Cross-cultural cognitive assessment of dementia: a meta-analysis of the impact of illiteracy on dementia screening and an evaluation of a transcultural short-term memory assessment

Abstract

Introduction: A major challenge in detecting dementia is that many of the tools used to screen for dementia are affected by linguistic, cultural, and educational factors. As the prevalence of dementia is expected to rise, exploring suitable methods of assessing dementia in diverse populations is becoming increasingly important. Aims: The first aim of this thesis was to assess research on dementia screening tools in literate and illiterate individuals to evaluate the impact of illiteracy on these tools, through a systematic review and meta-analysis. The second aim was to assess the predictive ability and construct validity of the Visual Short-Term Memory Binding Task (VSTMBT), a proposed transcultural tool. Methods: A systematic review identified 37 studies, 27 of which were included in a meta-analysis. In addition, data from a longitudinal study, which involved assessing older adults at three time points over two years, were examined. Data were collected from 72 healthy control participants and 82 participants with a diagnosis of Mild Cognitive Impairment (MCI). Linear mixed models and logistic regression models were used to assess how well the VSTMBT and other baseline measures of cognition predicted future cognitive decline and the development of Alzheimer’s Disease (AD). Finally, partial correlations between baseline neuropsychological assessments were conducted to examine the construct validity of the VSTMBT. Results: In the meta-analysis, multi-level random effects models revealed that literate participants had a significant advantage over illiterate participants in dementia screening tools. In the longitudinal study, the VSTMBT did not significantly predict cognitive decline or conversion from MCI to AD. Partial correlation analyses showed that the VSTMBT did not significantly correlate with other measures of memory and failed to discriminate between measures of executive function, processing speed and visuospatial ability. Conclusions: The meta-analysis highlighted the unsuitability of traditional dementia screening tools for individuals who are illiterate, emphasising the need for the screening tools that consider varying levels of literacy. The results of the longitudinal study were in line with the idea that the VSTMBT has higher predictive value at the preclinical stage compared to the MCI stage of AD. This study highlighted the need for different cognitive tests at different stages of AD progression.