A data science-based analysis of seasonal patterns in outpatient presentations due to olfactory dysfunction

Rhinology. 2020 Apr 1;58(2):151-157. doi: 10.4193/Rhin19.099.


Background: Changes in human olfactory function throughout the year appear to be a common perception due to the seasonal oscillations in some etiologies associated with olfactory loss. However, longitudinal data from large cohorts were rarely analysed for temporal patterns of human olfaction apart from oscillations in specific etiologies of olfactory loss.

Methods: Temporal patterns in the presentation of patients with olfactory disorders to a single centre were investigated as part of a cohort study. The time series analysis performed utilized a power spectrum analysis and an autoregressive integrated moving average (ARIMA) model in order to demonstrate repetitive fluctuations or trends in the monthly number of patients reporting from January 1999 to December 2017. The analyses additionally addressed temporal changes in the causes to which the olfactory disorder was attributed and in the degree of olfactory loss.

Results: A cohort of 7,014 patients was included. Descriptive analysis showed that the presentation of olfactory disorders had seasonal variation, high in March, without a trend. Power spectrum analysis showed a general seasonality of the numbers of patients, without further pattern in the causes or the degree of olfactory dysfunction.

Conclusions: The yearly periodicity in patient presentations at a specialized smell and taste clinic, was not readily attributable to seasonally changing medical causes of olfactory loss such as viral infections. This suggests that in addition to exploring the seasonality of olfactory etiologies, the changes in human olfactory acuity merit further assessments in longitudinal studies.

MeSH terms

  • Cohort Studies
  • Data Science
  • Humans
  • Olfaction Disorders* / epidemiology
  • Outpatients
  • Seasons
  • Smell*