A tale of two norms: the impact of normative sample selection criteria on standardized scores in older adults

Abstract

Objective: Normative sample composition can vary notably as a function of exclusionary criteria in older adult samples given that typical medical conditions of exclusion become more prevalent with age. Older adult norms may consist of typically aging or healthy aging individuals according to the stringency of the exclusion criteria, a difference illustrated when comparing two normative samples for the RBANS: the original standardization sample and the Oklahoma sample. Methods: The current study examined the clinical impact of norm selection in older adults by analyzing impairment classification rates for 105 older adult behavioral health inpatients. Rates of impaired performance (scaled score ≤4 or its normalized equivalent) were compared across RBANS subtests using standardization sample vs. Oklahoma sample normative data in patients aged 70–79 (n = 57) and 80–89 (n = 48). Results: Across most RBANS subtests, utilization of the original standardization sample vs. Oklahoma sample norms resulted in a greater proportion of individuals classified as impaired. These findings occurred across both age groups and persevered when holding constant the sample normalization process. Overall, use of the standardization sample vs. Oklahoma sample normative data were 1–11 times more likely to indicate impairment, with one of the more notable differences seen on the Coding subtest in the 80–89 sample (65% vs. 8% labeled as impaired). Conclusions: Sample composition and exclusionary criteria require consideration when selecting an appropriate normative sample against which to compare older adult test performance. Salient differences between typically aging vs. healthy aging older adult norms are discussed.