Evidation Health, a technology company that generates insights from real-world behavioral data collected through devices, has kicked off a new effort to better understand individuals' health behaviors and concerns throughout the COVID-19 pandemic.

Yesterday the company announced that it has recruited 100,258 people living in the U.S. into a longitudinal survey through the Achievement app, Evidation's patient engagement and rewards product for health organization customers.

These participants – a subsample of the app's nearly four million users who live in all 50 states and the District of Columbia – will be regularly providing their COVID-19 perceptions, changes in their behavior and any device-tracked activity data that is synced with the app.

“One of the great things about Achievement is our ability to layer permissioned data from connected devices and apps with self-reported outcomes we capture through surveys," Ernesto Ramirez, senior data scientist at Evidation Health, told MobiHealthNews in an email. "The trust and transparency we build with our members allows us to conduct novel research into the relationship between person-generated health data and important health conditions and outcomes like influenza and COVID-19.

The digital survey has already collected some baseline data in its first few days, the company wrote on its website. For instance, of respondents answering between March 12 and March 15 fewer (23%) said they believed that the U.S. was prepared or very prepared for the outbreak (23%), while roughly half ( said they believed that the virus represented a major threat to public health. Respondents said that they were generally washing their hands more frequently during the previous seven days (64%), but less often reported that they were avoiding large gatherings (34%).

There were also differences in replies between participants who did or did not have health insurance. The latter group more often said that they would not seek care if they developed coronavirus symptoms (8% versus 21%), and those who would more often said that they would most likely go to an emergency room (19% versus 26%).

A state-by-state breakdown of participant responses also illustrated differences in opinion among the populace. Here, Evidation highlighted the contrast between Washington, D.C. and Louisiana, which as of March 15 had the third- and fourth-highest number of cases per 100,000 residents. The former's respondents were the least assured in the country's preparedness for the disease and most often reported preventive behavior changes, while the latter's residents were among the most confident, and did not change their behaviors to a similar extent.

WHY IT MATTERS

The COVID-19 pandemic is escalating at a rapid pace, with the CDC disclosing thousands of new cases reported within the last day alone. Evidation's approach to polling can quickly yield a large quantity of information from generally decentralized responders, and complement those qualitative responses with quantitative activity data collected by their devices.

"Our ability to tap into a large virtual cohort of Americans shows the potential for leveraging new models of engagement and data collection not only for research purposes but meaningful and rapidly scalable public health surveillance," Ramirez said. "And because we're able to communicate with our members, we can track how behaviors and attitudes change over time as social and health dynamics shift in the future.”

With that being said, the strengths of these types of large, app-based studies can come at a cost. According to a recent meta-analysis of eight digital studies with similar designs, these approaches are often subject to low retention rates and participant samples that often skew younger and whiter than the general population.

THE LARGER TREND

Even in the earlier days of COVID-19's U.S. spread, health organizations and startups alike had developed a number of data-driven tools to track the outbreak. In February, Boston Children's Hospital's John Brownstein and Buoy Health's CEO Dr. Andrew Le described one such example that combines the former's data science tools with the latter's triaging chatbot.

For its own part, Evidation Health has a hefty collection of app-based health studies both logged and ongoing. Some of these have also been conducted in collaboration with big names. It was just a few weeks ago that its name was attached to Apple and Johnson & Johnson's Heartline Study of Medicare seniors.