How smartphone-based study demonstrates research potential
A study at Mount Sinai Health System finds that large-scale asthma research involving human subjects can be conducted in its entirety with iPhones using the Apple ResearchKit framework and a healthcare app developed at Mount Sinai with some partners.
The project demonstrates the ability of consumer-facing technology to play a role in research, providing results that are granular and offer insights not usually found in clinical trials.
“We achieved secure bidirectional data flow between investigators and 7,593 participants from across the United States, including many with severe asthma,” researches report in study results published in the journal Nature Biotechnology. “Our platform enabled prospective collection of longitudinal, multidimensional data (e.g., surveys, devices, geolocation and air quality) in a subset of users over the study period. Consistent trending and correlation of interrelated variables support the quality of data obtained via this method.”
For instance, researchers detected increased reporting of asthma symptoms in regions affected by heat, pollen and wildfires.
Mount Sinai launched the Asthma Mobile Health Study in March 2015, and within six months nearly 50,000 iPhone users downloaded the free app. The app walked those who decided to participate through an informed consent process explaining what the study was about, the risks and benefits, and participant obligations.
“We essentially brought research to the masses, and they could do it on their own time,” says Yu-Feng Yvonne Chan, MD, principal investigator of the study, an emergency physician and an expert in clinical research.
Eligible participants had to be 18 years or old, living in the United States, doctor-diagnosed with asthma and taking an asthma medication. Once enrolled, participants filled out baseline surveys of their health status so any changes that occurred during the study can be measured. Participants were asked to check in daily via an electronic asthma diary that posed five to seven questions on symptoms, possible asthma triggers and whether they took their medications.
Mount Sinai also sent additional information to educate and empower participants, such as educational materials donated by National Jewish Hospital to help participants better understand and manage their disease. Researchers further sought patient feedback on tools they need and developed a “doctor dashboard” that summarized patients’ most useful clinical information, which could be shown to their doctors.
Unlike many clinical research studies that pay participants, this study didn’t offer remuneration, so researchers wanted to give participants information of value, Chan says, adding, “We relied on altruism and a wave of citizen science.”
A total of 7,593 individuals enrolled in the study with 85 percent completing at least one survey; a core group of 2,317 filled out multiple surveys during the six-month trial period.
Among the findings, researchers were able to correlate increased daily asthma symptoms among Washington State participants during an outbreak of wildfires, and they also were able to collaborate patient symptoms with pollen levels and heat. When the six-month period of the study ended, researchers decided to go another full year to pick up seasonal data.
Researchers looked at the quality and validity of data collected from participants, comparing self-reported data to see if it correlated with trusted external resources.
“We wanted to make sure we were not randomly putting in garbage; that peak flow measurements conform to what we would expect, based on patient characteristics such as measuring use rate of an inhaler at night,” Chan says.
Before the formal survey period began, Mount Sinai researchers did two small pilot projects. The first used a module from biotechnology company 23andMe that enabled people to share their genetic data—this was a proof of concept project to demonstrate the willingness to share data, results of which may inform a new study.
Researchers also piloted with Sinai’s Epic electronic health record, having patients enter peak flow (lung function) data into the ResearchKit mobile app to determine if this data can inform decision support and generate an alert to clinicians when an individual is not doing well.
“We need a signal of when a patient is sick,” Chan concludes. “I’m a big believer and advocate for mobile health. 23andMe demonstrates the impact mobile apps can have on clinical research.”