Practicality of a computerized system for cognitive assessment in the elderly

Abstract 

Background

Early detection and diagnosis are critical to dementia care. However, many early cases remain undiagnosed as a result of the impracticality of neuropsychological testing, particularly in primary care. Mindstreams is an office-based computerized system for measuring cognitive function in multiple domains, with demonstrated validity, test-retest reliability, and sensitivity to treatment effects. This study evaluated its feasibility for assessment of the elderly.

Methods

Usability data were collected after each of 2,888 consecutive initial-visit testing sessions at the first 11 clinical centers to use Mindstreams. The χ² goodness-of-fit test was employed to determine whether patients and supervisors more often rated tests easy versus hard to use. Separate analyses were run for non–computer users, patients older than 75 years, and poor performers (≤1 standard deviation on overall battery performance).

Results

For all patients (n = 2,888; age, 64.7 ± 18.2 years), 83% rated the tests easy to use (P < .001). Seventy-three percent of non–computer users, 70% of patients older than 75, and 69% of poor performers rated them easy to use (Ps < .001). Supervisor ratings and ease of understandability ratings were similar. For all patients, 76% of supervisor ratings indicated no patient frustration (P < .001). Seventy-eight percent of ratings for non–computer users, 76% for patients older than 75, and 74% for poor performers indicated no frustration (Ps < .001).

Conclusions

Mindstreams was easily employed, including in patients with considerable cognitive impairment, supporting its practicality for in-office cognitive assessment of the elderly. The availability of such valid and practical assessment suggests the feasibility of integrating the technology within a clinical algorithm for improved detection of cognitive decline.

Keywords: Computerized cognitive assessment, Dementia, Early detection, Cognitive assessment, Usability, Neuropsychology, Neuropsychological testing, Clinical algorithm, Diagnosis, Cognitive, Memory, Mild cognitive impairment