• Disclosures: Portions of this manuscript were presented at the Third International Conference on Interpersonal Psychotherapy (March 2009), New York, New York, and at the Society for Social Work and Research 13th Annual Conference (January 2009), New Orleans, LA.

  • Conflict of interest: Dr. Grote, Dr. Spieker, Ms. Lohr, and Ms. Geibel report no competing interests. Dr. Swartz has received CME honoraria from Servier, Astra Zeneca, and Sanofi. She receives royalties from UpToDate. Dr. Frank serves on an advisory board and has received honoraria from Servier, International and receives royalties from Guilford Press and the American Psychological Association Press. Dr. Katon serves on an advisory board at Lilly and has received honoraria from Lilly, Forest, and Pfizer.

  • The authors disclose the following financial relationships within the past 3 years: Contract grant sponsor: National Institute of Mental Health; Contract grant number: K23 MH67595; Contract grant sponsor: Staunton Farm Foundation; Contract grant sponsor: NIH/NCRR/GCRC; Contract grant number: MO1-RR000056.

Correspondence to: Nancy K. Grote, School of Social Work, University of Washington, Campus Box 354900, 4101 15th Ave. East, Seattle, WA 98105. E-mail:



Childhood abuse and neglect have been linked with increased risks of adverse mental health outcomes in adulthood and may moderate or predict response to depression treatment. In a small randomized controlled trial treating depression in a diverse sample of nontreatment-seeking, pregnant, low-income women, we hypothesized that childhood trauma exposure would moderate changes in symptoms and functioning over time for women assigned to usual care (UC), but not to brief interpersonal psychotherapy (IPT-B) followed by maintenance IPT. Second, we predicted that trauma exposure would be negatively associated with treatment response over time and at the two follow-up time points for women within UC, but not for those within IPT-B who were expected to show remission in depression severity and other outcomes, regardless of trauma exposure.


Fifty-three pregnant low-income women were randomly assigned to IPT-B (n = 25) or UC (n = 28). Inclusion criteria included ≥18 years, >12 on the Edinburgh Postnatal Depression Scale, 10–32 weeks gestation, English speaking, and access to a phone. Participants were evaluated for childhood trauma, depressive symptoms/diagnoses, anxiety symptoms, social functioning, and interpersonal problems.


Regression and mixed effects repeated measures analyses revealed that trauma exposure did not moderate changes in symptoms and functioning over time for women in UC versus IPT-B. Analyses of covariance showed that within the IPT-B group, women with more versus less trauma exposure had greater depression severity and poorer outcomes at 3-month postbaseline. At 6-month postpartum, they had outcomes indicating remission in depression and functioning, but also had more residual depressive symptoms than those with less trauma exposure.


Childhood trauma did not predict poorer outcomes in the IPT-B group at 6-month postpartum, as it did at 3-month postbaseline, suggesting that IPT including maintenance sessions is a reasonable approach to treating depression in this population. Since women with more trauma exposure had more residual depressive symptoms at 6-month postpartum, they might require longer maintenance treatment to prevent depressive relapse.