Machine Learning Enhances the Efficiency of Cognitive Screenings for Primary Care.

BACKGROUND: Incorporation of cognitive screening into the busy primary care will require the development of highly efficient screening tools. We report the convergence validity of a very brief, self-administered, computerized assessment protocol against one of the most extensively used, clinician-administered instruments-the Montreal Cognitive Assessment (MoCA). METHOD: Two hundred six participants (mean age = 67.44, standard deviation [SD] = 11.63) completed the MoCA and the computerized test. Three machine learning algorithms (ie, Support Vector Machine, Random Forest, and Gradient Boosting Trees) were trained to classify participants according to the clinical cutoff score of the MoCA (ie, < 26) from participant performance on 25 features of the computerized test. Analysis employed Synthetic Minority Oversampling TEchnic to correct the sample for class imbalance. RESULTS: Gradient Boosting Trees achieved the highest performance (accuracy = 0.81, specificity = 0.88, sensitivity = 0.74, F1 score = 0.79, and area under the curve = 0.81). A subsequent K-means clustering of the prediction features yielded 3 categories that corresponded to the unimpaired (mean = 26.98, SD = 2.35), mildly impaired (mean = 23.58, SD = 3.19), and moderately impaired (mean = 17.24, SD = 4.23) ranges of MoCA score ( F = 222.36, P < .00). In addition, compared to the MoCA, the computerized test correlated more strongly with age in unimpaired participants (ie, MoCA ≥26, n = 165), suggesting greater sensitivity to age-related changes in cognitive functioning. CONCLUSION: Future studies should examine ways to improve the sensitivity of the computerized test by expanding the cognitive domains it measures without compromising its efficiency.

Subjective Versus Objective Assessment of Cognitive Functioning in Primary Care.

PURPOSE: This study examined the clinical utility of highly efficient subjective and objective screens of cognitive impairment. METHOD: Participants (N = 124, age ≥ 65, mean = 73.59, SD = 6.26) completed a 2-item questionnaire of subjective memory functioning, a brief computerized cognitive test, and the Montreal Cognitive Assessment (MoCA). Next, participants were assigned to 1 of 4 conditions, based on their subjective (low/high) and objective (impaired/unimpaired) levels of cognitive functioning. Further analysis divided the sample into age-based groups (ie, age < 75, age ≥ 75). RESULTS: The proportion of participants in the impaired subsample (ie, MoCA < 26), who reported a high level of subjective concern about their memory, was low (ie, 0.15). Among unimpaired participants, analysis detected significant group differences across subjective memory levels (P < .0003) and age (P < .005) categories on one of the three tasks of the computerized test (ie, cognitive control). In contrast, the MoCA offered no differentiation between these groups. CONCLUSION: Screening protocols in which cognitive testing is administered subsequent to patient complaint are prone to underdiagnosis. In addition, common dementia screens are insensitive to subjective deficits and healthy cognitive aging. Therefore, they may lead to dismissing valid concerns that deserve preventive attention. Primary care needs efficient screening tools that are sensitive to prodromal decline.