Prediction of cognitive impairment in the elderly by analysing their performance in verbal fluency and in sustained attention

Cognitive decline is related to different kind of biopsychosocial variables in the elderly. AIM. To identify which cognitive variables are important to differentiate between healthy elderly people and elderly people with cognitive impairment in order to propose a brief screening assessment for cognitive impairment in which education is not relevant. 264 subjects were divided into two groups taking into account the cutoff point 23/24 in the MEC, and all of them were assessed in cognitive functioning, sustained attention, cognitive plasticity, working memory, and verbal fluency. Discriminant analysis showed that the cognitive variables which best explain cognitive impairment were verbal fluency, sustained attention and cognitive plasticity. The area under the ROC curve for the verbal fluency task was 0.811 and for the sustained attention task was 0.777. A cutoff point with 10 words in the verbal fluency task showed a sensitivity of 74% and a specificity of 80% for healthy elderly people, while a cutoff point with 33 correct answers in the sustained attention task showed a sensitivity of 68.4% and a specificity of 68.6% for the same group. These findings suggest that a high performance on verbal fluency and sustained attention protect against cognitive impairment in the elderly so that verbal fluency and sustained attention tasks are good tests for cognitive impairment screening purposes.