Do group responses mask the effects of air pollutants on potentially sensitive individuals in controlled human exposure studies?

To establish primary National Ambient Air Quality Standards (NAAQS) for criteria air pollutants such as nitrogen dioxide (NO2), ozone (O3), and sulfur dioxide (SO2), US EPA relies in part on controlled human exposure studies. It has been suggested that evaluating average responses for all participants in these studies may not reflect the responses of sensitive participants in these studies. To evaluate this, we identified controlled exposure studies with multiple exposure concentrations or durations that provided individual-level lung function data. Based on individual lung function responses at specific exposure concentrations and the slope of individual concentration-response curves, we identified 12 participants out of a total of 208 participants in 12 studies who were potentially sensitive to O3, SO2, or sulfuric acid (H2SO4). We did not identify any participants sensitive to NO2. All of these participants were found to be potentially sensitive only at concentrations that were well above the NAAQS (SO2), above likely ambient concentrations (H2SO4), or at concentrations at which the study reported significant lung function effects for all participants (O3). Based on our analysis, average responses for all participants combined adequately reflect lung function responses for potentially sensitive study participants at concentrations in the range of the current NAAQS.