Skip to main content
. 2021 Apr 4;35(6):830–839. doi: 10.1177/19458924211004800

Table 4.

Multivariable Logistic Regression Models of Outcome Ageusia (N = 438, Events = 219)*.


Full Model

Final Model
Adjusted OR (95% CI) P-Value Adjusted OR (95% CI) P-Value
Age (per 1-year increase) 0.98 (0.97, 0.99) 0.004 0.98 (0.97, 0.99) 0.005
Male (reference = female) 0.56 (0.38, 0.83) 0.004 0.56 (0.38, 0.82) 0.003
Black (reference = white) 1.00 (0.46, 2.18) 1.00
Latino (reference = white) 1.06 (0.48, 2.34) 0.88
Other/unknown race (reference = white) 1.19 (0.50, 2.81) 0.69
Eosinophil count at time of SARS-CoV-2 test (per 0.1 K/μL increase in eosinophils) 0.09 (0.01, 0.95) 0.046 0.10 (0.01, 0.97) 0.047
History of hypertension (reference = no) 1.08 (0.66, 1.77)
History of hypercholesterolemia (reference = no) 1.12 (0.68, 1.85)

*Hosmer-Lemeshow goodness of fit test: p = 0.42 for full model; p = 0.09 for final model.

Full model includes past medical history and laboratory variables with sufficient available data and p < .05 for association with anosmia or ageusia at the bivariate level, as well as demographic characteristics. Final model includes variables with p-value < .05 after applying stepwise backward selection.p<0.05 are shown in bold and italic format.