Household levels of nitrogen dioxide and pediatric asthma severity

Abstract

Background: Adverse respiratory effects in children with asthma are associated with exposures to nitrogen dioxide (NO2). Levels indoors can be much higher than outdoors. Primary indoor sources of NO2 are gas stoves, which are used for cooking by one-third of U.S. households. We investigated the effects of indoor NO2 exposure on asthma severity among an ethnically and economically diverse sample of children, controlling for season and indoor allergen exposure.

Methods: Children 5-10 years of age with active asthma (n = 1,342) were recruited through schools in urban and suburban Connecticut and Massachusetts (2006-2009) for a prospective, year-long study with seasonal measurements of NO2 and asthma severity. Exposure to NO2 was measured passively for four, month-long, periods with Palmes tubes. Asthma morbidity was concurrently measured by a severity score and frequency of wheeze, night symptoms, and use of rescue medication. We used adjusted, hierarchical ordered logistic regression models to examine associations between household NO2 exposure and health outcomes.

Results: Every 5-fold increase in NO2 exposure above a threshold of 6 ppb was associated with a dose-dependent increase in risk of higher asthma severity score (odds ratio = 1.37 [95% confidence interval = 1.01-1.89]), wheeze (1.49 [1.09-2.03]), night symptoms (1.52 [1.16-2.00]), and rescue medication use (1.78 [1.33-2.38]).

Conclusions: Asthmatic children exposed to NO2 indoors, at levels well below the U.S. Environmental Protection Agency outdoor standard (53 ppb), are at risk for increased asthma morbidity. Risks are not confined to inner city children, but occur at NO2 concentrations common in urban and suburban homes.

Figure 1

Distribution of health outcomes: observations by season of monitoring for asthma severity score (A), days of wheeze (B), night symptoms (C) and rescue medication use (D).

Figure 2

Distribution of any wheeze (A), rescue medication use (B), maintenance medication use (C) and asthma severity score (D): observations for all monitoring periods by mother’s education level.

Figure 3

Exposure-response relationships between health outcome and NO₂ (log concentration as a continuous variable) illustrated with constrained, natural spline functions (solid lines) with 95% confidence limits (small dashed lines) and threshold function (bold dashed line) from fully adjusted, hierarchical ordered logistic regression models for asthma severity score (A), wheeze (B), night symptoms (C), and rescue medication use (D). Also shown is a histogram of NO₂ levels measured in subjects’ homes (panel D) for all observations (thin border) and observations taken in homes of gas stove users (bold border).