Table 3.
Dose–response relationship between childhood maltreatment and medical illnesses by mood disorder status
| Bipolar disorder group (n = 72) | Unipolar depression group (n = 248) | Control group (n = 354) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Adjusted ORa | 95% CI | P | Adjusted ORa | 95% CI | P | Adjusted ORa | 95% CI | P | |
| At least one medical illness | |||||||||
| Number of childhood maltreatment histories | |||||||||
| None (reference) | – | – | – | – | – | – | – | – | – |
| One | 4.71 | 1.30–17.09 | 0.019 | 0.68 | 0.34–1.35 | 0.270 | 0.81 | 0.36–1.86 | 0.622 |
| Two or more | 5.45 | 1.30–22.85 | 0.020 | 1.71 | 0.94–3.12 | 0.078 | 0.50 | 0.15–1.69 | 0.266 |
| Adjusted for age of mood disorder onset and illness duration | |||||||||
| None (reference) | – | – | – | – | – | – | – | – | – |
| One | 5.24 | 1.19–22.89 | 0.028 | 0.75 | 0.34–1.63 | 0.463 | – | – | – |
| Two or more | 5.07 | 1.11–23.18 | 0.036 | 1.12 | 0.54–2.32 | 0.759 | – | – | – |
| Number of medical illnesses | |||||||||
| Number of childhood maltreatment histories | |||||||||
| None (reference) | – | – | – | – | – | – | – | – | – |
| One | 4.26 | 1.31–13.84 | 0.016 | 0.80 | 0.40–1.61 | 0.528 | 0.77 | 0.34–1.74 | 0.532 |
| Two or more | 5.13 | 1.42–18.50 | 0.012 | 1.71 | 0.98–2.98 | 0.060 | 0.52 | 0.16–1.70 | 0.275 |
| Adjusted for age of mood disorder onset and illness duration | |||||||||
| None (reference) | – | – | – | – | – | – | – | – | – |
| One | 4.52 | 1.24–16.47 | 0.022 | 0.70 | 0.33–1.47 | 0.340 | – | – | – |
| Two or more | 4.10 | 1.05–15.93 | 0.042 | 1.32 | 0.65–2.60 | 0.431 | – | – | – |
OR, odds ratio derived from binary logistic regression for ‘at least one medical illness’ and from ordinal logistic regression for ‘number of medical illnesses’.
a.
Adjusted for the effects of gender and age.