DIAGNOSING COGNITIVE DECLINE: COMPARING NIA-AA TO DSM-5 APPROACHES

Abstract

Introduction Both the DSM-5 and NIA-AA (Albert et al., 2011) stipulate methods for assessing and diagnosing cognitive decline but differ in their approach for estimating cognitive ability. The DSM suggests that evidence of cognitive decline is demonstrated either through culturally normed neuropsychological evaluation or through the use of “bedside” assessments such as the Mini Mental State Examination (MMSE) or Montreal Cognitive Assessment (MoCA). By contrast, the 2011 recommendations from the National Institute of Aging – Alzheimer Association (NIA–AA) emphasize the use of objective testing ideally via a neuropsychological evaluation of cognitive impairment in one or more cognitive domains. The current study compared differences in diagnostic rates between these two standards. Since the NIA-AA standards emphasize a more detailed cognitive assessment, we hypothesized that these criteria would identify a larger proportion of individuals with cognitive impairment than the DSM-5 criteria. We then examined possible clinical predictors of this discrepancy, including demographic, psychosocial, health-related, and genetic factors. Methods Data was drawn from 604 participants who were recruited into the Prevention of Alzheimer's Disease With Cognitive Remediation Plus tDCS in Mild Cognitive Impairment and Depression (PACt-MD; ClinicalTrials.gov Identifier: NCT02386670) trial. This is a randomized, controlled trial to evaluate brain stimulation and computerized brain training interventions for the prevention of cognitive decline in older adults. Participants either had a clinical history of mild cognitive impairment (MCI), major depressive disorder in remission, both, or were recruited as healthy controls without history of cognitive difficulty or mental disorder. All participants underwent a clinical and neuropsychological evaluation. Cognitive diagnoses were determined during biweekly consensus conferences. Both the DSM-5 and NIA-AA criteria were used during each consensus conference. The current analysis is based on the baseline data (i.e., prior to the intervention) of the 455 participants for whom a consensus conference was completed. We first compared diagnostic rates of MCI (NIA-AA) vs. mild neurocognitive disorder (DSM-5), as well as dementia (NIA-AA) vs. major neurocognitive disorder (DSM-5). We then used binary logistic regression models to examine demographic (age, education, race, sex, premorbid intellectual ability), psychosocial (English as a primary language, history of depression), health-related (severity of medical illness burden), and genetic (APOE e4 carrier status) predictors of diagnostic discrepancy between DSM-5 and NIA-AA criteria. Analyses were conducted separately for two groups of participants: (i) the group for which NIA-AA diagnoses reflected greater cognitive impairment than did the DSM-5 diagnoses (e.g., NIA-AA diagnosis of dementia but DSM-5 diagnosis of mild neurocognitive disorder); and (ii) the group for which DSM-5 diagnoses reflected greater cognitive impairment (e.g., DSM-5 diagnosis of major neurocognitive disorder but NIA-AA diagnosis of MCI. Results At baseline, there were 103 (22.6%) cases of discrepant diagnoses between the two classification systems: 50 cases had a NIA-AA diagnosis of MCI and no DSM-5 cognitive diagnosis (i.e., normal cognitive status); 41 cases had a NIA-AA diagnosis of dementia and a DSM-5 diagnosis of mild neurocognitive disorder; 12 cases had a DSM-5 diagnosis of mild neurocognitive disorder and no NIA-AA diagnosis. Compared to the 352 cases in agreement, the 91 cases with NIAA diagnoses reflecting greater cognitive impairment were more likely to have a history of major depressive disorder in remission (46% vs. 39%; B=-1.058, p=.003), be an APOE4 carrier (50% vs. 27%; B=-1.120, p=.001), and be older (72?vs. 71 years; B=.063, p=.02). APOE4 status, history of depression, and age were independent predictors. Compared to the 352 cases in agreement, the 12 cases with DSM-5 diagnosis reflecting greater cognitive impairment were less likely to have a history of depression (17% vs. 41%; B=2.414, p=.025). Conclusions While there was overall agreement with cognitive diagnoses when using NIA-AA or DSM-5 criteria, the two systems were discrepant in 22.6% of cases. Discrepancies were more likely to occur in participants with a history of depression, who were APOE4 carriers, or who were older. Overall, NIA-AA criteria yielded diagnoses reflecting greater cognitive impairment, especially in participants with a history of prior depression or those who were APOE4 carriers. These results suggest that more detailed neuropsychological evaluations may be indicated in older adults with a history of depression or who have a genetic vulnerability to dementia. This research was funded by: This project has been made possible by Brain Canada through the Canada Brain Research Fund, with the financial support of Health Canada and the Chagnon Family.