RAHES: Remote Assessment of Health Exposures and Environment in SG100K Study Using a Low-Cost Indoor Air Quality Monitoring System

Air pollution is a significant driver of adverse health outcomes, contributing to an estimated 6–7 million premature deaths in 2019. According to the World Health Organization (WHO), nearly half of these fatalities are attributed by indoor air pollution, including particulate matter (PM), a critical concern as populations spend the majority of their time indoors. Exposure to PM_2.5 is positively associated with cardiovascular diseases, lung cancer, neurodegenerative conditions, and elevated oxidative stress in cells. Despite growing recognition of its health relevance, characterizing indoor air quality (IAQ) remains challenging due to building designs, ventilation systems, and human activity. Large-scale residential IAQ monitoring is limited, as indoor data are difficult to collect at scale. While low-cost sensors (LCS) offer a promising approach for improving spatial and temporal coverage, they face challenges related to inter-model variability and limited accuracy for gaseous pollutants such as NO, NO₂ and O₃.

Singapore is a complex urban city located at the southern tip of the Malay Peninsula. It is characterized by high population density, a major petrochemical complex, and one of the world’s busiest shipping ports and airports. Several previous studies have reported seasonal and spatial variability of black carbon (BC), brown carbon (BrC) and organic aerosols (OA). Our previous work using drone-based light absorption measurements reported elevated outdoor BC and BrC, near the 10th and 20th floors, respectively, compared to ground level. Notably, 77% of Singaporeans live in Housing and Development Board units. Of these buildings, 82% exceed 10 floors and 11% exceed 20 floors. As a result, vertical gradients in air pollution may have important implications for indoor exposure.

In this study, we explore feasibility of collecting IAQ data in the SG100K cohort. The cohort includes 100,000 participants with detailed clinical assessments and linkage to health records, allowing for a direct link between environmental exposure and health outcomes to be examined. This feasibility study consists of 200 participants, informing protocols for scaling up measurements in the SG100K cohort over 5 years. Each participant undergoes three months of continuous indoor measurements in their living room and bedroom. Our low-cost sensor system collects PM2.5, CO2, temperature, and humidity data remote. Data are supported by a proprietary remote data acquisition and quality-control pipeline. A subgroup of participants will also measure particle number concentration to explore the impacts of cooking and traffic emissions.

Preliminary indoor measurements indicate significant vertical variation in PM2.5 concentrations within the same building. Higher concentrations were consistently observed on the 11^th floor (12.1±3.9 µg/m³) compared with the 5^th and 7^th floors (6.3±2.5 and 8.0±3.3 µg/m³, respectively). In addition, indoor PM concentrations do not always follow outdoor ground-level diurnal patterns, suggesting that indoor PM is influenced by indoor human activity, indoor–outdoor air exchange, filtration systems, transport, and secondary processes. Initial cross-sensor comparisons demonstrate consistent performance among individual devices. Further analysis will include a wider range of room types, locations, and floor heights. Data collected from RAHES will be synthesized with longitudinal health records to elucidate how the residential environment influences human behaviours and health outcomes.