Smartphone App Floodlight Found to Reliably Measure MS Data
“Detection of progression onset or worsening is critical to optimally adapt the therapeutic strategy” patients are using to treat the neurodegenerative disease, the researchers noted, adding that “a more frequent estimate of function during daily life routine is likely to have greater value in tracking MS impairment.”
“Results from this study indicate that test features derived from the Floodlight PoC [proof-of-concept] app hold potential for use in clinical research and practice,” the team wrote.
Data from the analyses were published in the Multiple Sclerosis Journal, in a study titled “A smartphone sensor-based digital outcome assessment of multiple sclerosis.” The study was funded by Roche, which makes Floodlight.
Floodlight is an application designed to perform various MS-related assessments — of the sort that might typically be done by a healthcare professional in a hospital or clinic — using only a smartphone. The app includes three kinds of tests: measures of cognition, assessments of finger dexterity, and measures of walking and gait abnormalities.
The new analyses were based on data from a clinical trial (NCT02952911), completed in 2018, that tested the feasibility of using smartphone-based sensors in MS. That study, which monitored participants over 24 weeks, or about six months, enrolled 76 people with MS, as well as 25 healthy controls. Most of the MS patients had relapsing-remitting MS (90.8%) and relatively mild disease.
The participants tested a proof-of-concept version of the app, and researchers reported high satisfaction among its users. Now, a team at Roche and other institutions conducted a battery of statistical analyses to test how useful the measurements were for assessing cognition and function.
First, the researchers looked at test-retest reliability, which is just what it sounds like — ensuring that, if conditions don’t change, additional testing will continually get a similar result. This was calculated with a mathematical measure called the intraclass correlation coefficient or ICC, looking at scores taken two weeks apart.
In people with MS, the ICCs “were moderate or good, suggesting that reliable data can be captured with the Floodlight PoC [proof-of-concept] app,” the researchers wrote.
The team then looked for correlations between the different tests measured by the app, and comparable real-world assessments. In general, they found that the smartphone-based results closely mirrored the real-world tests, particularly for measuring cognition. For example, scores on a digital version of the symbol digit modalities test — a measure of cognition — were highly similar to scores on an oral version of the test.
The only app-based measure that did not significantly correlate with its real-world equivalent was an assessment balance. Also of note, most of the app-based measures were significantly associated with the participants’ levels of disability. Significant correlations with MRI-based measures of disease also were noted.
“Here, we provide the first evidence that the Floodlight PoC app can reliably capture clinically relevant data measures of functional impairment in [people with MS],” the researchers concluded.
Importantly, because the app is on a smartphone, it has the ability to record data passively — that is, as people go about their lives with their phones in their pockets — as well as measuring active movements. Analyses showed significant correlations between passive measurements taken by the app, and assessments of gait, or a person’s way of walking, done in a clinical setting.
“It has been suggested,” the researchers wrote, “that signs of gait alteration may be more pronounced during daily life than in conventional in-clinic metrics, thereby highlighting the importance of capturing out-of-clinic performance through passive monitoring.”
Data from such passive monitoring may help healthcare providers to “improve the translation of clinical findings to meaningful care as it informs on the patients’ true abilities during daily life activities,” the team wrote.
Also of note, statistical analyses indicated that two measures of dexterity on the app — the Pinching Test and the Draw a Shape Test — were both independently associated with an in-person measure of dexterity, the 9-Hole Peg Test. In other words, these analyses indicated that the two digital measures were assessing distinct aspects of dexterity that affect the in-person measure.
“This supports the concept that specific sensor-based test features can capture performance outcome information currently not recorded with commonly used in-clinic assessments,” the researchers wrote, adding that the result “exemplifies the potential of sensor data to characterize functional impairment beyond a single summary score that is typically recorded for in-clinic performance outcome measures.”
The investigators said more and better information may be culled by incorporating additional test features.
“Future work should explore the use of this technology in broader clinical applications and focus on establishing the clinical relevance for the additional information it can provide,” the team concluded.