Application of an Adaptive, Digital, Game-Based Approach for Cognitive Assessment in Multiple Sclerosis: Observational Study

• Main Authors: Hsu, Wan-Yu, Rowles, William, Anguera, Joaquin A, Zhao, Chao, Anderson, Annika, Alexander, Amber, Sacco, Simone, Henry, Roland, Gazzaley, Adam, Bove, Riley
• Format: Article
• Language: English
• Published: JMIR Publications 2021-01-01
• Series: Journal of Medical Internet Research
• Online Access: http://www.jmir.org/2021/1/e24356/

BackgroundCognitive impairment is one of the most debilitating manifestations of multiple sclerosis. Currently, the assessment of cognition relies on a time-consuming and extensive neuropsychological examination, which is only available in some centers. ObjectiveTo enable simpler, more accessible cognitive screening, we sought to determine the feasibility and potential assessment sensitivity of an unsupervised, adaptive, video game–based digital therapeutic to assess cognition in multiple sclerosis. MethodsA total of 100 people with multiple sclerosis (33 with cognitive impairment and 67 without cognitive impairment) and 24 adults without multiple sclerosis were tested with the tablet game (EVO Monitor) and standard measures, including the Brief International Cognitive Assessment for Multiple Sclerosis (which included the Symbol Digit Modalities Test [SDMT]) and Multiple Sclerosis Functional Composite 4 (which included the Timed 25-Foot Walk test). Patients with multiple sclerosis also underwent neurological evaluations and contributed recent structural magnetic resonance imaging scans. Group differences in EVO Monitor performance and the association between EVO Monitor performance and standard measures were investigated. ResultsParticipants with multiple sclerosis and cognitive impairment showed worse performance in EVO Monitor compared with participants without multiple sclerosis (P=.01) and participants with multiple sclerosis without cognitive impairment (all P<.002). Regression analyses indicated that participants with a lower SDMT score showed lower performance in EVO Monitor (r=0.52, P<.001). Further exploratory analyses revealed associations between performance in EVO Monitor and walking speed (r=–0.45, P<.001) as well as brain volumetric data (left thalamic volume: r=0.47, P<.001; right thalamic volume: r=0.39, P=.002; left rostral middle frontal volume: r=0.28, P=.03; right rostral middle frontal volume: r=0.27, P=.03). ConclusionsThese findings suggest that EVO Monitor, an unsupervised, video game–based digital program integrated with adaptive mechanics, is a clinically valuable approach to measuring cognitive performance in patients with multiple sclerosis. Trial RegistrationClinicalTrials.gov NCT03569618; https://clinicaltrials.gov/ct2/show/NCT03569618