Discussion on a Selection of Key Scientific Articles

By John Atkinson,
Associate Director, Intergovernmental Veterinary Health
Follow us on Twitter

Assisting differential clinical diagnosis of cattle diseases using smartphone-based technology in low resource settings: a pilot study. Tariku Jibat Beyene, Amanuel Eshetu, Amina Abdu, Etenesh Wondimu, Ashenafi Feyisa Beyi, Takele Beyene Tufa, Sami Ibrahim, Crawford W. Revie. (2017) BMC Veterinary Research, 13:323. doi 10.1186/s12917-017-1249-3

Key Thoughts:
In an ideal world every diagnosis would be confirmed by means of a laboratory investigation. However, limited resources and laboratory availability means that this is often not possible and so low-cost tools that increase the likelihood of a correct diagnosis can play a significant role in developing countries. As mobile phone usage across these regions increases, it makes sense that the veterinary world should look to digital solutions, like the app featured in this paper, that are easy to use and reliable. The disease profile found in this study was consistent with that published by the Ministry of Agriculture, although it was also interesting to note that lumpy skin disease and foot-and-mouth disease (FMD) were more frequently diagnosed than apparently expected.

Article Summary:
Livestock disease is one of the most important factors affecting livestock production and productivity, and it damages the livelihoods of the majority of families in Ethiopia. This is exacerbated by livestock owners not being able to access reliable veterinary advice from experienced animal health professionals. This lack of veterinary input means that cattle diseases may be misdiagnosed. Tools that support veterinary decision-making may help improve the diagnosis of disease, especially when diseases present with similar clinical signs.

In this paper the authors describe a pilot study involving VetAfrica-Ethiopia, which is a smartphone-based application that aids in the differential diagnosis of cattle diseases. Fifteen final-year veterinary medicine students were allocated to veterinary clinics across Ethiopia, provided with low-cost smartphones, and given basic training on how to use the smartphone app. The app included 12 diseases which were identified by experienced veterinarians in Ethiopia as being particularly important economically and/or challenging to diagnose on clinical signs.

Under the supervision of the duty veterinarian, the veterinary students examined sick animals brought to their respective clinics and recorded relevant case details in the app. The students made a diagnosis based on the clinical signs and case history, whilst the app gave a range of tentative diagnoses with associated likelihoods. The level of agreement between the student diagnoses and the app predictions were evaluated. Laboratory confirmation was not possible because of the distance from laboratories and the level of resources required.

From 928 cattle examined, over 70 different diseases were diagnosed and around 70% of the cases were associated with a disease that had been included in the app. Parasitic gastroenteritis was the most common diagnosis, whilst lumpy skin disease, lungworm, and FMD were in the top 11 diseases but had not been included in the app. The highest level of matching between students and the app was for babesiosis and the lowest was for pasteurellosis, which may reflect how well characterised/how similar to other diseases these are. Multiple-variable logistic regression analysis indicated that the likelihood of a match was influenced by factors including the disease being diagnosed, the veterinary student, and the level of certainly assigned to the most likely diagnosis in the app.

This pilot study found an acceptable level of matching between the student diagnosis and the app prediction, which suggests that smartphone technology can help animal health professionals in resource-limited settings.

Cost-effectiveness of mass dog rabies vaccination strategies to reduce human health burden in Flores Island, Indonesia. Ewaldus Wera, Monique C.M. Mourits, Henk Hogeveen. (2017) Vaccine. 35:6727-6736.

Key Thoughts:
It is always important to evaluate why we do things, and determine if the current strategy we are using is the most appropriate one for whatever we are trying to achieve. Resource limitations can significantly impact the strategy chosen, but sometimes an increased investment can really pay off. We know that mass vaccination of dogs works for stopping rabies in both dogs and people, so it is essential that we employ the right vaccine at the right frequency and coverage for the population concerned. The study here shows that all human deaths due to rabies can be prevented. Surely that’s the goal we must strive to reach.

Article Summary:
Rabies is a serious and costly threat to the inhabitants of Flores Island, with the vast majority of registered bite cases receiving post-exposure treatment (PET) free of charge from the local government. However, mass vaccination of dogs is needed to eliminate the disease in the dog population and prevent rabies in humans.

Annual mass vaccination using a long-acting vaccine at a coverage of 70% has been shown as the most cost-effective strategy in reducing dog rabies cases, but did not explore the impact on saved human life and prevented PET. So the authors used a dynamic simulation model to evaluate the cost-effectiveness (defined as the costs per year of life lost [YLL] averted) of mass dog vaccination strategies on Flores Island.

The previously developed SEIVR model was used to simulate the transmission of rabies over 10 years for a typical village of 1,500 people and 400 dogs, starting with the introduction of one infectious dog. Actual vaccination expenditures in Flores Island were used to calculate the cost of vaccination strategies. Various strategies were evaluated because they had previously been found to effectively control a dog rabies epidemic, including short- (52-week duration of immunity [DOI]) and long-acting (156-week DOI) vaccines at varying frequency and coverage. The current strategy used in Flores Island of short-acting vaccine with 50% coverage was also included, as was a PET-only strategy.

The base strategy of no mass dog vaccination and no PET resulted in 881 YLLs, or 30 cases of human rabies during the 10-year simulation period. Dog vaccination was generally more effective than PET alone. The current strategy of annual vaccination with short-acting vaccine at a coverage of 50% was not cost-effective. The use of long-acting vaccine at 70% coverage appeared to be most cost-effective in reducing YLL. Combining this vaccination strategy with PET is slightly more expensive, but would prevent all human deaths due to rabies. Increasing the coverage of dog vaccination and/or switching to long-acting vaccine is needed to reduce human rabies and would be cost-effective.


Copyright © 2019 Intervet International B.V., also known as MSD Animal Health. All rights reserved.