Ninth Annual Maternal and Child Health Epidemiology Conference / December 10-12, 2003
Women, Infant, and Children (WIC) Program Participation
and the Effect on Breastfeeding Practices in Louisiana
DIONKA PIERCE: Good afternoon. I’m pleased to be presenting an analysis that we did looking at the effect of WIC on breastfeeding in Louisiana mothers. I want to acknowledge Dr. *Tretran who is a PRAMS analyst. He’s here and he’s very surprised I’ve just acknowledged him, and also former PRAMS data manager, Joanna *Hobble, who also participated. As far as background, Louisiana continues to have lower rates of breastfeeding for more than one month after delivery. In 1997, the rate was 26%, in ’98 31%, and in 1999, it was 35%. A previous analysis revealed implications of hospital breastfeeding counseling and practices on continued breastfeeding in Louisiana mothers. That study found that a woman is likely to breastfeed if she was: One, given a formula pack. Two, if the hospital staff did not provide her with information about breastfeeding. Three, if her baby was not in the same room with her at the hospital, and four, if she was black, less educated, unmarried and lived in a rural area.
This study examines the relationship between WIC participation and breastfeeding initiation and duration in Louisiana mothers from 1997 through 1999. The methods used in this analysis were, we used Louisiana PRAMS data from 1997 through 1999, multi-varied logistic regression with hierarchical modeling, our alpha level was set at .05 for statistical significance, and the software we used was (inaudible) 7.5. The outcome that we looked at was no breastfeeding or breastfeeding less than one week. The question used here was, “Did you breastfeed or pump breast milk to feed your infant after delivery?” And we grouped the responses, “No, I did not breast feed,” or if the mother responded she breastfed less than one week. The risk factors that were included in our model were WIC status which was our main risk factor, maternal smoking, pregnancy intention, the number of previous live births, breastfeeding counseling, Medicaid status, and maternal demographics such as race, age, and education.
Table one shows a description of the Louisiana PRAMS population from 1997 through 1999. For each predictor, the category with the highest percentage is highlighted, so in terms of age, we have 30.7% that are in the 20 to 24 category, education 38.6% had at least 12 years of education, 59.7% of the PRAMS population for these years were white women, and 66.3% responded “no” to the question, “Did you smoke during the three months before pregnancy or the--” I’m sorry, the question is, “Have you smoked at least 100 cigarettes in your lifetime?” And 66.3% of the women for 1997 through 1999 reported that they did not smoke or they smoked less than 100 cigarettes in their lifetime. I also want to make a note here. We chose the race black and white because other races in Louisiana make up less than 5% of our population.
This table two shows a continued description of the PRAMS population. Here we see, again, the categories for each predictor with the highest percentage are highlighted, so 53% or roughly 54% of our PRAMS population are on WIC. On Medicaid status, they are evenly distributed there. Fifty-two percent did not intend their pregnancies, 83% did receive breastfeeding counseling, and 42.8% were (inaudible). This was their first live birth. Table three shows the percent of no breastfeeding or breastfeeding less than one week by maternal demographics and social characteristics in WIC participants. So, among women who do not breastfeed or breastfed less than one week and who were WIC participants, 83.2% of them were less than 20 years, 79.9% of them had less than a 12th grade education, 77% were black, and 61% smoked greater than or equal to 100 cigarettes and were current smokers. The percent of no breastfeeding or breastfeeding less than one week and these predictors, 82% were on Medicaid, 67% did not intend their pregnancy, 58% did receive breastfeeding counseling, and 58% were (inaudible), and these are among mothers who were on WIC and who did not breastfeed or breastfed less than one week.
Now, this final table shows our adjusted odds ratios of the statistically significant predictors of no breastfeeding or breastfeeding less than one week. In here, we see that if a woman is on WIC, she is 1.7 times more likely not to breastfeed or breastfeed less than one week, and that is our main risk factor. You can reference your handout for the others. In conclusion, this study has revealed an association between WIC participation and non-initiation or discontinuation of breastfeeding in Louisiana mothers, 1997 through ’99. However, the data used in this analysis does not reflect the more recent breastfeeding promotional activities conducted in Louisiana WIC clinics, and I have listed those activities. Those are mandatory breastfeeding training for staff, breastfeeding counseling for mothers, available breast pumps, encouragement by staff at initial contact and follow-up visits, and promotional out reaches mainly held during the month of August. In order to increase the rates of breastfeeding in Louisiana, we recommend that breastfeeding education and outreach effort should be framed to reach black women, women who are less educated, and women who are current smokers. Thank you.