Correlation and agreement of olfactory perception assessed by the Connecticut Chemosensory Clinical Research Center olfactory test and the Brief-Smell Identification Test

Abstract

Introduction: Assessing olfactory perception in olfactory disorders is of utmost importance in therapy management. However, the University of Pennsylvania Smell Identification Test and the Sniffin' Sticks are the only tests validated in Brazil.

Objectives: To evaluate the correlation and agreement between the Chemosensory Clinical Research Center olfactory test and the Brief-Smell Identification Test - University of Pennsylvania Smell Identification Test - in healthy participants and in participants with olfactory disorders based on the results and technical aspects of both tests.

Methods: Fifty participants without olfactory complaints and 50 participants with olfactory disorders who underwent the Chemosensory Clinical Research Center olfactory test and the Brief-Smell Identification Test were included. The following tests were used for statistical analysis: Mann-Whitney U test, Spearman's correlation, intraclass correlation coefficient and Bland-Altman plot. An alpha error (significance level) of 0.05 was considered in the statistical analysis.

Results: Both tests were effective in distinguishing the groups without the presence of overlapping values ​​for the measured markers. Additionally, there was a strong correlation between Spearman's correlation and intraclass correlation coefficient between the tests and for both nostrils. However, the correlations were lower when the groups were individually evaluated. The Bland-Altman plot showed no bias when all participants were simultaneously evaluated.

Conclusions: The tests to assess olfactory perception presented a high level of agreement. In our sample, we could infer that the Connecticut Chemosensory Clinical Research Center olfactory test is similar to the Brief-Smell Identification Test and can be used in the routine diagnosis of patients with complaints of olfactory disorders, considering the advantage of its low cost.

Keywords: B-SIT; Connecticut; Olfaction; Olfactory perception; Smell; UPSIT.

Figure 1

Preliminary test and identification test analysis for the Connecticut Chemosensory Clinical Research Center olfactory test. In the identification test, the first bottle is the Ammonia test (positive control).

Figure 2

Bland Altman plot for the z-score results of the Brief-Smell Identification Test and Connecticut Chemosensory Clinical Research Center olfactory test. (A) Analysis of all participants comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (right nostril) with the result of the Brief-Smell Identification Test. (B) Analysis of all participants comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (left nostril) with the result of the Brief-Smell Identification Test. (C) Analysis of participants with olfactory disorders comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (right nostril) with the result of the Brief-Smell Identification Test. (D) Analysis of participants with olfactory disorders comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (left nostril) with the result of the Brief-Smell Identification Test. (E) Analysis of healthy participants comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (right nostril) with the result of the Brief-Smell Identification Test. (F) Analysis of healthy participants comparing the result bias of the Connecticut Chemosensory Clinical Research Center olfactory test (left nostril) with the result of the Brief-Smell Identification Test. The data are presented by the z-score of each marker. The figures show: line of equality (difference = 0); 95% Confidence Interval (CI) of mean difference; 95% CI of limits of agreement.

Figure 3

Representation of the participants with olfactory disorders and healthy participants considering the olfactory classification for Connecticut Chemosensory Clinical Research Center olfactory test and brief-smell identification test. The brief-smell identification test was administered by a physician if the participant had difficulties performing the test and the test consisted of a kit with four booklets with specific microencapsulated fragrance labels to be scrapped. The participant identified the perceived substance/odor from four alternatives. The final result summed the correct responses, and the test was considered normal if the participant responded nine or more correct items or as hyposmia in cases of correct smell identification for less than nine items.