By linking a population’s activity patterns to time- and location-specific air quality data, our research team is able to develop models which can be used to study differing rates of ozone exposure—a factor that could contribute to higher rates of medical emergencies.

urban skyline with visible variations in air quality

Problem

Metropolitan analytics research has revealed that the consequences of elevated ozone exposure may be more immediate—and deadly—than public health decision-makers first suspected.

Ambulance deployed to scene of medical emergency

Our project builds off a comprehensive study from researchers at Rice University and the city of Houston, Texas. This group found a significant link between incidents of out-of-hospital cardiac arrest and increases of 20 parts per billion (ppb) in ozone within the previous three hours.

For policymakers, these findings raised a number of questions about population health: Which populations regularly face the highest levels of ozone exposure as they move throughout their daily lives? Are there areas of the city or times of day that are closely associated with ozone levels which can increase the risk of cardiac arrest?

New methods were needed to link a city's air quality data to the population’s activities throughout the area. Such an analysis would have the potential to identify public health threats before they could develop into medical emergencies. 

Methods

Working in concert with the Biocomplexity Institute’s Network Dynamics and Simulation Science Laboratory, SDAL developed an information model to link ozone exposure to demographically representative “synthetic populations.”

map displaying regional variations in ozone levels

The resulting simulation system ran 4.4 million individuals through the course of a “normal” day in Houston, Texas, spread out across 1.8 million activity locations. By linking these movement patterns to time- and location-specific air quality data, our researchers were able to create a detailed representation of the population demographics facing health risks from ozone exposure.

These methods offer several advantages over conventional policy studies, including the ability to identify at-risk populations proactively.

Impact

Our initial findings have confirmed that ozone levels can vary widely within a geographic area. Where individuals live, work, and play all contribute to their level of risk, making targeted health policy measures all the more essential for effective intervention.

The methods employed in this project could enable cities like Houston to issue warning when ozone levels are high, encouraging vulnerable populations to remain inside. They also have the potential to help emergency managers and medical providers strategize the deployment of medical resources to the specific demographic groups and geographic locations where they are most needed.

Publications

  1. B. Pires, G. Korkmaz, K. Ensor, D. Higdon, S. Keller, B. Lewis, and A. Schroeder. (2015). Towards an in silico Experimental Platform for Air Quality: Houston, TX as a Case Study. Computational Social Science Society of America Conference. Santa Fe, NM.
     
  2. K. Ensor, L. Raun, and D. Persse. (2013). A case-crossover analysis of out-of-hospital cardiac arrest and air pollution.  Circulation, AHA-113.

Back to top