Google search interests and new cases of COVID-19 in Bangladesh: a vector autoregression analysis for disease surveillance

This study explores the application of Google Search Trends (GST) data as a tool for COVID-19 surveillance in Bangladesh. Using the Vector Autoregression (VAR) approach, it investigates how GST search can predict new COVID-19 cases. The findings show that GST can offer timely insights into public health responses. The results also show that GST search items can predict the outbreak before 1–2 weeks. While VAR analysis confirms the causal relationship between the GST terms and the new COVID-19 cases, this study also acknowledges the limitations of inappropriate regional data at the outset of any new pandemic. It also acknowledges the misinformation from GST can underestimate real time surveillance. Policymakers can use GST for real-time risk communication and informed decision during pandemic. The results also suggest the importance of infodemiology to integrate it into the public health system for disease surveillance as the country’s’ resources are constrained. This study highlights the contribution of infodemiology towards the enhancement of traditional methods and systems of surveillance, which is relevant both nationally and globally. GST offers a great opportunity for advancing digital epidemiology and innovation in public health, nonetheless obstacles rest in the way.