Measuring cognitive change in older adults: reliable change indices for the SIDAM

Abstract

The interpretation of changes in test scores in older adults should account for potential practice effect, normal decline and regression to the mean.

The Assessment of Changes in Cognitive Functioning in the Elderly: Age- and Education-Specific Reliable Change Indices for the SIDAM

“…Consistent with the findings of previous research [16], group level analyses via Friedman and Wilcoxon tests indicated small but significant evidence for practice effects. The small magnitude of practice effects (<1 point) could be explained by the fact that older adults’ abilities to benefit from learning were found to have declined in comparison to middle-aged adults’ abilities to benefit from previous exposure to test materials due to age-related changes in fluid intelligence, perceptual speed and attention/concentration capacity [29,30,31].…”

“…Results of Kruskal-Wallis testing may reflect that extreme pre-test SIDAM test scores may yield to less extreme post-test SIDAM test scores closer to the SIDAM group mean score, a statistical phenomenon that was also observed and reported in other studies [16,32]. In this study, data are provided to calculate individual RCI scores according to the method of Hsu [11], which takes into consideration possible effects of regression to the mean and may substantially contribute to the precise measurement of changes in test scores.…”

“…The test-retest reliability coefficients observed in this study were acceptable and similar to the test-retest reliabilities observed in a previous study [16]. In the next step, RCIs were derived from the psychometric qualities of the SIDAM to estimate changes.…”

“…reliable) changes in cognitive test performance and underlying cognitive abilities and most likely reflect measurement error, practice effects and/or normal aging effects. In comparison, Hensel et al [16] reported RCI scores ranging from at least 4 to up to 7 points on the SIDAM, indicating reliable change. Presumably, the larger sample size in this study allowed more precise estimations, leading to smaller ranges of RCI scores.…”

“…Change norms (i.e. RCIs) for the accurate classification of cognitive change in the elderly according to SIDAM test scores have been published [16]. However, that study was based on a relatively small sample of older adults (n = 119), and the computation of RCIs did not include correction for the potential influences of sociodemographic variables such as age, education and gender.…”