AI-driven analysis of passive nightly cough data collected through smartphone applications can provide a robust one-week early warning indicator for community flu and COVID-19 outbreaks, according to a research study published by the UK Health Security Agency (UKHSA) and sleep technology company Sleep Cycle.
The joint study evaluated whether cough data gathered during sleep could track respiratory illness trends across England. Researchers discovered that population-normalised cough metrics closely reflected levels of acute respiratory infection reported through NHS 111 triage calls, often signaling rising infection rates roughly seven days before official surges appeared in standard laboratory and healthcare data.
How Passive Sleep Data Tracks Respiratory Illness
The surveillance model relies on the Sleep Cycle smartphone app, which uses AI-powered sound analysis to help users understand their sleep patterns. During normal sleep, the software automatically records audio signals to detect coughing episodes using privacy-preserved, passively collected data. According to the research findings published by UKHSA and Sleep Cycle, these metrics update on a daily basis to offer a near real-time view of syndromic illness burden.
Traditional public health surveillance systems depend on individuals actively seeking medical care through the National Health Service. Consequently, those metrics face delays driven by healthcare-seeking behavior, service availability, demographic differences, laboratory processing workflows, and backfilling.
Complementing Traditional Public Health Surveillance
Public health officials emphasize that digital health signals are designed to enhance rather than replace established monitoring networks. The study found that increases in nightly coughing demonstrated short-term leading relationships of approximately one week for both influenza and COVID-19 surveillance indicators.
“No single surveillance system provides a complete picture of respiratory disease activity,” said Professor Steven Riley, Chief Data Officer at UKHSA. He noted that combining established approaches with novel digital health signals contributes to a richer, more resilient understanding of population respiratory health without being affected by reporting delays or laboratory turnaround times.
Dr. Emil Carlsson, Research Scientist and co-lead author of the study, stated that the findings demonstrate how consumer-generated health data can be transformed into epidemiologically meaningful surveillance signals using rigorous scientific methods while maintaining strong privacy protections.
Validating a New Category of Health Data
The collaboration marks the first time that passively generated smartphone data has been shown to produce robust population-level health intelligence at a national scale. Dr. Mikael Kågebäck, Chief Technology Officer and acting Chief Executive Officer at Sleep Cycle, explained that the validation creates opportunities for researchers, healthcare organizations, and industry partners to build new services for situational awareness and operational decision support.
While the study evaluated statistical relationships between surveillance indicators rather than real-time operational availability, researchers suggest the findings warrant prospective evaluation. Continuous digital health monitoring could soon provide public health authorities with the advanced warning required to manage seasonal respiratory illness trends more effectively.
Worth a look