Spatial and spatiotemporal machine learning models for studying COVID-19 dynamics: A review of methodology and reporting practices (Revised version accepted for publication in Epidemiologic Reviews)

COVID-19 has transitioned from a pandemic to an endemic state, but the emergence of novel variants continues to pose significant public health challenges. This study aimed to systematically review the application of spatial and spatiotemporal machine learning (ML) models in understanding the dynamics of COVID-19 and the local-level drivers, including demographic, socioeconomic, environmental, epidemiological, healthcare, housing conditions, behavioural, and vaccination. A systematic search was conducted across Scopus, Web of Science, PubMed, Emcare (via Ovid), the WHO COVID-19 database, and grey literature, adhering to PRISMA guidelines. Data extraction followed the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS) checklist, and study quality was assessed using a validated scoring system. A total of 42 studies met the inclusion criteria. Our findings indicate that global-scale spatial and spatiotemporal ML models dominate the field. Standalone factors such as demographic, environmental, and socioeconomic variables are frequently used as local-level drivers. However, integrating these into composite indicatorsaggregating multiple standalone factors into a single score-is notably lacking. This review highlights critical gaps in the current use of spatial and spatiotemporal ML models to understand the spatial epidemiology of COVID-19. Addressing these gaps could significantly enhance the understanding of COVID-19 dynamics and inform the development of effective public health strategies to mitigate future threats.