Relationship between smell identification testing and the neuropsychological assessment of dementia in community-dwelling adults

The objective of this research was to investigate the relationship of the University of Pennsylvania Smell Identification Test (UPSIT) with neuropsychological tests and identify the utility of the UPSIT in detecting cognitive impairment. This research was an archival study of neuropsychological test results obtained from 70 clients (30 male/40 female) of a community-based memory clinic. The sample had an average age of 69.7 (SD = 9.7) and education of 14.6 (SD = 2.8) years. Hypotheses were tested using correlations, receiver operating characteristic (ROC) curves, and logistic regression. The UPSIT showed significant, weak to moderate correlations with neuropsychological tests. The UPSIT raw score correlated significantly with all but one cognitive ability domain. The UPSIT T-score was significantly correlated with all cognitive domains. Obtained areas under the ROC curve (AUC) for the UPSIT ranged from .60 to .87. The AUCs of the UPSIT did not differ significantly from verbal semantic fluency tests, but the Repeatable Battery for the Assessment of Neuropsychological Status Total Scale and Delayed Memory index tended to produce larger AUCs than the UPSIT. Results from step-wise logistic regressions suggest that the UPSIT raw score provides unique information beyond its relationship to age. Olfaction relates broadly to cognitive ability and may be sensitive to early symptoms of cognitive decline. Further research is needed to explore the relationships between smell identification tests and neuropsychological assessment.