Press Releases from the ATS International Conference

HomeAboutNewsroomPress ReleasesPress Releases from the ATS International Conference ▶ Air Pollution May Disrupt Sleep
Air Pollution May Disrupt Sleep

Embargoed until May 21, 4:45 p.m. ET


Dacia Morris
ATS Office 212-315-8620 (until May 17)
Cell Phone 917-561-6545


PRESS CONFERENCE:  May 21, 2017, 4:45 p.m.; Walter E. Washington Convention Center Room, 148 (Middle Bldg., Street Level)

Session: B20 Big Data Comes to Sleep Medicine
Abstract Presentation Time: Monday, May 22, 10:45 a.m. ET
Location: Walter E. Washington Convention Center Room 152A-B (Middle Bldg., Street Level)


ATS 2017, WASHINGTON, DC─High levels of air pollution over time may get in the way of a good night’s sleep, according to new research presented at the ATS 2017 International Conference.

“Prior studies have shown that air pollution impacts heart health and affects breathing and lung function, but less is known about whether air pollution affects sleep,” said lead author Martha E. Billings, MD, MSc, assistant professor of medicine at the University of Washington. “We thought an effect was likely given that air pollution causes upper airway irritation, swelling and congestion, and may also affect the central nervous system and brain areas that control breathing patterns and sleep.”

The researchers analyzed data from 1,863 participants (average age 68) in the Multi-Ethnic Study of Atherosclerosis (MESA) who also enrolled in both MESA’s Sleep and Air Pollution studies. The researchers looked at two of the most common air pollutants: NO2 (traffic-related pollutant gas) and PM2.5, or fine-particle pollution. Using air pollution measurements gathered from hundreds of MESA Air and Environmental Protection Agency monitoring sites in six U.S. cities, plus local environment features and sophisticated statistical tools, the research team was able to estimate air pollution exposures at each participant’s home at two time points:  one year and five years. 

Wrist actigraphy, which measures small movements, provided detailed estimates of sleep and wake patterns over seven consecutive days. This was used to calculate “sleep efficiency”—a measure of the percentage of time in bed spent asleep vs. awake. Researchers found that the sleep efficiency of the lowest 25 percent of participants was 88 percent or less. The research team studied if pollution exposures differed among those in this low sleep efficiency group.

The population was divided into “fourths” according to levels of pollution. The quarter of those who experienced the highest levels of pollution was compared to the quarter with the lowest levels.

The study found:

  • The group with the highest levels of NO2 over five years had an almost 60 percent increased likelihood of having low sleep efficiency compared to those with the lowest NO2 levels.
  • The group with the highest exposures to small particulates (PM2.5) had a nearly 50 percent increased likelihood of having low sleep efficiency.

The authors adjusted for a range of factors, including age, body mass, obstructive sleep apnea, race/ethnicity, income and smoking status. They also adjusted for neighborhood socioeconomic status.  

The researchers were particularly interested in chronic exposure to air pollution and what that long-term exposure might mean for sleep health. “There may be acute sleep effects to short-term exposure to high pollution levels as well, but we lacked the data to study that link,” Dr. Billings said, noting that the parent MESA study is investigating the chronic effects of air pollution on cardiovascular health.

“These new findings indicate the possibility that commonly experienced levels of air pollution not only affect heart and lung disease, but also sleep quality. Improving air quality may be one way to enhance sleep health and perhaps reduce health disparities,” Dr. Billings said.

Future studies, she added, need to explore the association between other air pollutants and sleep, the mechanisms by which these pollutants may disrupt sleep patterns and whether traffic noise is the driving factor contributing to poor sleep quality.

Contact for Media: Martha E. Billings, MD, MSc,


Abstract 11211

Relationship of Air Pollution to Sleep Disruption: The Multi-Ethnic Study of Atherosclerosis (MESA) Sleep and MESA-Air Studies

Authors: M.E. Billings1, D.R. Gold2, P.J. Leary1, A. Szpiro1, C.P. Aaron3, J.D. Kaufman1, S.S. Redline4; 1University of Washington - Seattle, WA/US, 2Harvard School of Public Health - Boston, MA/US, 3Columbia University Medical Center - New York, NY/US, 4Harvard Medical School and Brigham and Women's Hospital - Boston, MA/US

Exposure to air pollution is associated with cardiovascular mortality and pulmonary morbidity, including asthma, COPD, lower respiratory infections, and possibly sleep apnea. Although air pollution also may influence sleep quality through alterations in inflammatory or autonomic nervous system pathways, the relationship between air pollution and sleep has not been well studied. We evaluated the relationship between participant-level estimates of long-term ambient-derived traffic-related air pollution exposure with objective sleep fragmentation.

We analyzed data from a subpopulation of the Multi-Ethnic Study of Atherosclerosis (MESA) who participated in both MESA Sleep and AIR studies. Exposure to traffic related air pollutants (oxides of nitrogen) were estimated at participants’ homes using spatio-temporal models based on cohort-specific monitoring averaged for one and five years prior to sleep assessment. Objective sleep fragmentation was evaluated with wrist actigraphy recorded over seven 24 hour periods. We used multivariate logistic regression models to evaluate for an association of traffic related air pollution with low sleep efficiency (<88%) and increased wake after sleep onset (WASO; > 60 mins). We adjusted for socio-demographics, sleep apnea (AHI>15), short sleep duration (< 6 hrs) and residential socio-economic status (SES).

MESA participants (n=1863) were an average age 68 (+/- 9) years, 46% male, 36% white, 24% Hispanic, 29% black and 12% Asian. A quarter of the sample had < 88% sleep efficiency and 11% had WASO > 60 mins. The highest quartile NO2 exposure level (> 23.7 ppb) over 5 years compared to the lowest (< 10 ppb) was associated with a 57% greater odds of low sleep efficiency in fully adjusted models with a significant test for trend (table 1). The highest quartile compared to the lowest quartile NO2/x average 1 and 5-years exposure levels were also associated with 71-91% greater odds of > 60min WASO.


Higher levels of traffic-related air pollution are associated with greater odds of objectively measured sleep disruption after adjusting for individual and residential socio-demographics. Further research is needed to identify the mechanisms and whether associations are attributable to oxides of nitrogen, traffic noise, other pollutants or environmental exposures that co-vary with traffic.

Table 1: Quartile of NO2 exposure levels averaged over 5 years as predictor of low sleep efficiency after adjusting for age, sex, BMI, < 6 hrs sleep duration and OSA (AHI≥15) (model 1) plus race/ethnicity, income, smoking status (model 2), plus residential SES (model 3), presented as OR (95% CI)


Table 1

Model 1

Model 2

Model 3

NO2 5 yr average exposure




Quartile I

1.00 (ref)

1.00 (ref)

1.00 (ref)

Quartile II



1.19 (0.86,


1.20 (0.86,


Quartile III



1.07 (0.76,


1.11 (0.78,


Quartile IV

1.57 (1.15,


1.46 (1.03,


1.57 (1.06,


1.57 (1.06,