Comparability of two tests of olfactory functioning

Abstract

In the first of two comparisons, 50 patients who came to the Connecticut Chemosensory Clinical Research Center (CCCRC) with various chemosensory complaints were evaluated by a dual threshold/odor identification test administered in the CCCRC and by the self-administered University of Pennsylvania Smell Identification Test (UPSIT). Agreement between tests proved high, with a correlation coefficient of 0.92. The UPSIT correlated better with the odor identification component than with the threshold component of the CCCRC test. This disparity stemmed largely from limitations in the reliability of the threshold component. An increase in the stringency of the criterion for determination of threshold increased the reliability of this component and, in a second comparison with a total of 58 patients and controls, the correlation between the UPSIT and CCCRC test increased. With the more stringent criterion, the correlation equalled 0.96.