage of above 90%. This is hard to achieve in resource-poor countries due to the high costs involved. Short conclusion Modeling has made a significant contribution in addressing challenges associated with the epidemiology and control of LSD, especially in the areas of risk factors, disease transmission, diagnosis and forecasting as well as intervention strategies. However, more studies are needed in all these areas to address the existing gaps in knowledge.