Predictive Models and Factor Analysis for Dengue Fever Healthcare Management in South America

This study investigates the influence of 29 socioeconomic factors on dengue incidence across nine South American countries from 1990 to 2020. Dengue, a mosquito-borne disease, continues to pose significant public health challenges, exacerbated by socioeconomic disparities and environmental changes. However, high multicollinearity among the variables complicates traditional regression analyses. To address this, Partial Least Squares (PLS) regression was employed, a method adept at handling correlated predictors while ensuring accurate estimation and prediction. The study’s methodology included determining the optimal number of PLS components through the minimization of the Root Mean Squared Relative Error (RMSRE), ensuring robust model performance. Results indicate that PLS regression successfully mitigates multicollinearity and identifies key socioeconomic determinants influencing dengue cases. These factors provide valuable insights into the socioeconomic drivers of dengue transmission, offering a more nuanced understanding of disease dynamics in diverse settings. The study also underscores the potential of advanced statistical methods to enhance disease prediction and inform evidence-based decision-making in public health. This research contributes to the field of public health by demonstrating the applicability of PLS regression in complex epidemiological datasets. Furthermore, it highlights the importance of integrating socioeconomic dimensions into dengue control strategies, particularly in resource-limited regions. These findings underscore the need for interdisciplinary approaches in addressing vector-borne diseases and tailoring interventions to specific socioeconomic contexts.