Examination of the accuracy of Cogniciti’s self‐administered, online, Brain Health Assessment in detecting amnestic mild cognitive impairment

Abstract

Background: As the population ages, there is an increasing need for easily administered online assessments sensitive to mild cognitive difficulties not only for clinical practice but also to support research trial initiatives. There are currently few digital cognitive assessments readily available for self-administration online. Cogniciti's Brain Health Assessment (BHA) provides statistically normed results for those aged 20-94, allowing for its use across the adult lifespan. The BHA is, therefore, uniquely positioned to provide self-administered clinical screening for cognitive impairment. Objectives: Our team has recently presented data indicating the accuracy of the BHA. The aim of the present study was to further examine the utility of the BHA compared to that of the Montreal Cognitive Assessment (MoCA) in detecting amnestic mild cognitive impairment (aMCI) in a sample of community dwelling older adults, and to provide indication of convergent validity between tasks on the BHA and standard clinical neurocognitive assessment tasks measuring similar constructs. Method: Using a cross-sectional design, community-dwelling older adults aged 60-89 completed a gold standard neuropsychological assessment to determine a diagnosis of normal cognition (NC) or aMCI (by consensus of 3 staff neuropsychologists). Each participant also completed the BHA and MoCA. Penalized logistic regression (PLR) analyses were used to examine which specific BHA tasks and measured demographic variables contributed to this test's predictive utility in detecting aMCI; MoCA variables were similarly modeled along with demographics, in a separate PLR analysis. Diagnostic accuracy of the PLR models for the BHA and MoCA were compared using area under the receiver operating characteristic curve (ROC-AUC) analyses. Pearson correlations were examined between traditional neuropsychological measures and BHA tasks to assess convergent validity. Result: 91 participants met inclusion criteria (51 aMCI, 40 NC). PLR modelling for the BHA indicated Face-Name Association, Spatial Working Memory, and age predicted aMCI (ROC-AUC = 0.76; 95%CI: 0.66, 0.86). Optimal cut-points resulted in 21% classified as aMCI (positive), 23% negative, and 56% inconclusive. For the MoCA, digits, abstraction, delayed recall, orientation, and age predicted aMCI (ROC-AUC = 0.71; 95%CI: 0.61, 0.82). Optimal cut-points resulted in 22% classified positive, 8% negative, and 70% inconclusive (standard logistic regression [LR] results will also be presented). The BHA model classified fewer participants into the inconclusive category and more as negative for aMCI, compared to the MoCA model (Stuart-Maxwell p = 0.004). Convergent validity of the BHA tasks was supported by moderate to large effect correlations between BHA and standard clinical tasks. Conclusion: There is a need for brief cognitive assessment tools that can detect early signs of cognitive decline in clinical practice and aid in appropriate participant selection into research trials. Online, self-administered measures provide potential for broad reach of cognitive screening to individuals who may otherwise not be easily reached to receive clinical or research assessments. The self-administered BHA showed similar detection of aMCI as compared to the clinician-administered screener (MoCA), with fewer participants classified inconclusively. Given the BHA is an online, self-administered task, this measure has the potential to save practitioners and researchers time, decrease unnecessary referrals for comprehensive assessment to determine presence of aMCI, and improve selection into trials of potentially efficacious early interventions.