THE MISINTERPRETED AGE EFFECT ON COGNITIVE TEST RESULTS: A PRESENTATION OF TEST NORMS FROM PERSONS WITHOUT UNDERLYING PATHOLOGIES

Abstract

The need for cognitive test norms that are based on persons without underlying brain pathologies becomes increasingly important as the diagnostic work-up of neurocognitive disorders advances and tries to identify cognitive diseases at earlier stages. The aim was to establish cognitive test norms from cognitively healthy elderly persons without Alzheimer (AD) or cerebrovascular pathology and compare the results to traditionally derived test norms. 333 cognitively healthy elderly persons were included from the Swedish BioFINDER study. Mean scores and standard deviations were established for tests of executive function (Trail Making Test, Symbol digit, Stroop and verbal fluency), attention (A Quick Test of Cognitive Speed), memory (immediate and delayed word recall), and visuospatial function (clock drawing). The following groups were created: A) entire cohort, B) no clinical progression during ≥2 years, C) no cerebrovascular pathology (white-matter lesions or infarcts), D) no AD pathology (abnormal levels of cerebrospinal fluid [CSF] Aβ42/40), E) no measurable pathology (all above and no increase in CSF neurofilament light as a measure of axonal injury). In the total population, and in cohorts B and D, there were significant associations between age and the cognitive tests. Interestingly, in those without cerebrovascular pathology (cohort C), the age effect disappeared in tests of attention, memory, and visuospatial function, but remained in all executive tests. When excluding persons with underlying pathologies (cohort E), the cutoffs for cognitive impairment were overall stricter compared to the total population (cohort A) and the traditional method for establishing robust norms (cohort B). Here, we present numerous cognitive test norms derived from truly cognitively normal persons without preclinical AD or cerebrovascular disease. The well-established association between age and cognitive test scores can to a large extent be explained by cerebrovascular pathology in all domains except executive function where other pathological mechanisms might explain the age association. We propose that age is just a proxy for underlying pathological mechanisms. Stratifying norms according age could result in a delayed identification of early cognitive impairment, especially due to cerebrovascular disease.