Date of Completion


Embargo Period



breastfeeding, postpartum depressive symptoms, Latin American women, public healh, policy, Ecuador, Puerto Rico, Mexico

Major Advisor

E. Carol Polifroni

Associate Advisor

Cheryl Tatano Beck

Associate Advisor

Heather Evans

Field of Study



Doctor of Philosophy

Open Access

Open Access


The Relationship Between Breastfeeding and Postpartum Depressive Symptoms in Women

from Ecuador, Mexico, and Puerto Rico

Jan Weingrad Smith, PhD

University of Connecticut, 2012

This study used a secondary data analysis to describe the relationship between breastfeeding and the development of postpartum depressive symptoms in a low-income cohort of women from Ecuador, Mexico, and Puerto Rico in the United States. It also explored the risk of developing postpartum depressive symptoms in low-income women from these communities regardless of their decision to breastfeed. Data were collected between 2006 and 2012 at a Federally Qualified Community Health Center Prenatal Center in New England. A sample of 128 women had complete data for analysis and self-identified as Ecuadorian, Mexican, or Puerto Rican. Because the number of Ecuadorian women (12) was statistically too small to be included, the final sample was 116 women from Mexico and Puerto Rico. The analysis examined the possibility that the relationship between breastfeeding and postpartum depressive symptoms varied according to the nationality of the mother. The data suggest that Puerto Rican women are 1.8 (OR 1.758, 95% CI .680-4.543) times as likely to have PPD symptoms (X2 =.241, df = 1, p = .987) compared to women from Mexico. The odds ratio for Mexican women who breastfed their infant and experienced postpartum depressive symptoms was 2.000 (95%CI .219-18.232, X2= .390, df=1, p=.987). While not statistically significant, the data seem to imply that Puerto Rican women have a higher probability of having PPD symptoms regardless of their decision to breastfeed. Due to incomplete data, some women in the database could not be included in the study, creating a small sample size for analysis. The power analysis proposed for the study initially called for 200 cases. Only 116 cases had complete data entry, which limited the sample and did not reach the required power for statistical significance. This information will add to the specificity of the evidence that can be used to develop targeted programs to increase breastfeeding rates and decrease rates of postpartum depression in these groups.