Regression equations for predicting scores of persons over 65 on the Rey Auditory Verbal Learning Test, the mini‐mental state examination, the trail making test and semantic fluency measures

Abstract

Scores on neuropsychological tests are often used to detect abnormal changes in cognition in older persons. Accordingly, it is important to have normative data that allow the abnormality of a test score to be determined precisely and accurately. Regression equations that estimate an expected score based on demographic or premorbid factors can be an efficient method of making normative comparisons. Our aim was to compute regression equations with age, gender and estimated premorbid IQ as predictors of scores on the Rey Auditory Verbal Learning Test (AVLT), the Trail Making Test (TMT), Mini-mental State Examination (MMSE) and measures of semantic fluency. All measures were administered to a group of 272 healthy older persons aged between 65 and 90 during the pre-treatment phase of a study evaluating the effect of nutritional supplements on cognition. Premorbid IQ was estimated using the National Ault Reading Test (NART). Stepwise multiple regression procedures were used to determine the weights to be applied to the predictor variables. Age and premorbid IQ were found to be significantly correlated with all test variables; gender correlated significantly with most scores. Regression equations based on the 3 predictor variables explained between 10% and 30% of the variance of the range of test scores. The use of these equations in clinical practice was illustrated. The significant correlations between the predictor variables and test scores justified computing a set of equations for use in interpreting data from older persons. The abnormality of the difference between predicted and obtained scores provides a convenient index of an individual's current level of neuropsychological functioning.