Evaluation of the usability of a brief computerized cognitive screening test in older people for epidemiological studies.

Background: Computerized cognitive testing has the potential to be an effective way to assess and monitor cognition in large neuroepidemiological studies. CogState is a game-like computerized test with demonstrated validity and reliability that has shown sensitivity to decline in older individuals over time. This study aimed to evaluate the serial usability of the test specifically within an older community cohort. Methods: The test battery was administered to healthy volunteers aged 50 years and above at 3-month intervals over 12 months in a community setting. Test usability was examined in terms of acceptability, efficiency and stability. Results: Of 301 subjects (age: 61.9 ± 7.2 years), 87% completed the study. In addition, 85% completed the first test within the allowed time and passed integrity criteria with their performance improving and stabilizing at subsequent visits. The computerized battery required 15 min for administration on average, allowing 263 patients to be assessed on 5 occasions by 2 assessors. All tasks showed stability and a high test-retest reliability with serial administration. Conclusions: This computerized test was shown to have good acceptability, efficiency and stability for the repeated assessment of cognitive function in older people. Together with its demonstrated sensitivity to cognitive impairment and cognitive change, these data suggest that it would be a useful tool for application in neuroepidemiological studies.