To the Editor:
Why do different air pollution studies find conflicting results for the same pollutants? Holliday et al1 illustrate the importance of accurate exposure, showing that more accurate exposure for PM10 produced stronger estimates of health associations (although with larger standard deviations). Accurate exposure measurement is even more important for PM2.5 species—notably black carbon, which has higher local spatial variability than PM10 or PM2.5.
Using personally monitored measures of elemental carbon, PM2.5, and sulfate concentrations in Atlanta, Suh and Zanobetti2 found that elemental carbon (but not other pollutants) was strongly associated with 5 measures of heart rate variability. When centrally monitored data were substituted for personally monitored concentrations, the elemental carbon associations became considerably smaller and statistically consistent with a null effect.
Hsu et al3 found no associations of PM2.5 nickel with pulse rate in New York City when using centrally monitored measurements of airborne PM2.5 nickel as a proxy for personal exposure. However, when exposure was measured inside or just outside the home, or measured personally, associations were found.
Grahame4 examined whether the association of black carbon with heart rate variability was lower in studies using centrally monitored black carbon across wide geographic areas than in studies using more accurate proxy measures. The associations were most robust when subject exposure was reasonably accurate and weakened with worsening exposure error—even suggesting the possibility that, when exposure was most poorly characterized, associations might shift from black carbon to pollutants with less exposure error (a finding needing further confirmation). Subsequently, Suh and Zanobetti2 confirmed the reduction of heart rate variability associations when central monitor readings are a proxy for reasonably accurate exposure. These studies strongly suggest the importance of accurate assessment of exposure when studying the health effects of PM2.3 compounds and black carbon, in particular.
Thomas J. Grahame
US Department of Energy
1. Holliday KM, Avery CL, Poole C, et al. Estimating personal exposures from ambient air pollution measures: using meta-analysis to assess measurement error. Epidemiology. 2014;25:35–43
2. Suh HH, Zanobetti A. Exposure error masks the relationship between traffic-related air pollution and heart rate variability. J Occup Environ Med. 2010;52:685–692
3. Hsu S-I, Ito K, Lippmann M. Effects of thoracic and fine PM and their components on heart rate and pulmonary function in COPD patients. J Expo Sci Environ Epidemiol. 2011;21:464–472
4. Grahame TJ. Does improved exposure information for PM2.5 constituents explain differing results among epidemiological studies? Inhal Toxicol. 2009;21:381–393