Smartphones Add New Dimension to Exposure Studies

Scientists from UC Berkeley are the first to demonstrate how apps  installed on smart phones can substantially improve pollution exposure estimates used in environmental and public health studies.

The paper by Edmund Seto and Michael Jerrett from the School of Public Health and collaborators at the Centre for Research in Environmental Epidemiology in Barcelona, Spain, and Imperial College London, UK, will change how exposure assessments are developed in epidemiology studies, the authors predict.

Traditionally, epidemiological studies have relied on data collected at fixed-site monitoring stations, which measure pollution on a city-wide scale. More recently, researchers have modeled within-city pollution levels using portable monitors at the homes of study subjects. But people are mobile throughout the day, leaving some exposures unaccounted for when only home address estimates are used.

Now, web-enabled smartphones are allowing scientists to track a person’s location using Global Positioning Systems (GPS), improving estimates of exposure to pollution and other environmental hazards.

The app developed by Seto, called CalFit, not only records time-location patterns, it simultaneously measures personal energy expenditure using the phone’s accelerometer. Scientists then combine the information with maps of air pollution to develop personalized exposure estimates.

Accelerometers have been used with GPS in previous physical activity studies, according to the authors, but as separate devices rather than an integrated system like CalFit, which simultaneously measures location and intensity of movement.

The study team recruited 36 participants living in Barcelona, Spain, from November 2010 to February 2011. In addition to using CalFit on an android phone, participants manually recorded their movements. Their self-reported activities were cross-validated with CalFit.

For each participant, researchers overlaid data from CalFit with pollution exposures from street-scale maps of nitrogen dioxide (NO2) then estimated time-weighted daily concentrations. They found as much as a 15 micrograms per cubic meter (µg/m3) difference in NO2 between personal exposure concentrations and home address concentrations. When they applied additional spatial and micro-environmental adjustment factors, the difference in NO2 exposure concentrations increased by up to 50 µg/m3.

“One of the major innovations of the study is that the activity data collected allows us to estimate inhalation simultaneously as the person moves through the pollution exposure field, “ says Jerrett. “We were able to show, for example, that time in transit accounted for only six percent of people’s time budget, but in transit exposures accounted for 24 percent of their daily inhaled NO2.”

“The findings are not that surprising,” says Seto. “We know that air pollution levels are generally higher when a person is on or near a busy road than at their residence.  And, we know that people will potentially breathe in more air pollution when they are physically active. We just haven’t had a simple all-in-one instrument that can be used to measure how important time-location and activity patterns are on real-world exposures until now.”

“This is a huge innovation in exposure science that is going to allow us to assess more accurately the health effects of air pollution and other environmental exposures,” notes Jerrett.

“This is just the beginning,” says Seto. “The CalFit app was designed to be flexible so that it can be used in a variety of environmental health studies. CalFit is currently being used in research assessments of greenspace exposure, physical activity, diet, and stress.”

The National Academies of Sciences, commissioned by EPA and NIEHS, described the shift toward individualized exposure assessments in a report titled, Exposure Science in the 21st Century: A Vision and a Strategy. For more information, visit

Read the journal article:

back to top