Assessing Agreement in Exposure Classification between Proximity-Based Metrics and Air Monitoring Data in Epidemiology Studies of Unconventional Resource Development

Open-access methodologic study examining the agreement between proximity-based exposure metrics (well activity, or WA) and ambient air monitoring data in epidemiologic studies of unconventional resource development. The study compared exposure quartiles derived from WA calculations against measured concentrations of six pollutants at Pennsylvania monitoring sites from 2011 through 2015. For every pollutant tested, agreement between the two methods was poor, with weighted kappa coefficients falling between -0.1 and 0.1. The result challenges the validity of WA proximity surrogates as a basis for inferring URD-related exposure in epidemiology.

Read the full open-access paper.