5.4. Atopy of the first degree relatives and depression of the proband
The CHAID technique was used to analyze the effects of maternal, paternal, and siblings’ atopies on a proband’s depression. As Figure 3 shows, in females the strongest risk factor in predicting self-reported lifetime depression was the presence of parental atopy (chi-square test, χ2 = 12.3, df = 1, p < 0.001) (IV: Figure 1). Another statistically significant predictor of depression was a proband’s regular daily smoking (χ2 = 11.9, df = 1, p < 0.001): When either mother or father were atopics, the proportion of cohort members with depression was 12.7%. Among non-smokers, whose parents and who themselves were atopics, the corresponding proportion was 9.0%. Interestingly, if there existed no parental history of atopy, then living in an urban area as well as a city dweller’s own atopy, were both statistically significant predictors of depression.
Figure 4 presents the results for hospital-treated depression in females (IV: Figure 2). Both sibling’s atopy as well as parental atopy were significant predictors for hospital-treated depression.
Figure 3. CHAID flowchart of the statistically significant predictors for self-reported lifetime depression involving female cohort members, the Northern Finland 1966 Birth Cohort study (IV: Figure 1).
Figure 4. CHAID flowchart of the statistically significant predictors for hospital-treated depression involving female cohort members, the Northern Finland 1966 Birth Cohort study (IV: Figure 2).
In the CHAID-analysis, none of the family atopy variables predicted depression among males. However, regular daily smoking among males was associated with both self-reported lifetime depression (χ2 = 4.1, df = 1, p = 0.044) and with hospital-treated depression (Fisher’s Exact test, p = 0.009).
In subsequent analyses, the effects of maternal and paternal atopies to a proband’s self-reported lifetime depression were investigated separately (Table 5; IV: Table 2). Logistic regression analyses showed that maternal atopy increased a female proband’s risk of lifetime depression up to 1.9-fold (OR = 1.9, 95% CI 1.1–3.0), while paternal atopy alone did not associate statistically significantly with a proband’s depression. The corresponding risk increased over 4-fold, when a female proband’s atopy was combined with parental (OR = 4.5, 95% CI 1.6–13.2) or maternal (OR = 4.1, 95% CI 1.8–9.4) atopy.
Table 5. Risk of depression associated with family atopy variables and with a cohort member’s own atopy for males and females in the Northern Finland 1966 Birth Cohort (IV: Table 2).
| The variables for own and familial atopy | Males | Females |
| OR (95% CI) | OR (95% CI) | |
| Single risk factor* | ||
| Parental atopy | 1.1 (0.2–4.6) | 2.1 (1.0–4.1) |
| Paternal atopy | 0.7 (0.3–1.7) | 1.6 (0.9–2.9) |
| Maternal atopy | 0.6 (0.3–1.5) | 1.9 (1.1–3.0) |
| Sibling’s atopy | 1.1 (0.6–2.0) | 1.4 (0.9–2.0) |
| Own atopy | 1.0 (0.4–2.0) | 1.6 (1.0–2.8) |
| Combined risk factors** | ||
| Own atopy and parental atopya | Not feasible | 4.5 (1.6–13.2) |
| Own atopy and paternal atopyb | Not feasible | 2.7 (0.9–7.8) |
| Own atopy and maternal atopyc | 1.2 (0.3–4.1) | 4.1 (1.8–9.4) |
| Own atopy and sibling atopyd | 1.0 (0.4–2.6) | 3.2 (1.7–6.1) |
| *Adjusted odds ratios (OR). In addition to family atopy variables, the following confounders were used in the logistic regression analyses: The dwelling place in 1966 (urban/rural), father’s social class in 1966 (classes I–II/classes III–IV/farmers), maternal smoking during pregnancy (yes/no), cohort member’s regular daily smoking at the age of 31 (regular smokers [i.e., smoking on 7 days per week]/occasional smokers [i.e. smoking less than on 7 days per week] and non-smokers), and mother’s parity in 1966 (grand multi-parity i.e. ≥6 deliveries/non-grand-multiparity i.e., 1–5 deliveries). **Reference category: neither own atopy nor parental atopya / paternal atopyb / maternal atopyc / sibling atopyd. | ||