Yazenia Linares Vega
Acute respiratory infections (ARI) are the leading cause of morbidity and mortality worldwide, mainly in young children and in immuno compromised and elderly patients. Respiratory viruses are the main etiology of respiratory disease worldwide, it is estimated that 90% of ARIs are related to viruses. In Cuba, the highest% of positivity is provided by Influenza, RSV and of positivity is provided by Influenza, RSV and currently SARS CoV2, which is generating the largest pandemic. The significant increase in climate anomalies produced by anthropogenic climate change leads to changes in the multiplicative capacity of the virus and in its circulation, both on a temporal and spatial scale. Therefore, forecasting the viral dynamics would allow the decision makers of the health system to take the necessary measures to control it. Obtaining seasonal models for the prediction and early warning of the impact of climate variability on influenza, RSV and SARS CoV2 viruses both on a temporal and spatial scale in Cuba is an essential objective. An ecological study was carried out with retrospective-prospective analysis of the virus series and the climatic anomalies described by the climatic indices of Bultó for the period 2010-2020. For the spatial structure, the data series interpolated from a 10km weight matrix by the Kriging method was used, which allows creating a continuous mesh for the country. Autoregressive spatial and heteroscedastic time series models with hexogenous variables (Climate Index) were confirmed to be adequate for predicting the impact of climate variability on the circulation and spread viruses. The quality of fit of the temporal model with significant values of agreement and Skill factor was 0.95% - 0.96% and for the spatial model, 0.9-% - 0.91%. The increase in temperatures, rainfall and humidity create favorable conditions for the circulation of the viruses that cause ARI. A monthly early warning system was obtained to monitor the circulation of viruses in the country and the provinces with the highest viral activity were identified in different months of the year.
Conclusions: Both temporal and spatial models are obtained for the prediction and early warning of the circulation of the influenza viruses, RSV and the recent SARS CoV2, conditioned by the impact of seasonal climate variability with high reliability. This model methodology could be used in other respiratory viruses