Background: Olfactory disturbances including anosmia and parosmia are unique features of SARS-CoV-2 infection with high positive predictive value. Change in olfaction appears to be the consequence of multiple parallel processes including conductive and obstructive anosmia, destruction of nasal cilia and olfactory sensory neurons, and retrograde destruction of higher-order olfactory neural processes. The global pooled estimate for olfactory disturbance among patients with a positive RT-PCR is approximately 40%. To date, there has been no investigation of regional differences in the prevalence of olfactory disturbance in SARS-CoV-2 infection, and no analysis of this sequelae in the southern US.
Methods: A retrospective cohort analysis was performed using the Informatics for Integrating Biology and the Bedside data tool at University of Florida Shands Hospital (i2b2). Outcomes of interest were “SARS coronavirus 2 RNA” and “R43.0 anosmia” OR “R43.1 parosmia”. Covariates were gender (female, male, not recorded, unknown), race (American Indian, Black Hispanic, Black or African American, Hispanic, Indian, Native Hawaiian, White, White Hispanic, Multiracial, and other), and age (0-9, 10-17, 18-34, 35-44, 45-54, 55-64, 65-74, 75-84, 85+).
Results: 8572 patients with COVID-19 were identified, of which 229 (2.8%) had olfactory disturbance. Among patients with olfactory disturbance, 157 (68.9%) were female and the mean age was 31.7 years (95% CI: 17.8- 45.6). Less than 3 patients were American Indian, Black Hispanic, Hispanic, Indian, Native Hawaiian, White Hispanic, or Multiracial. 62 (27.1%) were Black or African American, 141 (61.6%) were White, and 16 (7.0%) were other. Less than 3 patients were 0-9 years, 12 (5.2%) were 10-17 years, 119 (52%) were 18-34 years, 31 (13.5%) were 35-44 years, 26 (11.4%) were 45-54 years, 18 (7.9%) were 55-64 years, 11 (4.8%) were 65-74 years, 5 (2.2%) were 75-84 years, and less than 3 were 85+ years. Prevalence of anosmia and parosmia was more than double in female patients compared to male patients. R² for age group vs. olfactory disturbance diagnosis was .2382.
Conclusion: In this study we found the prevalence of anosmia and parosmia among patients with a positive RT-PCR at University of Florida Shands was much lower than global pooled estimates. Olfactory disturbance in this study reflected age-dependent anosmia, with patients 75-84 years reflecting the lowest prevalence. Given these findings we conclude that while anosmia is regarded as an early marker of COVID-19 infection, it may not be as reliable of a predictor in elderly populations in the southern United States.