Researchers have found that computerised neuropsychological assessment devices (CNADs) need to improve in order to reach the same level of reliability as traditional pen and paper assessments.
Dr Nicola Gates and Dr Nicole Kochan, clinical neuropsychologists and researchers from the Centre for Healthy Brain Ageing (CHeBA) at UNSW, conducted a review of CNADs against the development standards recommended by the American Academy of Clinical Neuropsychology and National Academy of Neuropsychology in 2012.
The review, published in Current Opinions in Psychiatry, analysed 17 CNADs developed since 2012 when the standards were published. The researchers found that the field has not significantly advanced, with the majority of recommendations inadequately addressed by current CNADs.
The demand for CNADs as a more cost-effective alternative for early detection of late-life neurocognitive disorders has grown in the face of an ageing population and increasing dementia prevalence. CNADs offer a range of potential advantages, including ease of access and fewer requirements for clinician expertise and resources. To date, however, the development of new CNADs has been characterised by insufficient validity and reliability measures.
“Proponents of CNADs assert that technology-based assessments improve upon traditional neuropsychological tests. However, there remain fundamental questions of validity, reliability, normative data and administration, raising the question of whether CNADs are appropriate alternatives,” say co-authors Dr Gates and Dr Kochan.
A range of issues were found to limit the reliability of CNADs, including the application of existing CNADs to new clinical populations, insufficient clinical guidelines, small sample sizes used for normative data, and individual patient factors, such as language and sensory skills and familiarity with technology.
“Cost-effective, valid, efficient and accessible tests for the detection of late-life neurocognitive disorders are crucial, as early identification facilitates appropriate early intervention,” says Dr Gates.
However, Dr Gates argues significant research-based improvement is needed before CNADs are as reliable as traditional testing methods. “Whilst there is a pressing need for innovative and readily applicable cognitive tests, these requirements do not outweigh the necessity for valid measures. The risk of diagnostic errors is potentially high and poor clinical decisions could potentially arise, having significant impact upon individuals in terms of their well being and access to treatment”.
“We recommend clinicians and researchers make informed decisions about CNAD suitability for their clients and their individual requirements based upon published psychometric and other test information.”
