Reliability of online, remote neuropsychological assessment in people with and without subjective cognitive decline

TL;DR

This study examines the reliability of online neuropsychological tests for people with and without subjective cognitive decline, finding some tests are reliable and could offer a cost-effective alternative to in-person assessments.

Contribution

The study evaluates the reliability of remote neuropsychological tests in people with and without subjective cognitive decline, identifying tests suitable for online monitoring.

Findings

Test-retest reliability of the NeurOn tests ranged from poor to good, with the strongest reliability for Sustained Attention to Response Test and Picture Recognition.

SCD participants scored better than Non-SCD on some baseline tests, but differences were not significant after Bonferroni correction.

No differences were found between SCD and Non-SCD in longitudinal change over six months.

Abstract

Online, remote neuropsychological assessment paradigms may offer a cost-effective alternative to in-person assessment for people who experience subjective cognitive decline (SCD). However, it is vital to establish the psychometric properties of such paradigms. The present study (i) evaluates test-retest reliability of remote, online neuropsychological tests from the NeurOn software platform in people with and without SCD (Non-SCD) recruited from the general population; and (ii) investigates potential group differences in baseline performance and longitudinal change. Ninety-nine participants (SCD N = 44, Non-SCD N = 55) completed seven tests from the NeurOn battery, covering visual and verbal memory, working memory, attention and psychomotor speed. Sixty-nine participants (SCD N = 34, Non-SCD N = 35) repeated the assessment six (+/- one) months later. SCD was classified using the Cognitive Change Index questionnaire. Test-retest reliability of the NeurOn test outcome measures ranged from poor to good, with the strongest evidence of reliability shown for the Sustained Attention to Response Test and Picture Recognition. The SCD group was significantly older than the Non-SCD group so group differences were investigated using analysis of covariance whilst controlling for the effect of age. SCD scored significantly better than Non-SCD for Digit Span Backwards (maximum sequence length) and Picture Recognition (recall of object position) at baseline. However, these were not significant when using the Bonferroni-adjusted alpha level. There were no differences between SCD and Non-SCD in longitudinal change. The results suggest online, remote neuropsychological assessment is a promising option for assessing and monitoring SCD, however more research is needed to determine the most suitable tests in terms of reliability and sensitivity to SCD.

A considerable proportion of the older adult population experiences subjective decline in their thinking skills even though they score within ‘normal’ limits on screening tests for mild cognitive impairment or dementia. Research suggests that, for a small percentage of these people, their experience of a decline in their thinking skills might indicate an early stage of dementia. It is important for research to identify the earliest markers of dementia as this is when treatments may be most effective. By harnessing computing technology to improve on the accuracy and availability of cognitive assessments, we may be able to identify early and subtle cognitive changes caused by dementia. This study investigated whether online and remote cognitive assessment is a reliable method to assess and monitor thinking skills in the general older adult population. We were able to identify tasks which showed the best evidence for reliability when completed online and remotely by people with and without a subjective experience of cognitive decline, and therefore may be appropriate for monitoring thinking skills in people who are concerned about their cognitive ability. Our findings suggest online cognitive assessment may be a useful and cost-effective alternative to in-person clinic-based assessment, although care should be taken to determine reliability and sensitivity of tests used.

Figures