Tuberculosis Spread Analysis through the Integration of Principal Component Analysis and Vector Autoregressive Models

Tuberculosis (TB) continues to be a pressing health issue in developing countries like Indonesia. In West Java, climatic factors such as temperature, humidity, and precipitation significantly impact TB incidence. This study integrates Principal Component Analysis (PCA) and Vector Autoregressive (VAR) models to investigate how these climatic variables affect TB spread. We analyzed weekly data on TB incidence, precipitation, relative humidity, and air temperature from five locations in West Java over 53 weeks. PCA was used to reduce the dimensionality of the climatic data, identifying key components like precipitation and temperature as primary factors influencing TB, with relative humidity having a secondary role. The VAR model examined the temporal relationships and revealed significant lagged effects of these climatic variables on TB incidence, indicating that climate changes influence TB spread with some delay. The coefficient of determination explains how far the dependent data can be explained by the independent data (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) is 0.6727. The integration of PCA and VAR models effectively captured the interaction between climate and TB dynamics, providing a robust framework for understanding these relationships. This approach highlighted the critical role of climate in TB forecasting and emphasized the need for targeted public health strategies. Future research should expand this model to other regions and incorporate socio-economic factors to further enhance TB forecasting and public health interventions.