Table 6.
Multinomial regression between types of child maltreatment and types of eating disorders in the total sample
| Type of lifetime eating disorder | |||||
|---|---|---|---|---|---|
| Child maltreatment type | Estimate | No eating disorder | Anorexia nervosa | Bulimia nervosa | Binge eating disorder |
| Harsh physical punishment | % (n) | 17.8 (6468) | 25.8 (72) | 30.6 (24) | 35.1 (100) |
| AOR (95% CI) | 1.00 | 1.64a (1.17, 2.30)** | 2.08a (1.04, 4.18)* | 2.52a (1.86, 3.40)*** | |
| Physical abuse | % (n) | 17.3 (6290) | 28.0 (75) | 23.0 (21) | 37.0 (103) |
| AOR (95% CI) | 1.00 | 1.91a (1.33, 2.74)*** | 1.46a (0.79, 2.72) | 2.83a (2.12, 3.79)*** | |
| Sexual abuse | % (n) | 11.0 (4113) | 26.3 (64) | 28.2 (22) | 29.7 (83) |
| AOR (95% CI) | 1.00 | 2.11a (1.46, 3.04)*** | 2.43a (1.23, 4.80)* | 3.05a (2.20, 4.24)*** | |
| Emotional abuse | % (n) | 10.0 (3696) | 21.6 (59) | 26.5 (24) | 29.8 (87) |
| AOR (95% CI) | 1.00 | 2.25a (1.53, 3.33)*** | 2.97a (1.59, 5.56)*** | 3.67a (2.61, 5.15)*** | |
| Emotional neglect | % (n) | 9.9 (3788) | 12.3 (38) | 27.4 (23) | 18.4 (49) |
| AOR (95% CI) | 1.00 | 1.18a (0.81, 1.71) | 3.18b (1.66, 6.09)*** | 1.99b (1.43, 2.76)*** | |
| Physical neglect | % (n) | 27.5 (9904) | 35.6 (93) | 38.5 (28) | 51.5 (146) |
| AOR (95% CI) | 1.00 | 1.60a (1.21, 2.11)*** | 1.79a,b (0.95, 3.40) | 2.92b (2.20, 3.87)*** | |
| Exposure to intimate partner violence | % (n) | 17.7 (6505) | 28.9 (72) | 32.1 (22) | 38.6 (111) |
| AOR (95% CI) | 1.00 | 1.76a (1.23, 2.53)** | 2.06a (1.01, 4.20)* | 2.84a (2.13, 3.79)*** | |
| Any child maltreatment | % (n) | 48.6 (17428) | 62.4 (157) | 65.6 (46) | 73.6 (210) |
| AOR (95% CI) | 1.00 | 1.78a (1.28, 2.48)*** | 2.05a,b (1.001, 4.19)* | 2.98b (2.20, 4.03)*** | |
| Any family dysfunction | % (n) | 29.0 (10428) | 42.8 (107) | 54.4 (35) | 55.6 (151) |
| AOR (95% CI) | 1.00 | 1.72a (1.29, 2.28)*** | 2.75a,b (1.50, 5.04)*** | 2.98b (2.16, 4.11)*** | |
Note. All n values were unweighted, and all percentages were weighted. Individuals with more than one eating disorder were removed from analyses. AOR = adjusted for sex. Estimates with different superscripts are significantly different from one another at p ≤ .05.
*p ≤ .05; **p ≤ .01; ***p ≤ .001.