BACKGROUND: Genesis and the prognostic value of olfactory dysfunction (OD) in COVID-19 remain partially described. The objective of our study was to characterize OD during SARS-CoV-2 infection and to examine whether testing of OD may be a useful tool in clinical practice in order to early identify patients with SARS-CoV-2 infection. METHODS: Olfactory function assessment was objectively carried out using the u-Smell-it® test. In a cross-sectional study part, we evaluated this test in a control cohort of SARS-CoV-2 negative tested patients, who attended the University Hospital Frankfurt between May 2021 and March 2022. In a second longitudinal study part, sensitivity and specificity of OD was evaluated as a diagnostic marker of a SARS-CoV-2 infection in Frankfurt am Main, Germany in SARS-CoV-2 infected patients and their close contacts. RESULTS: Among 494 SARS-CoV-2 negative tested patients, OD was detected in 45.7% and was found to be significantly associated with the male gender (p < 0.001), higher age (p < 0.001), cardiovascular and pulmonary comorbidities (p < 0.001; p = 0.03). Among 90 COVID-19 positive patients, OD was found in 65.6% and was significantly associated with male gender and positive smoking status (p = 0.04 each). Prevalence and severity of OD were significantly increased in infections with the Delta variant (B.1.617.2) compared to those with the Omicron variant (BA.1.1.529). Diagnostic sensitivity and specificity of OD for diagnosis of SARS-CoV-2 infection were 69% and 64%, respectively. CONCLUSION: OD is common in COVID-19 negative and positive tested patients with significantly different prevalence rates observed between different variants. Diagnostic accuracy of OD is not high enough to implement olfactory testing as a tool in diagnostic routine to early identify patients with a SARS-CoV-2 infection.