ORIGINAL RESEARCH

ABSTRACT

This study quantified the impact of long‑term PM₂.₅ exposure on natural all-cause mortality among adults aged ≥30 years in Tbilisi, using the WHO AirQ+ framework and multi‑year mortality and air quality data for 2019−2024. At a six‑year mean PM₂.₅ concentration of 17.3 µg/m³, an estimated 9.0% of natural all-cause deaths in this age group were attributable to PM₂.₅, corresponding to about 1,081 premature deaths per year and 154.6 deaths per 100,000 population aged ≥30 years. These findings indicate a substantial mortality burden associated with ambient PM₂.₅ exposure in Tbilisi.

​

Several recent studies using the AirQ+ tool provide useful comparisons for interpreting these findings. A national AirQ+ assessment in Serbia estimated approximately 3,585 premature deaths annually attributable to long-term PM₂.₅ exposure across 11 major cities, including around 1,796 deaths in Belgrade. 17 Similarly, a study conducted in Mashhad during 2016-2021 reported annual PM₂.₅ concentrations ranging from 24 to 33 µg/m³ and estimated approximately 1,800-2,400 attributable deaths annually, with attributable fractions between 10% and 15% depending on the year and baseline incidence. 18 Although differences in population size, exposure levels, and methodological assumptions limit direct comparison, the estimated burden observed in Tbilisi is consistent with findings from other urban settings with elevated PM₂.₅ exposure.

​

A major strength of this study is the use of multi-year exposure and mortality averages, which reduces the influence of short-term fluctuations and unusual events, including the marked increase in mortality observed during the COVID-19 pandemic in 2021. The use of the standardized WHO AirQ+ framework also improves comparability with international studies and provides a transparent approach for health impact assessment.

​

Several limitations should also be considered. First, population estimates for adults aged ≥30 years were derived using age- and sex-specific proportions from the 2014 census, which may not fully reflect demographic changes during the 2019−2024 study period, including population aging, migration, and urban population growth, As these estimates were used to derive the population denominator for the health impact assessment, any discrepancies may have influenced the estimated mortality burden. Second, exposure assessment was based on city-wide average PM₂.₅ concentrations and therefore did not capture intra-urban variability in exposure. Individuals residing near major roads or industrial sources may experience substantially higher pollution levels than are reflected in the citywide average. Future studies incorporating higher-resolution exposure assessment methods could better characterize spatial variability in PM₂.₅ exposure and support more geographically refined health impact assessments. Third, the concentration-response functions used in AirQ+ are derived primarily from cohort studies conducted outside Georgia and may not fully represent local population characteristics or exposure patterns. In addition, the analysis did not account for indoor air pollution, socioeconomic factors, behavioral risk factors, or the combined effects of multiple pollutants. Furthermore, the 95% confidence intervals reported by AirQ+ reflect uncertainty in the concentration-response function (relative risk estimates) and do not account for additional sources of uncertainty, including exposure assessment, population estimation, mortality data, or the spatial representativeness of the monitoring network. Consequently, the estimated mortality burden should be interpreted as an approximation of the health impacts attributable to PM₂.₅ exposure. Future studies could extend this work by examining cause-specific mortality outcomes to provide more clinically relevant evidence for public health policy, and by estimating population-weighted PM₂.₅ exposure at the district level to better account for spatial variation in both pollution concentrations and population distribution across Tbilisi.

​

The hypothetical reduction scenarios demonstrated substantial potential health benefits associated with improving air quality in Tbilisi. Reducing the long-term mean PM₂.₅ concentration from 17.3 µg/m³ to 12.3 µg/m³ and 7.3 µg/m³ was associated with marked declines in attributable mortality. In contrast, attainment of the WHO annual guideline value of 5 µg/m³ was associated with eliminating modeled PM₂.₅-attributable mortality above the counterfactual concentration. These findings suggest that even moderate reductions in PM₂.₅ concentrations could prevent a considerable number of premature deaths.

​

Overall, this study provides one of the first standardized estimates of PM₂.₅-attributable mortality in Tbilisi based on multi-year data. The findings support integrating air-quality improvement into urban health and environmental policy and highlight the importance of interventions targeting major emission sources, particularly traffic emissions, construction-related dust, and residential combustion sources.