Measuring emotion recognition: Added value in diagnosing dementia of the Alzheimer's disease type

Neuropsychological tests, particularly for episodic memory, are used to classify patients in memory clinics. Still, the differential diagnosis between dementia of the Alzheimer's disease type (Dementia-AD), mild cognitive impairment (MCI), or major depressive disorder (MDD) is challenging. However, impairments in other domains, such as emotion recognition, an aspect of social cognition, might have additional value in distinguishing Dementia-AD from MCI and MDD and hence signal progression of neurodegeneration. We evaluated this in patients visiting a memory clinic. Sixty healthy controls (HC) and 143 first time attendants of an academic hospital memory clinic who were eventually classified as Dementia-AD (n = 45), MCI (n = 47), MDD (n = 27), or No Impairment (NI, n = 24) were included. We assessed group differences in Emotion Recognition (Ekman 60 Faces Test (EFT)) and episodic memory (Dutch Rey Auditory Verbal Learning Test (RAVLT)). With multinomial and binomial regression analysis, we assessed whether EFT was added to RAVLT in distinguishing patient groups. Dementia-AD patients had significantly worse emotion recognition than HC, MCI, MDD, and NI groups, but no other between-group differences were found. Episodic memory was impaired in Dementia-AD and MCI patients. We found no memory impairments in the MDD and NI groups. Emotion recognition in addition to episodic memory was significantly better in predicting group membership than episodic memory alone. In conclusion, emotion recognition measurement had added value for differentiation between patients first visiting memory clinics, in particular in distinguishing Dementia-AD from MCI. We recommend the standard inclusion of emotion recognition testing in neuropsychological assessment in memory clinics.