January 2009

Document Type


Degree Name



Dept. of Public Health and Preventive Medicine


Oregon Health & Science University


Background Breastfeeding represents the healthiest form of nutritional intake for infants. Historical research indicates the benefits of breastfeeding include fewer infections, protection against chronic diseases, and financial savings over purchased infant formula. Research aimed at increasing breastfeeding rates has identified possible risk factors for early cessation of breastfeeding, including the maternal intention status of pregnancy. Several studies suggest that unintended pregnancies may be associated with a decreased likelihood of postpartum breastfeeding and nearly half of the pregnancies in the United States each year are unintended. Such studies, however, commonly regard mistimed and unwanted pregnancies as equivalent, finding that women with unintended pregnancies (mistimed and unintended) were less likely to breastfeed than women with intended pregnancies and failing to compare mistimed pregnancies to unwanted. Distinguishing breastfeeding outcomes between mistimed and unwanted pregnancies may inform and/or change health policies regarding women and infants. This study tests the hypothesis that postpartum Oregon women whose pregnancy was classified as unwanted are less likely to initiate breastfeeding and complete at least 8 weeks of non-exclusive (any) breastfeeding than for an infant whose pregnancy was classified as either mistimed or intended pregnancies. Methods Using the 2005 Oregon PRAMS data set, this cross-sectional study evaluated the relationship between pregnancy intention status and any subsequent breastfeeding duration of at least eight weeks postpartum (classified as binary: yes, no). STATA (version 10.0) was used for analysis of data. Postpartum mothers’ survey responses were classified according to a three-part pregnancy intention status (intended, mistimed, and unwanted). Simple logistic regression analysis was used to identify associations between breastfeeding and individual predictor variables. Backward elimination model-building removed statistically non-significant variables (p > 0.10) from the model based on highest insignificant p-values. Multivariate logistic regression was used to evaluate and control for risk factors known to influence breastfeeding, including age, marital status, race/ethnicity, SES, education, and parity. Sampling weights were accounted for in all analyses owing to the complex sampling design of PRAMS. Results The sample size for 2005 Oregon PRAMS analysis was 1,915 (response rate of 68.2% unweighted, 75.6% weighted). Among respondents, 75.3% breastfed ≥ 8 weeks. Breastfeeding prevalence according to pregnancy intention was 81.4% (intended), 67.5% (mistimed), and 57.6% (unwanted). Compared to women whose pregnancies were unwanted, women with mistimed pregnancies were significantly more likely to breastfeed (OR 1.99, 95% C.I.: 1.00, 3.96) as were women with intended pregnancies (OR 2.45, 95% C.I.: 1.27, 4.72). Covariates significantly associated with breastfeeding at eight weeks included maternal non-smoking at time of survey administration (OR 1.99, 95% C.I.: 1.19, 3.34), increasing maternal age (p = 0.011), absence of maternal postpartum depression (OR 1.85, 95% C.I.: 1.10, 3.12), and being married (OR 1.72, 95% C.I.: 1.15, 2.58). Discussion This study used a three-category pregnancy intention predictor variable to reveal that the association between breastfeeding at eight or more weeks and unwanted pregnancies is significantly different from mistimed pregnancies. This study also confirmed a significant association between overall pregnancy intention and breastfeeding at eight weeks. Women with intended pregnancies were also more likely to breastfeed than those with mistimed pregnancies, although this finding was not significant. This study’s outcome provides useful data on how breastfeeding education and support dollars might be best targeted, by focusing on the 7.49% of pregnancies identified as unwanted instead of including the 30.4% that were merely mistimed. Given the greater potential risks and needs associated with unwanted pregnancies, this category should be emphasized in public health and pediatric research involving breastfeeding and/or pregnancy intention. Future studies should build on these data and evaluate the impact, outcome, and cost-benefit of incorporating pregnancy intention status into clinical counseling affects breastfeeding prevalence among unintended pregnancies. Subsequent cross-sectional studies may also choose to examine if decreasing the incidence of unwanted pregnancies within a community correlates with an increase in breastfeeding and improvement in community-wide health status measures.




School of Medicine



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.