Image courtesy of Propeller Health
A recent study concludes that there is a strong and significant association between daily exposure to fine particulate matter (PM2.5), which refers to tiny particles or droplets in the air that are two and a half microns or less in width, and the frequency of asthma rescue medication usage. Because the particulates are so small, they can get deep into the lungs and greatly affect respiratory conditions.
“This study is unique in that it looked at how air pollution impacts someone’s day-to-day life, in terms of their symptoms,” said Meredith Barrett, vice president of research at Propeller Health, in an interview with MD+DI. She said past studies have had to rely on mortality or hospitalization data to understand how air pollution affects asthma, which represents more rare and severe exacerbation events. “By using this medication-use data, we’re able to look at the daily burden of disease,” she said.
Data was gathered by fitting small sensors onto inhaled medication devices of some 2800 people across the United States. The sensors collected information on when and how frequently (by measuring the number of puffs) people were using their medicine. The sensors were paired with the patient’s smartphones, which provided a GPS location.
This data was merged with environmental information such as temperature, humidity, and precipitation, as well as air pollution data, all of which were collected from government sources. “From that data, we’re able to understand the conditions in which symptoms may occur,” Barrett said.
Propeller has developed models that will look at a person’s past record of medication use. As more and more data is collected on each individual, the models learn what their particular environmental sensitivities are.
“For each Propeller user, we gather information about the environmental conditions in which they use their rescue inhaler. We can use that information to learn about whether environmental conditions on any given day in their location may lead to symptoms for them,” explained Barrett. “It’s helpful to get that information back to the patient, so they learn about their disease, so they can plan their day, whether or where they exercise, or be sure to carry their rescue medication, if it’s going to be a high-risk day.”
The information helps inform clinical care as well, Barrett said. “Providers are getting information so that they can see when their patients are having rescue-medication-use spikes, and then are able to target their care. It helps them to focus in on the patient and give them the right care at the right time,” she said.
A similar model has also been created for the entire country. “We’ve collected hundreds of millions of data points about where and when rescue medication use occurs. We use that data to develop models that estimate when environmental conditions may lead to symptoms for any location in the United States,” said Barrett. “We share that information widely--it’s freely available for people to better self-manage their health.”
“We’re really trying to use the data we’re collecting to better understand respiratory disease in general. What are the conditions that may cause symptoms?” Barrett said. “What are the particular environmental sensitivities, and how do they differ across people? We are going to use this information to gain a better understanding of how the environment influences respiratory disease and try to inform public health.”
In fact, the company has already made strides in the public health arena. Propeller recently collaborated with government and nonprofit partners in Air Louisville, a community program that used Propeller’s connected inhalers to capture data from asthma sufferers in Louisville, KY.
As a result of the information gathered from the sensors, the city is increasing tree coverage in high-risk asthma areas, identifying alternate truck routes that would take traffic away from high-risk neighborhoods, and considering other changes to city-wide zoning policies to address the health impacts of highways and industrial emissions.
“We’re really trying to connect the individual data to a policy level in terms of having impact,” Barrett concluded.