ABSTRACT
Late-Life Depression (LLD) is often associated with cognitive impairment. However, distinction between cognitive impairment due to LLD and those due to normal aging or mild Alzheimer's Disease (AD) remain difficult. The aim of this study was to present and compare the multivariate base rates of low scores in LLD, mild AD, and healthy control groups on a battery of neuropsychological tests. Participants (ages 60-89) were 352 older healthy adults, 390 patients with LLD, and 234 patients with mild AD (i.e., MMSE ≥ 20). Multivariate base rates of low scores (i.e., ≤ 5th percentile) were calculated for each participant group within different cognitive domains (verbal episodic memory, executive skills, mental processing speed, constructional praxis, and language/semantic memory). Obtaining at least one low score was relatively common in healthy older people controls (from 9.4 to 17.6%), and may thus result in a large number of false positives. By contrast, having at least two low scores was unusual (from 0.3 to 4.6%) and seems to be a more reliable criterion for identifying cognitive impairment in LLD. Having at least three low memory scores was poorly associated with LLD (5.9%) compared to mild AD (76.1%) and may provide a useful way to differentiate between these two conditions [ χ ( 1 ) 2 = 329.8, p < 0.001; Odds Ratio = 50.7, 95% CI = 38.2-77.5]. The multivariate base rate information about low scores in healthy older people and mild AD may help clinicians to identify cognitive impairments in LLD patients, improve the clinical decision-making, and target those who require regular cognitive and clinical follow-up.
ABSTRACT
BACKGROUND/AIMS: It is well established that healthy adults obtain low performances when simultaneously interpreting the results of multiple tests. The aim of this study was to estimate the proportion of French-speaking healthy older adults with low scores for the RAPID (Réseau d'Aide au diagnostic et à la PrIse en charge des Détériorations cognitives et de maladies neurologiques chroniques en Franche-Comté et au niveau national) battery test and consider different combinations of test scores within a specific domain and across different domains. METHODS: The prevalence of low scores (i.e., ≤5th percentile) on the 14 RAPID primary measures was calculated from the RAPID normative sample (n = 476), based on 4 ages (50-89 years) and 3 levels of education. RESULTS: A high percentage (40.1%) of the normative sample obtained at least one or more low scores (i.e., false positives). In contrast, the risk of having low scores was much less important (<2%) when considering the combinations of 2 test-scores. CONCLUSION: Low scores are very common in healthy older subjects and are thus not necessarily pathological or indicative of truly impaired functioning. The information derived from a cognitive profile may provide a greater clinical relevance in an individual, since very few of the healthy older adults obtained low scores on combinations of 2 test-scores.
