Discordance Between Subjective and Objective Measures of Smell and Taste in US Adults

Abstract

Objective Examine the rates and factors associated with under- and overreporting of subjective changes in smell or taste as compared with objective measures. Study Design Cross-sectional analysis. Setting National Health and Nutrition Examination Survey (2013-2014). Methods We examined participants ≥40 years old who completed subjective questionnaires (smell, n = 3510; taste, n = 3089), validated objective 8-odor pocket smell tests, and NaCl/quinine taste tests. Over- and underreporting was determined by the difference in subjective and objective results. Univariate and multivariate logistic regression analyses incorporated sampling weights. Results A majority of participants correctly classified impairment: smell (73.7%; 95% CI, 71.2%-76.1%) and taste (78.3%; 95% CI, 75.6%-80.7%). Age ≥65 years (odds ratio, 2.23; P = .001) was associated with underreporting impairment, and persistent cold symptoms (odds ratio, 2.15; P = .001) were associated with overreporting smell impairment. Smoke, onion, and natural gas scents were incorrectly identified more frequently by individuals aged ≥65 years after Bonferroni correction. No factors were associated with under- and overreporting taste impairment. Conclusion Although the concordance rate between subjective and objective assessment of smell and taste impairment remains high, we found that older age was associated with incorrect report of impairment. This suggests that the subjective perception of smell varies across demographical and clinical factors, and it is important to not overlook such factors in clinical practice. Potentially using a simplified odor assessment regularly in the clinical setting may aid in early detection and intervention.