Skip to main content
. 2024 May 9;19(5):e0302579. doi: 10.1371/journal.pone.0302579

Table 4. Determinants of breakthrough infections among vaccinated patients over a 6-month period using univariate analysis and multivariable logistic regression models.

Univariate analysis Multivariable logistic regression models
Model 1 i Model 2 ii
Categorical variables % P adj. OR 95%CI P adj. OR 95%CI P
Sex
    Female1 6.81
    Male 2.15 0.16 0.35 0.05–1.63 0.21 0.30 0.04–1.37 0.15
Smoking
    No1 3.57
    Yes 5.80 0.48
Alcohol consumption
    No1 5.09
    Yes 0.00 0.60
Blood type & Rhesus factor
    A-1 0.00
    A+ 3.92
    AB+ 10.0
    B- 0.00
    B+ 11.5
    O- 0.00
    O+ 3.33 0.59
Chronic disease
    No1 2.91
    Yes 6.41 0.29 2.48 0.55–13.1 0.25
Vaccine type
    Pfizer-BioNTech (BNT162b2)1 2.29
    AstraZeneca 10.0 0.04 5.35 1.22–27.9 0.03 4.75 1.11–24.2 0.04
Side effects after the first shot
    No1 4.00
    Yes 4.58 1.00
Continuous variables Breakthrough infection No breakthrough infection P adj. OR 95%CI P adj. OR 95%CI P
Age (years) 51.4 52.2 0.86
Body Mass Index (kg/m 2 ) 28.1 27.1 0.74

Determinants of breakthrough infections were predicted using univariate (t-test and the chi-squared test for continuous and categorical variables, respectively) and multivariable analysis (logistic regression models). iThe variables tested by univariate analysis that had a P value (P) < 0.40 were included in Model 1 (multivariable logistic regression analysis). iiIn Model 2, a backward logistic regression model was created including only complete cases. 1Reference group. *During the study period. Individuals who had received a Sputnik V vaccination were removed from the analysis because of their low total number.