Table 2. Association between severe eczema and comorbid atopic disease.
| Comorbid atopic disease | Mild or moderate eczema | Severe eczema | P-value# | ||
|---|---|---|---|---|---|
| Freq | Percent (95% CI) | Freq | Percent (95% CI) | ||
| Asthma | |||||
| Ever asthma | 2226 | 24.3 (22.3, 26.4) | 201 | 36.9 (28.5, 45.2) | 0.002 |
| Asthma within the past 12 months | 1705 | 19.0 (17.0, 20.9) | 169 | 32.2 (24.1, 40.2) | 0.0003 |
| Asthma severity | < 0.0001 | ||||
| Mild | 1110 | 64.1 (58.3, 69.9) | 73 | 36.7 (23.9, 49.5) | |
| Moderate | 485 | 30.4 (24.7, 36.1) | 61 | 27.2 (16.2, 38.2) | |
| Severe | 108 | 5.5 (3.4, 7.6) | 35 | 36.1 (20.5, 51.7) | |
| Hay fever | |||||
| Hay fever within the past 12 months | 3252 | 34.1 (31.7, 36.5) | 253 | 39.0 (30.7, 47.3) | 0.26 |
| Hay fever severity | < 0.0001 | ||||
| Mild | 2124 | 61.4 (56.7, 66.0) | 86 | 31.1 (21.6, 40.7) | |
| Moderate | 960 | 33.9 (29.2, 38.6) | 109 | 39.7 (28.9, 51.5) | |
| Severe | 162 | 4.6 (3.3, 5.9) | 57 | 29.1 (18.5, 39.7) | |
| Food allergies | |||||
| Food allergies within the past 12 months | 1436 | 14.1 (12.5, 15.7) | 157 | 27.0 (19.2, 34.7) | < 0.0001 |
| Food allergy severity | < 0.0001 | ||||
| Mild | 660 | 47.9 (41.6, 54.2) | 35 | 18.1 (9.1, 27.1) | |
| Moderate | 421 | 28.0 (22.7, 33.3) | 60 | 32.5 (18.9, 46.0) | |
| Severe | 344 | 23.3 (18.2, 28.3) | 61 | 48.6 (31.8, 65.4) | |
Rao-Scott chi-square test
Parental refusal to answer a particular question or response of “don't know” occurred for the questions pertaining to eczema severity in 22 (0.2%), ever asthma in 104 (0.1%), current asthma in 107 (0.2%), asthma severity in 13 (0.3%), current hay fever in 154 (0.2%), hay fever severity in 49 (0.3%), current food allergy in 120 (0.2%) and food allergy severity in 30 (0.6%), respectively.
Note: The proportional odds assumption was not met (Score test, P<0.01). Therefore, ordinal logistic regression was not used. Rather, all analyses of severity used binary logistic regression on dichotomized severe vs. mild-moderate disease. However, similar results were obtained when severity was dichotomized moderate-severe vs. mild.