Short-term associations between air pollution and nocturnal cough frequency.

Abstract

Background: Air pollution remains a leading environmental risk factor for health, yet its short-term associations with early respiratory symptoms are not well characterized because population-scale symptom surveillance data are scarce. We aimed to quantify the association between short-term exposure to fine particulate matter (PM_2.5) and nocturnal cough frequency using large-scale mobile sensing data collected across multiple countries.

Methods: We analyzed nocturnal audio recordings from users of a widely used mobile sleep application (Sleep Cycle). An artificial intelligence-based cough detection model was applied to estimate nightly cough frequency. Data were aggregated at the city-day level across 32 cities in 12 countries over approximately 500 consecutive days. We used city-specific time-series models combined through random-effects meta-analysis and multi-city pooled models adjusting for meteorological conditions and influenza activity. In addition, we conducted an event-based analysis during a major wildfire episode to evaluate the acute respiratory impacts of extreme air pollution.

Results: Higher daily PM_2.5 concentrations were consistently associated with increased nocturnal cough frequency across all analytical approaches. In the city-specific meta-analysis, each 10 μg/m³ increase in PM_2.5 was associated with a 2.4% increase in nocturnal cough frequency (relative risk 1.024, 95% CI: 1.013-1.035). Pooled analyses showed a nonlinear exposure-response relationship, with measurable increases in cough frequency observed even at relatively low PM_2.5 concentrations. Elevated cough frequency was also observed during the wildfire episode, coinciding with substantial increases in ambient PM_2.5.

Conclusions: Short-term exposure to PM_2.5 is associated with increased nocturnal cough frequency at the population level, suggesting that respiratory symptoms may respond to air pollution at lower exposure levels than previously recognized. Large-scale mobile sensing of respiratory symptoms provides a feasible digital biomarker for real-time environmental health surveillance and could complement traditional public health monitoring systems.