Graduation Year


Document Type




Degree Name

Doctor of Public Health (Dr.PH.)

Degree Granting Department

Epidemiology and Biostatistics

Major Professor

Kathleen O'Rourke, Ph.D.

Committee Member

Kevin Kip, Ph.D.

Committee Member

Wei Wang, Ph.D.

Committee Member

Michael Custer, Dr.PH


perinatal health, Maternal and Child Health, birth outcomes, mental health, war


Introduction. Today women comprise 15% of the U.S. active-duty military, but are often overlooked in research of the Armed Forces. While some of the challenges faced by women are similar to men, they encounter unique stressors related to childcare while deployed, sexual harassment and assault, and gynecological needs. Women are also more likely than men to develop post-traumatic stress disorder (PTSD). Both stress and PTSD have been linked to the development of chronic hypertension and some adverse birth outcomes. We hypothesized that post-9/11/2001 era military women who deployed or who had indicators of PTSD would be at greater risk of developing a hypertensive disorder of pregnancy (HDP) than non-deployed or non-PTSD military women.

Methods. We conducted a retrospective cohort study using a U.S. Department of Defense database comprised of all active-duty women who gave birth to their first, live-born singleton infant using Tricare from January 1, 2004 to December 31, 2008. The database included maternal and infant birth hospitalization records, maternal mental health visits, and post-9/11/2001 deployment information which included Post-Deployment Health Assessment (PDHA) and Reassessment (PDHRA) screening responses. HDP was defined with ICD-9-CM codes in the maternal birth hospitalization record. We evaluated the risk of HDP associated with overall deployment and timing of deployment ending compared to non-deployed women, in addition to cumulative months of deployment. We also conducted Classification Tree Analysis (CART) to determine appropriate cut-points for categorization of deployment variables among mothers who deployed: cumulative weeks of deployment, percent of study time spent deployed, and dwell time between two most recent pre-birth deployments. We explored effect modification by covariates postulated to potentially modify the relationship between deployment history and risk of HDP. New variables were defined and used in multivariable logistic regression models for each deployment measure. Women fit into four PTSD case-definition categories: confirmed (ICD-9-CM diagnosis), probable (possible plus endorsement of “high risk” items on the PDHA), possible (endorsement of ≥3 PTSD items on the PDHA), or none (no PTSD indicators). We compared mothers with PTSD to non-cases using multivariable logistic regression to quantify the risk of HDP, and repeated the analyses using a ≥2 PTSD item endorsement case-definition criteria. All logistic regression models were adjusted for known confounders and important covariates.

Results. There were a total of 36,675 births, 13.4% of mothers experienced at least one HDP, and 35% of mothers deployed. No increased risk of HDP was observed for deployment overall (OR=1.02, 95% CI: 0.95-1.09), but black mothers who deployed were 13% more likely to develop an HDP (OR=1.13, 95% CI: 1.00-1.27). CART revealed an important cut-point for cumulative deployment length of ≥1 year, which was statistically significant among mothers <35 years old. Mothers with ≥1 year cumulative deployment were 17% more likely to have an HDP than mothers deployed <1 year (OR=1.17, 95% CI: 1.01-1.36). The prevalence of confirmed PTSD was 1.6% in the overall cohort. The prevalence of any PTSD among deployed mothers who completed a PDHA was 6.2%. Overall, PTSD was not significantly associated with HDP except among probable PTSD cases using the ≥2 item criteria (OR=1.30, 95% CI: 1.01-1.67) and among confirmed PTSD mothers identifying as “other” race (OR=6.62, 95% CI: 1.72-25.47).

Conclusion. Results are suggestive of an elevated risk of HDP among the military population among women who deployed for a year or longer and for black mothers. Although PTSD did not clearly confer additional risk in the overall cohort, there is evidence to support further research using more thorough screening especially across racial/ethnic groups. Our study likely underestimated PTSD and possibly attenuated results since individuals may purposely report inaccurately on the PDHA in order to go home sooner after deployment. Future studies should include information related to deployment-specific experiences and screen all participants for evidence of PTSD.

Included in

Epidemiology Commons