Utility of a three-item smell identification test in detecting olfactory dysfunction

Objective: Physicians rarely assess smell function, largely because of time considerations. Therefore, there is clinical need for very brief cranial nerve I screening tests. Although a few such tests exist, none have been adequately validated. The goal of this study was to empirically assess the utility of a three-item microencapsulated odor identification test in detecting olfactory dysfunction. Setting. Smell and taste center at a university medical center. Methods: The test was administered to 224 consecutive patients (98 men and 126 women ranging in age from 15-88 years). As part of their overall assessment, the well-validated 40-item University of Pennsylvania Smell Identification Test (UPSIT) was also administered. Sensitivity, specificity, and both negative and positive predictive values of the three-item test were established relative to UPSIT dysfunction categories. Test-retest reliability was determined in a subset of 39 patients. Results: The three-item test was abnormal in 99% (67/68) of patients with anosmia, 85% (35/41) of those with severe microsmia, 76% (31/41) of those with moderate microsmia, and 50% (17/34) of those with mild microsmia. Of the 40 normosmic patients, 62.5% (25/40) correctly identified all odors, 25% (10/40) two odors, and 12.5% (5/40) one odor. None of the normosmic patients missed all three items. Using a cut-off score of 2, the test's sensitivity and specificity were 99% and 40%, respectively, for detecting total anosmia. The corresponding negative and positive predictive values were 98% and 43%. For detecting anosmia and severe microsmia, these values were 93%, 45%, 88%, and 63%. For detecting any olfactory pathology, they were 82%, 63%, 42%, and 91%. The test-retest reliability was 0.87. Conclusion: The brief three-item test used in this study was found to be highly sensitive in identifying olfactory loss in patients with chemosensory complaints, particularly those with severe dysfunction. Although only moderately specific, its high reliability and negative predictive value suggests it may be an appropriate screening test for olfactory loss.