Forecasting the incidence of leishmaniasis in Ethiopia using global burden of disease 2021 data

visceral leishmaniasis is fatal vector-borne disease without timely treatment. East Africa accounts for nearly 45% of global visceral leishmaniasis cases, with Ethiopia among the most affected countries. Despite World Health Organization targets to reduce case fatality to below 1% by 2030, Ethiopia continues to face high mortality and poor treatment outcomes, worsened by limited resources. This analysis forecasts VL trends and provides to guide combined interventions supporting Ethiopia’s progress toward the Sustainable Development Goals. This study employed secondary data analysis from the Global Burden of Disease database (1990–2021) to forecast visceral leishmaniasis incidence in Ethiopia. LSTM-based models, including multistep long short-term memory, hybrid autoregressive integrate moving average and long short-term memory and Transformer approaches, were developed in TesnsorFlow, while autoregressive integrate moving average models were constructed using the statsmodels and pmdarima libraries in Python. Data stationery was tested at a 0.05 significance level. Model performance was assessed using Root Mean Squared Error, Mean Absolute Percentage Error, and Symmetric Mean Absolute Percentage Error. The best-performing model then applied to forecast VL incidence for the period 2021–2030. According to Global Burden of Diseases data, the incidence of Leishmaniasis in Ethiopia shows a long term downward trend, decreasing from 768.091 cases per 100,000 in 1990 to 86.820 by 2021. The analysis results revealed that Transformer model outperformed all, achieving mean absolute error: 9.55%, root mean square error: 10.075%, mean absolute percentage error: 12.78%, and symmetric mean absolute percentage error: 13.04%. The incidence of Leishmaniasis in Ethiopia is projected to decline slightly through 2030, according to Transformer model. The forecast estimates that the Leishmaniasis incidence will be decreasing 84.197783 cases per 100,000 by 2030. Overall, the transformer model demonstrated superior predictive performance. The forecasts indicate a continued decline in incidence; however, at the current trajectory, Ethiopia is projected to fall short of the World Health Organization neglected tropical diseases roadmap’s ambitious target of a 60% reduction by 2030. Enhancing integrated case management, scaling up targeted mass screening in high-burden regions like Amhara and Tigray, and improving access to diagnosis and treatment are essential.