UKHSA Study Finds Sleep App Cough Data Can Predict Respiratory Illness Outbreaks
The UK Health Security Agency (UKHSA) has collaborated with the AI sleep technology company Sleep Cycle to explore whether cough data collected from a sleep app can provide early warnings of respiratory illnesses in England. The findings from their study suggest that this passive data collection can serve as a reliable and regionally consistent indicator of respiratory illness in the community, particularly for influenza and COVID-19 activity. The research indicates that increases in cough data often precede reported cases of these illnesses by about a week, demonstrating the potential of digital health data in public health monitoring.
Traditionally, public health surveillance relies on individuals seeking medical care through the NHS, which can be influenced by various factors such as public awareness, availability of services, and demographic differences. Moreover, existing systems are often affected by delays in reporting and laboratory processing times. In contrast, the Sleep Cycle app collects cough data automatically during sleep, using privacy-preserving methods that allow for daily updates, thus providing a near real-time overview of respiratory illness trends.
The study highlights that the cough data aligns closely with levels of respiratory illness reported through NHS 111, with noticeable increases in coughing often occurring one week prior to spikes in flu and COVID-19 cases. This suggests that integrating passive digital health monitoring with traditional surveillance systems could enhance the understanding of seasonal respiratory illness trends.
Professor Steven Riley, Chief Data Officer at UKHSA, remarked that these findings indicate the value of combining established surveillance methods with innovative digital health signals to achieve a more comprehensive understanding of population respiratory health. He noted that no single surveillance system can fully capture the dynamics of respiratory diseases, but the use of passive cough monitoring could provide timely signals about upcoming disease trends, independent of healthcare-seeking behaviors and reporting delays.
Dr. Emil Carlsson, a research scientist and co-lead author of the study, emphasized that the data collected nightly through the app reflects significant changes in community respiratory health. He pointed out that this study demonstrates how consumer-generated health data can be transformed into valuable epidemiological signals while ensuring strong privacy protections.
Dr. Mikael Kågebäck, Chief Technology Officer and Acting Chief Executive Officer at Sleep Cycle, stated that this research validates a new category of health data, showcasing that passively generated smartphone data can yield robust population-level health insights at a national scale. This capability opens up opportunities for enhancing public health surveillance and enabling the development of new services aimed at improving situational awareness and operational decision-making in healthcare.




