Epidemiology researcher uses Google Trend search data to predict RSV pediatric hospitalizations

You do it all the time. Wake up with cold-like symptoms and start Googling. Google logs all search activity, allowing researchers to dig deep into data, glean insights into population behavior, and identify health-related trends.

Tarang Parekh, assistant professor of epidemiology at the University of Delaware College of Health Sciences, has done precisely that. Using historic modeling of Google Trends search activity from 2019 and a cross-sectional analysis of the 2019 Healthcare Cost and Utilization Project’s Kids’ Inpatient Database (HCUP-KID), Parekh and colleagues have identified a link between respiratory syncytial virus (RSV) searches and pediatric hospitalizations linked to RSV. 

The study, recently published in the American Academy of Pediatrics’ Hospital Pediatrics journal, is the first of its kind, showing lead-time predictions that could have significant ramifications on overburdened healthcare systems. 

“If people’s interest in RSV increases, and they’re searching for information on RSV, should we expect increased hospital admissions within a month or two? That’s the question we were asking,” Parekh said. 

The research was inspired by Parekh’s wife, Sahiti Pemmasani, a co-author of the paper and a pediatric emergency fellow at Beaumont Hospital in Michigan. She observed an influx of RSV cases during her residency at Mt. Sinai Hospital in New York in 2020. 

“She saw a significant increase in RSV cases, and ICUs were full of COVID patients,” he said. “So, in talking with pediatricians in West Virginia and California, we realized they were seeing similar increases in hospitalizations, and it made me wonder about potential patterns.

“Hospitals were not ready for the RSV spike,” Parekh said. “But we thought if we can predict an increase in RSV hospitalizations based on search data, hospitals could prepare and ensure they’re better equipped to handle an influx of pediatric hospitalizations.”