Logistic regression was applied to determine the risk of allergic rhinitis depending on markers of estrogen levels: digit ratio, age at menarche, menstrual cycle regularity, and WHR. These factors were considered as categorical variables. Secondly, the analyses were performed with 2D:4D, age at menarche, BMI, and WHR as continuous variables. Relationships between allergic rhinitis and markers of estrogen level were tested by analysis of variance. Due to the relationship between socio-economic factors and the prevalence of allergies, maturity rate, and anthropological parameters, socio-economic status was included in each analysis as a potential confounder. A p-values <0.05 was used to define statistically significant differences.

Medically diagnosed allergy was reported by 205 women (26.7% of all subjects), of which 168 women were diagnosed with allergic rhinitis (21.9% of all subjects). The results of logistic regression are shown in Table 2. In women with a high digit ratio, indicating exposure to a high estrogen level in the prenatal period, the risk of allergic rhinitis was almost twice as high as that in women with middle values of the ratio (p<0.05). The difference was greater for the right than left hand. A significantly higher risk of allergic rhinitis was also observed in women who experienced first menstruation at younger age. No differences in the odds ratio of allergic rhinitis due to obesity, abdominal obesity, or irregular menstrual cycles were found.