Abuse in childhood and adolescence as a predictor of type 2 diabetes in adult women.

BACKGROUND Although child abuse is associated with obesity, it is not known whether early abuse increases risk of type 2 diabetes. PURPOSE To investigate associations of child and adolescent abuse with adult diabetes. METHODS Proportional hazards models were used to examine associations of lifetime abuse reported in 2001 with risk of diabetes from 1989 to 2005 among 67,853 women in the Nurses' Health Study II. Data were analyzed in 2009. RESULTS Child or teen physical abuse was reported by 54% and sexual abuse by 34% of participants. Models were adjusted for age, race, body type at age 5 years, and parental education and history of diabetes. Compared to women who reported no physical abuse, the hazards ratio (HR) was 1.03 (95% CI=0.91, 1.17) for mild physical abuse; 1.26 (1.14, 1.40) for moderate physical abuse; and 1.54 (1.34, 1.77) for severe physical abuse. Compared with women reporting no sexual abuse in childhood or adolescence, the HR was 1.16 (95% CI=1.05, 1.29) for unwanted sexual touching; 1.34 (1.13, 1.59) for one episode of forced sexual activity; and 1.69 (1.45, 1.97) for repeated forced sex. Adult BMI accounted for 60% (95% CI=32%, 87%) of the association of child and adolescent physical abuse and 64% (95% CI=38%, 91%) of the association of sexual abuse with diabetes. CONCLUSIONS Moderate to severe physical and sexual abuse in childhood and adolescence have dose-response associations with risk of type 2 diabetes among adult women. This excess risk is partially explained by the higher BMI of women with a history of early abuse.