Statistical Appraisal of the Global Air Pollution Dataset: A Review and Empirical Re-analysis of City-Level Air Quality Indices

Ambient air pollution is a leading environmental risk factor for premature mortality, and city-level air qualityindices (AQI) are widely used to communicate exposure. The Global Air Pollution Dataset records the overallAQI and four pollutant sub-indices (PM2.5, ozone, nitrogen dioxide and carbon monoxide) for 23,463 citiesin 175 countries, yet its statistical structure and the validity of analyses built on it are rarely examinedtogether. This review combines a narrative synthesis of the literature with a full statistical re-analysis ofthe supplied file. We audit data quality and internal consistency, describe distributions, quantify pollutantdominance, correlation and relative importance, measure between-country heterogeneity, and evaluatepredictive models under two validation designs. Our synthesis finds that (i) the overall AQI equals themaximum sub-index in 99.4% of records, and the 141 exceptions all occur at extreme PM2.5 levels; (ii) PM2.5is the highest sub-index in 78.9% of cities and in every city rated Hazardous, and alone explains 96.9% ofAQI variance; (iii) country of location explains about half of the variation in log AQI (intraclass correlation0.54; Kruskal-Wallis ϵ2 = 0.49), with median AQI ranging from 20 (Bolivia) to 169 (Pakistan); and (iv) whencountry information is allowed to leak across cross-validation folds, the AUC for predicting unhealthy airfrom non-PM2.5 pollutants rises from 0.84 to 0.97, an artefact of the validation design. We propose a shortreporting checklist and six priority research directions.