Long-term associations of morbidity with air pollution: A catalog and synthesis
Frederick W Lipfert
DOI: https://doi.org/10.1080/10962247.2017.1349010
Abstract:I searched the National Institutes of Health MEDLINE database through January 2017 for long-term studies of morbidity and air pollution and cataloged them with respect to cardiovascular, respiratory, cancer, diabetes, hospitalization, neurological, and pregnancy-birth endpoints. The catalog is presented as an online appendix. Associations with PM2.5 (particulate matter with an aerodynamic diameter <2.5 μm), PM10 (PM with an aerodynamic diameter <10 μm), and nitrogen dioxide (NO2) were evaluated most frequently among the 417 ambient air quality studies identified. Associations with total suspended particles (TSP), carbon, ozone, sulfur, vehicular traffic, radon, and indoor air quality were also reported. I evaluated each study in terms of pollutant significance (yes, no), duration of exposure, and publication date. I found statistically significant pollutant relationships (P < 0.05) in 224 studies; 220 studies indicated adverse effects. Among 795 individual pollutant effect estimates, 396 are statistically significant. Pollutant associations with cardiovascular indicators, lung function, respiratory symptoms, and low birth weight are more likely to be significant than with disease incidence, heart attacks, diabetes, or neurological endpoints. Elemental carbon (EC), traffic, and PM2.5 are most likely to be significant for cardiovascular outcomes; TSP, EC, and ozone (O3) for respiratory outcomes; NO2 for neurological outcomes; and PM10 for birth/pregnancy outcomes. Durations of exposure range from 60 days to 35 yr, but I found no consistent relationships with the likelihood of statistical significance. Respiratory studies began ca. 1975; studies of diabetes, cardiovascular, and neurological effects increased after about 2005. I found 72 studies of occupational air pollution exposures; 40 reported statistically significant adverse health effects, especially for respiratory conditions. I conclude that the aggregate of these studies supports the existence of nonlethal physiological effects of various pollutants, more so for non-life-threatening endpoints and for noncriteria pollutants (TSP, EC, PM2.5 metals). However, most studies were cross-sectional analyses over limited time spans with no consideration of lag or disease latency. Further longitudinal studies are thus needed to investigate the progress of disease incidence in association with air pollution exposure. Implications: Relationships of air pollution with excess mortality are better known than with long-term antecedent morbidity. I cataloged 489 studies of cardiovascular, respiratory, cancer, and neurological effects, diabetes, and birth outcomes with respect to 12 air pollutants. About half of the studies reported statistically significant relationships, more frequently with noncriteria than with criteria pollutants. Indoor and cumulative exposures, coarse or ultrafine particles, and organic carbon were seldom considered. Significant relationships were more likely with less-severe endpoints such as blood pressure, lung function, or respiratory symptoms than with incidence of cancer, chronic obstructive pulmonary disease (COPD), heart failure, or diabetes. Most long-term studies are based on spatial relationships; longitudinal studies are needed to link the progression of pollution-related morbidity to mortality, especially for the cardiovascular system.