Clinical Efficacy of Halstead's Impairment Index Versus an Empirical Weighting Procedure

Abstract

The Halstead-Reitan Neuropsychological Battery is commonly used to determine the presence or absence of neurocognitive impairment. A principal indicator used to make this determination is the Impairment Index, an index comprising seven measures from the battery. The Impairment Index has been criticized for assuming equality among the measures. Using the original validation sample for the Halstead-Reitan Neuropsychological Battery, a discriminant function analysis was performed to determine the best set of weights for the seven measures. The results showed that the measures are of differential importance in discriminating between impaired and non-impaired subjects. Cross-validation of the obtained weights on a new sample indicated that the two weighting methods do not differ in classification accuracy. The findings suggest that an empirically derived set of weights is probably no more useful than the easier-to-compute equal weighting procedure commonly used.