Olfactory dysfunction can influence nutritional intake, the detection of environmental hazards, and quality of life. Prior research has found discordance between subjective and objective measures of olfaction. In people living with HIV (PLWH), olfactory dysfunction is widely reported; however, few studies have examined concordance between subjective olfactory self-ratings and performance on an objective psychophysical measure of olfaction and associated factors in men living with HIV (MLWH).

MLWH (n=51, mean age=54 years, 66.7% Black) completed two subjective olfaction ratings (two 5-point Likert scales), the Smell Identification Test (SIT), cognitive measures (HVLT-R, TMT), and self-report questionnaires assessing smell habits, mood, cognitive failures, and quality of life. Participants were categorized into one of four groups: true positives (TP; impaired subjective olfaction and objective olfaction dysfunction), false negatives (FN; intact subjective olfaction and objective olfaction dysfunction), false positives (FP; impaired subjective olfaction and objective normosmia), and true negatives (TN; intact subjective olfaction and normosmia). Established formulas were used to calculate the sensitivity and specificity of subjective olfaction, and t-tests and ANOVA were used to examine potential demographic, clinical, and cognitive factors contributing to discordance between subjective and objective olfaction dysfunction.

Across both subjective self-report items, 35.3% reported olfactory dysfunction, whereas 60.8% had objective olfaction dysfunction on the SIT (score < 33). Black MLWH had significantly higher rates of subjective (Black 41.2% vs. White 35.3%) and objective (Black 73.5% vs. White 35.3%) olfactory dysfunction (X2(1)=9.22, p=.002). We found discordance between subjective and objective olfaction measures, with 29.4% of the sample having objective olfaction dysfunction and not recognizing it (FN). In comparison, 3.9% with self-rated olfaction impairment had normal objective olfaction scores (FP). Additionally, there was concordance in subjective self-reports compared with objective olfaction, with 35.3% correctly identifying normal olfaction (TN) and 31.4% correctly identifying olfactory dysfunction (TP). Those unaware of olfaction dysfunction (FN) reported using less scented products in daily life on the Smell Habits Questionnaire. Although the FN group had faster TMT scores, these findings were no longer significant after the removal of three outliers in the TP group (e.g., time to complete greater than 350 seconds).

Our findings cohere with work in healthy older adults, traumatic brain injury, and Parkinson’s disease, documenting that subjective olfaction may inadequately capture the full range of a person’s olfactory status. We extend these findings to a sample of MLWH, in which discordance rates ranged from 35-61% for subjective and objective olfactory dysfunction. Unawareness of olfactory dysfunction in MLWH was associated with less daily smell habits and paradoxically faster TMT performance. A higher number of smell habits in the TP group indicate that more frequent odor exposure may increase sensitivity to olfactory declines. Future studies with larger samples will be helpful in understanding the full nature of these relationships. Lastly, given that one-third of the sample had discordance between subjective and objective olfaction, objective olfaction measures may be useful to consider in the neuropsychological assessment and standard clinical care for PLWH.