Autistic Traits: A Highly Prevalent Risk Indicator for Childhood Abuse, Posttraumatic Stress, and Depression

Friday, May 16, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
A. L. Roberts, 401 Park Drive, Harvard School of Public Health, Boston, MA
Background:  Women with autistic traits (the “broad autism phenotype”) may be at elevated risk for abuse in childhood. Children who noticeable differ from others are at increased risk of being targeted for abuse in childhood, as are children with communication impairments and children with disabilities.  As depression and posttraumatic stress disorder (PTSD) are common sequelae of childhood abuse, women with autistic traits may be at elevated risk of these outcomes due to increased exposure to childhood abuse.

Objectives:  To examine the association of autistic traits with childhood physical, emotional and sexual abuse, and depression and PTSD symptoms.  To further examine the extent to which higher exposure to childhood abuse and lower social support explain possible associations between autistic traits and depression and PTSD.

Methods:  We examined prevalence of childhood abuse, PTSD symptoms, and depression by autistic traits.  We estimated odds ratios (OR) of childhood sexual abuse (Gallup Poll) and the highest quartile of childhood physical/emotional abuse (Conflict Tactics Scale) associated with quintiles of autistic traits in separate models.  We modeled PTSD symptoms and lifetime depression as the dependent variables in separate models with autistic traits as the independent variable. To ascertain the extent to which childhood abuse and social support might account for possibly elevated risk of PTSD and depression in women with higher levels of autistic traits, we separately added childhood abuse and social support to models.  In sensitivity analyses we restricted to women without a child with ASD (control mothers).

Results:  Women in the highest versus lowest quintile of autistic traits were more likely to have been sexually abused (lowest quintile=26.7%; highest quintile=40.1%) and to have experienced physical/emotional abuse (lowest quintile=14.3%, highest quintile=23.9%).  PTSD symptoms and lifetime depression were also more prevalent in the highest versus lowest quintile of autistic traits (high PTSD symptoms, lowest quintile=4.5%, highest quintile=10.7%; depression, lowest quintile=7.9%, highest quintile=29.2%). Adjusted for covariates, women in the highest versus lowest quintile of autistic traits were at significantly elevated odds of sexual abuse (OR=1.7, 95% confidence interval (CI)=1.2, 2.6), and women in the top two quintiles of autistic traits were at elevated odds of physical/emotional abuse.  Odds of PTSD were elevated in women in the top three quintiles of autistic traits compared with the reference group (OR range=1.4 to 2.0).  Odds of depression were substantially higher in all women above the lowest quintile of autistic traits (OR range= 2.4 to 4.5). Adjustment for childhood abuse slightly to moderately attenuated associations with PTSD and depression. Social support did not further mediate these associations.

Conclusions:  We identify for the first time autistic traits as a risk indictor for childhood abuse and PTSD. As we characterized autistic traits by quintiles among women without a child with ASD, our findings suggest that levels of autistic traits associated with elevated risk for depression and PTSD are highly prevalent in the general population.  Risk of depression was approximately two to four and a half times greater in women above the lowest quintile of autistic traits and increased roughly monotonically with autistic traits.