Results from logistic regression analyses (odds ratios [with 95% CI] and regression coefficients) for two models with demographic variables, variables with respect to clinical history and specific IgE.
| Asthmaa | Asthmab | |||
|---|---|---|---|---|
| 3–4 years of age at inclusion | 2.0 (1.0 to 4.0) | 0.7 | 1.7 (0.8 to 3.7) | 0.5 |
| Positive family history of allergy for pollen | 3.3 (1.4 to 7.5) | 1.2 | 2.4 (0.9 to 6.1) | 0.9 |
| Wheeze at inclusion | 7.2 (2.6 to 20.1) | 2.0 | 17.9 (5.0 to 63.5) | 2.9 |
| Specific IgE (0.5 IU/ml) | 18.7 (5.4 to 64.3) | 2.9 |
All variables included in the model were present (= 1) or absent (= 0) at inclusion. Dependent variable is asthma, children without asthma used as references.
↵a model with demographic variables and clinical history (n = 123), intercept of the model: −2.7. Model 1: score = 0.7*(3–4 years of age) + 1.2*familial allergy for pollen + 2.0*wheeze.
↵b as a but model includes specific IgE, intercept of the model: −4.3. Model 2: score = 0.5*(3–4 years of age) + 0.9*familial allergy for pollen + 2.9*wheeze + 2.9*specific IgE. For both scores the corresponding probabilities of developing asthma can be calculated from Pr = 1/(1+e-(score+intercept)). IgE = immunoglobulin E.