Dissertations, Theses, and Capstone Projects

Date of Degree


Document Type


Degree Name



Public Health


Lorna E. Thorpe

Subject Categories



Background: Perinatal depression is a significant public health issue that affects women and their families. Studies conducted outside of the United States (US) have found a higher prevalence of perinatal depression in immigrant compared to native-born women. US studies have been less consistent but have relied on convenience samples and lacked appropriate comparison groups.

Objectives: To characterize the relationship between migration to the US and risk for perinatal depressive symptomatology and to examine the role of the post-migration psychosocial environment on the occurrence of perinatal depressive symptoms.

Methods: The dissertation used two data sources, the NYC Pregnancy Risk Assessment Monitoring System (PRAMS) and the Centering Pregnancy Plus (CPP) Project. In cross-sectional analysis of PRAMS data (2009¬¬-2010) using log-binomial regression, we assessed the association between nativity and early postpartum depressive symptomatology, including by duration and age of entry into US for immigrant women. Psychosocial mediators were tested with joint tests of significance. In longitudinal analysis using growth mixture modeling and multinomial logistic regression, we characterized perinatal depressive symptom trajectories among adolescent women and examined the effects of generational status and acculturation strategy on trajectories, testing also for psychosocial mediators.

Results: Prevalence of postpartum depressive symptoms was comparable between immigrant and US-born women (aPR=1.08, 95% CI 0.74-1.58), but varied by race/ethnicity, with non-Hispanic White immigrant women at elevated risk compared to their US-born counterparts. There was a slightly elevated, non-significant risk of postpartum depressive symptoms among immigrant women with greater compared to less exposure to the US, and partner stress partially mediated this relationship. Among adolescent women, we identified three distinct trajectories of depressive symptoms: `stable no/low' (58%), `moderate declining' (32%), and `high stable' (11%). Compared to second or greater generation women, first generation women had a 69% lower odds of being classified in the `high stable' group versus the `stable no/low' group (aOR=0.31, 95% CI 0.17-0.57). Social support partially mediated the association between immigrant generation and chronically high symptomatology.

Conclusions: Routine screening and referral to culturally and age appropriate support/treatment might be offered to immigrant and second generation adolescent women as well as non-Hispanic White immigrant women of all ages.

Included in

Epidemiology Commons