LAUSR

Objective: Robust norms for neuropsychological tests may offer superior clinical utility to conventional norms, in their ability to distinguish normal cognitive aging from prodromal dementia. However, the availability of robust norms from midlife, where cognitive changes in those at risk of disease may arise, is limited. This study presents demographically stratified robust norms for tests of verbal memory in Australian women. Method: Participants were from the population-based Women’s Healthy Ageing Project. Baseline (1999 to 2002; n = 368; age range = 53–67years) and follow-up (2012 to 2014; n = 291; age range = 65–80years) measures of word-list and story recall were administered at least 10 years apart. Four samples were identified: conventional (derived from a cross-sectional sample), robust (derived from a longitudinal sample), mild cognitive impairment (MCI) or Alzheimer’s disease (AD), and lost to follow-up. Area under the curve (AUC) values were generated to assess the diagnostic ability of conventional and robust norms using 1 standard deviation and 1.5 standard deviation cut-offs. Results: There were differences between conventional Australian and American normative data for the Consortium to Establish a Registry for Alzheimer’s Disease word-list recall. Individuals who declined to MCI/AD over the follow-up displayed poorer performance at baseline, however no differences in classification ability of robust (AUC range .54 to.64) and conventional (AUC range .51 to .65) norms were observed. Conclusion: Neuropsychological performance in midlife predicted clinical cognitive decline 1 decade later, but conventional and robust norms was similarly predictive of conversion to disease in this cohort. The use of country-specific, representative conventional norms remains a valuable tool for neuropsychologists to assess cognitive performance throughout midlife.