Assessing COVID-19 Vaccine Effectiveness and Risk Factors for Severe Outcomes through Machine Learning Techniques: A Real-World Data Study in Andalusia, Spain.

Background: COVID-19 vaccination has become a pivotal global strategy in managing the pandemic. Despite COVID-19 no longer being classified as a Public Health Emergency of International Concern, the virus continues affecting people worldwide. This study aimed to evaluate risk factors and vaccine effectiveness on COVID-19-related hospital admissions, intensive care unit (ICU) admission and mortality within the Andalusian population throughout the pandemic.

Methods: From March 2020 to April 2022, 671,229 individuals, out of 9,283,485 with electronic health records in Andalusia, experienced SARS-CoV-2 infection and were included in the analysis. Data on demographics, medical history, vaccine administration, and hospitalization records were collected. Associations between medical history, COVID-19 vaccines, and COVID-19 outcomes were assessed.

Results: Our study identified 48,196 hospital admissions, 5,057 ICU admissions, and 11,289 deaths linked to COVID-19. Age, male sex, and chronic diseases were identified as risk factors, while the COVID-19 vaccine demonstrated protective effects, although with reduced effectiveness during the omicron variant period. However, the risk for these outcomes increased over time after receiving the last vaccine dose, particularly after six months, especially among those aged 60 or older.

Conclusion: The global health challenge of COVID-19 persists, marked by emerging variants with higher virulence and severity, particularly among the unvaccinated and those beyond six months post-vaccination, especially those aged 60 and above. These findings highlight the need for robust surveillance systems targeting new variants and administering booster doses, particularly for individuals aged 60 or older with underlying health conditions, to mitigate the global burden of COVID-19.