Researchers who combined data from wearables, handheld devices, and digital apps were able to tell the difference between people with early dementia and those without the progressive disease. That’s according to the results of a 12-week study involving older participants in real-world settings with known cognitive status. The study indicates that digital biomarkers may help diagnose the chronic condition in earlier stages when treatment might be most helpful.
The study was a collaboration between researchers from Eli Lilly, Evidation Health, and Apple. It was published in The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD ’19), August 4–8, 2019, Anchorage, AK, USA.
“Over the past few years, we’ve seen how data and insights derived from wearables and mobile consumer devices have enabled people living with health conditions, along with their clinicians, to better monitor their health,” said Nikki Marinsek, Ph.D., a first author and data scientist at Evidation Health in a press release.
“We know that insights from smart devices and digital applications can lead to improved health outcomes, but we don’t yet know how those resources can be used to identify and accelerate diagnoses,” Marinsek said. “The results of the trial set the groundwork for future research that may be able to help identify people with neurodegenerative conditions earlier than ever before.”
The goal of the current study was to determine whether, in real-world settings, Apple devices in combination with mobile applications could help identify cognitive and behavioral differences among the study participants with and without mild cognitive impairment.
The 113 participants in the study, ages 60–75, either had no dementia, mild cognitive impairment, or mild Alzheimer’s disease dementia. The devices they used included Apple’s iPhone, Apple Watch, iPad, and Beddit sleep monitoring device. These were combined with digital apps developed by Evidation Health, a company whose research focuses on using behavioral data in the context of health.
Evidation established a secure study platform to obtain study participants’ consent to collect and analyze 16 terabytes of data across a number of sources, including: passively derived sensor data from the smart devices, questionnaires about mood and energy, and simple assessment activities on the Digital Assessment App. The App included psychomotor tasks, such as dragging one shape onto another or tapping a circle as fast and as regularly as possible, reading tasks and a typing task.
Data were collected in six categories: gross motor function, autonomic (heart rate), circadian, behavioral/social/cognitive, fine motor control and language. The biomarkers that differentiated the two groups included: slower typing, less regularity in walking and later first steps taken during the day, fewer text messages received and sent, greater reliance on helper apps (Clock and Siri) and poorer survey compliance. For example, symptomatic individuals answered the daily one-question surveys less often than healthy controls and, when they did respond, tended to respond later in the day
The authors concluded that data obtained through the use of Apple devices suggested an ability to differentiate between individuals with mild cognitive impairment and mild Alzheimer’s disease dementia, and those without symptoms in ways not previously detected through common clinical screening tools.
The authors also pointed to some limitations of their work. “First, some of the patterns we found are associated with behaviors that are modifiable. Shifts in behaviors not associated with the progression of the underlying disease must be properly accounted for in future work. Further, there is the potential that a passive measure of cognitive performance could be self-reinforcing; without the knowledge of actions to take to mitigate any potential decline, the knowledge of the decline might cause decline itself,” they wrote.
The findings could lead to early screening or detection tools for neurodegenerative conditions, he added. Using everyday devices in this way is the goal, said Christine Lemke, co-founder and president of Evidation Health. “These early findings suggest the potential of novel digital measures that are based on data generated and controlled by individuals.”
