Validation of the self-administered Montreal Cognitive Assessment (MoCA) XpressO: A self-administered digital cognitive screening tool for the adult population in the general public
Sivan Klil-Drori,Katie Bodenstein,Lara Kojok,Johanna Gruber,Shuo Sun,Soham Rej,Harmehr Sekhon,Youssef Ghantous,Ziad Nasreddine,Shuo (Mila) Sun
DOI: https://doi.org/10.1016/j.jagp.2023.02.022
IF: 7.996
2023-09-02
American Journal of Geriatric Psychiatry
Abstract:Introduction The Montreal Cognitive Assessment (MoCA) XpressO is a digital cognitive screening test for use by the general adult population that can be self-administered at home. The purpose of MoCA-XpressO is to allow quick and accessible screening for cognitive concerns in the aging population and to establish if further cognitive testing is needed. This study aims to validate the MoCA-XpressO compared to a digital version of the standard MoCA test (version 8.1) for adults aged 50 years and older. Methods This was a crossover, randomized study design where participants were allocated to undergo both the MoCA-XpressO and the standard digital MoCA test (version 8.1). The order for the administration in the cross-over design was determined randomly for each participant. Participants were recruited from two sites: the MoCA Clinic and a family practice clinic. The MoCA-XpressO is a brief (5-10 minutes) digital screening tool that is self-administered on a tablet device and evaluates the following cognitive domains: processing speed, executive functions, and memory. For the analysis, the standard digital MoCA test scores were dichotomized into two groups according to the total score: "normal" (score ≧25 points) and "abnormal" (<25 points). The following associations between the standard MoCA test and the MoCA XpressO were explored: (1) Total scores, (2) Memory Index Score (MIS) on standard digital MoCA test versus the memory task on MoCA-XpressO, and (3) Executive Index Score (EIS) on standard digital MoCA test versus the logical task on MoCA-XpressO. To analyze these associations, we performed the Chi-square test for categorical variables and the Student's t-test for continuous variables. Logistic regression model was built to quantify the association between the digital standard MoCA and the MoCA-XpressO scores; the accuracy of the model was evaluated by the sensitivity, specificity, and the area under the Receiver Operating Characteristic (ROC) curve (i.e., AUC). Two optimal probability cutoff values for the model were considered, resulting in higher sensitivity and lower specificity for each cutoff, and vice versa. Results We recruited 44 participants aged 50 years and older during October and November 2022. Participants completed the digital standard MoCA test and the MoCA-XpressO in a randomized order. Data analysis was completed in December 2022. The analysis showed a strong association between the standard MoCA MIS score and the MoCA-XpressO memory task score (p-values<.001). Additionally, the standard MoCA EIS score, including the Trails B, clock drawing, and letter-F-fluency subtests showed a strong association with the MoCA-XpressO logical task score (p-values<.03). Finally, the total score of the standard MoCA test showed strong association with the total score of the MoCA-XpressO (p-values<.005). The ROC curve of the logistic model had an AUC around 0.85. With a cutoff probability of 0.45, the sensitivity was 89.2% and the specificity was 62.5%; with a cutoff probability of 0.6, the sensitivity was 71.4% and specificity was 87.5%. Conclusions The high values of AUC (0.85), sensitivity of 89.2% with a cutoff probability of 0.45, and specificity of 87.5% with a cutoff probability of 0.6, indicate that the logistic regression model fits the data well. The MoCA-XpressO scores are strongly associated with scores on the digital standard MoCA test. These results indicate that the score of the MoCA-XpressO memory task may predict the standard MoCA MIS score; also, the results indicate that the score of the MoCA-XpressO logical task may predict the standard MoCA EIS score; finally, the results indicate that the total score of the MoCA-XpressO may predict the total score of the standard MoCA . Therefore, the MoCA-XpressO may fill the unmet need for a valid home-based, self-administered pre-screening tool. This research was funded by This study was funded by MoCA Test Inc.
psychiatry,geriatrics & gerontology,gerontology