Back

Global generalisability of AI-driven COVID-19 vaccination policies: a cross-sectional observational study

Awasthi, R.; Nagori, A.; Nasri, B.

2023-02-02 health informatics
10.1101/2023.01.31.23285300 medRxiv
Show abstract

The effectiveness of a vaccine depends on vaccine uptake, which is influenced by various factors, including vaccine hesitancy. Vaccine hesitancy is a complex socio-behavioral issue, influenced by misinformation, distrust in healthcare providers and government organizations, fear of side effects, and cultural or religious beliefs. To address this problem, AI models have been developed, but their global generalizability remains unclear. Therefore, this study aimed to identify global determinants of vaccine uptake and develop a generalizable machine learning model to predict individual-level vaccine uptake. The study used publicly available survey data from 23 countries and employed Bayesian networks and generalized mixed effects models to identify key determinants of vaccine uptake. The results showed that trust in the central government and vaccination restrictions for national and international travel were key determinants of vaccine uptake. A generalized mixed effects model achieved an AUC of 89% (SD=1%), precision of 90% (SD = 4%), and recall of 82% (SD=2%) on unseen testing data from new countries, demonstrating the models generalizability. The findings of this study can inform targeted interventions to improve vaccine uptake globally.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.