SEARCH

SEARCH BY CITATION

Keywords:

  • breastfeeding;
  • first time mothers;
  • infant health;
  • initiation;
  • peer counselors;
  • WIC

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

Objective

This study was designed to determine whether singleton women who had not previously breastfed and who had a women, infant and children (WIC) peer counselor were more likely to initiate breastfeeding than women not exposed to the WIC peer counselor.

Design and Sample

The retrospective cross-sectional study used data from the 2009 Texas Department of State Health Services (DSHS) WIC Infant Feeding Practices Survey (IFPS) administered through 73 local WIC agencies. Of the 5,427 responses to the 2009 Texas DSHS WIC IFP Survey, 56.6% (N = 3,070) were included in this study.

Measures

The Texas DSHS WIC IFPS, a 55-item survey with multiple-choice and two open-ended questions, was used to evaluate breastfeeding beliefs, attitudes, and practices among women receiving WIC services.

Results

Women who had peer counselor contact during pregnancy, in the hospital, and after delivery were more likely to initiate breastfeeding than women without such contacts, OR = 1.36, 2.06, 1.85, respectively.

Conclusion

Women's decision to initiate breastfeeding is significantly associated with WIC peer counselor contacts. Continued WIC peer counselor program services may increase breastfeeding initiation rates among WIC participants.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

Breastfeeding rates in the most vulnerable populations, including women in low-income brackets, women with less than a high school education, and women from minority populations, are well below Healthy People 2020 goals (Dennis, 2002; Flower, Willoughby, Cadigan, Perrin, & Randolph, 2008; Gill, Reifsnider, & Lucke, 2007a; Sparks, 2010). Nationally, approximately 10 million women, children, and infants receive Special Supplemental Nutrition Program for Women, Infant and Children (WIC) benefits each month. Of these WIC beneficiaries, approximately 52.9% are children, 23.7% are infants, and 23.4% are women (Connor et al., 2011). According to the U.S. Department of Agriculture (2008), 59% of women enrolled in WIC in 2008 breastfed their children; the goal of Healthy People 2020 is 81.9% (U.S. Department of Health and Human Services, 2010).

A major goal of WIC programs is to improve infant nutrition, and encouraging women to breastfeed is a vital component of this effort (U.S. Department of Agriculture, 2008; U.S. Department of Health and Human Services Office of the Surgeon General, 2011). However, often low-income women do not have the social support needed to feel confident about their ability to breastfeed. Thus, strengthening the confidence of a mother who has never breastfed may have a positive impact on her decision to breastfeed subsequent children.

WIC peer counselors are the major source of breastfeeding information and support for the majority of low-income women. Support from WIC peer counselors could thus prove beneficial for low-income women during the critical breastfeeding initiation phase (Arlotti, Cottrell, Lee, & Curtin, 1998; Gill, 2009; Gill, Reifsnider, Lucke, & Mann, 2007b; Humphreys, Thompson, & Miner, 1998; James & Lessen, 2009; Meier, Olson, Benton, Eghtedary, & Song, 2007; Mitra, Khoury, Hinton, & Carothers, 2004; Olsen, Haider, Vangjel, Bolton, & Gold, 2010; Schmied, Beake, Sheehan, McCourt, & Dykes, 2011; U.S. Department of Health and Human Services Office of the Surgeon General, 2011). Typically, WIC peer counselors have been recipients of WIC services themselves and have had successful breastfeeding experiences; they also receive 20 hr of training. In Texas, WIC peer counselors speak to all women about breastfeeding at the first WIC appointment, even if they have other children; they also consult with women who have requested formula; model breastfeeding for expectant women; visit with women in hospitals for mother-to-mother support; staff the breastfeeding hot line after hours; and provide breast pumps after hours when WIC agencies are closed (Texas Department of State Health Services, 2011).

Mothers who are breastfeeding for the first time (or trying again after previous failures) receive emotional support, appraisal and informational support from the peer counseling program (Bandura, 1977; Dennis, 1999; Dennis, Hodnett, Gallop, & Chalmer, 2002; Dykes, Moran, Burt, & Edwards, 2003). In addition, WIC peer counselors provide culturally relevant care to women, which can help to establish a trusting relationship.

Research question

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

No published research, however, has examined the association between WIC peer counselor contact and increased breastfeeding initiation among primiparas and women who have not previously breastfed other children. Therefore, the purpose of this study was to determine whether WIC peer counselor contact was associated with increased breastfeeding initiation in WIC-enrolled singleton women.

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

Design and sample

The sample for this study included 3,070 surveys or 56.6% (N = 5,427) of the completed 2009 Texas Department of State Health Services (DSHS) biennial WIC Infant Feeding Practices Surveys (IFPS). The study was approved by the Institutional Review Boards of The University of Tennessee Health Science Center and the Texas Department of State Health Services. De-identified data from eligible surveys were provided to the investigators by the Texas Department of State Health Services.

Measures

The 2009 Texas DSHS WIC IFPS, a staff-administered survey, provided data to the 73 local WIC agencies to assist them in planning program-specific interventions (Texas Department of State Health Services, 2009). WIC staff members in the local WIC agencies were instructed to administer Scantron paper surveys to eligible women with an appointment for subcertification for their 1-year-old child. The subcertification appointment confirms continuing eligibility in WIC and evaluates nutritional risk and other factors (L. Duong, personal communication, March 26, 2013). Staff members were told not to administer the survey to women who brought in twins, or other multiples for subcertification, or to foster parents. Participants took the survey in English (68.4%) or Spanish (31.6%), based on personal preference. The survey included two questions to determine eligibility. The first question, “Did you give birth to this child,” determined biological mother status. Respondents who answered “no” were instructed to stop the survey. To determine WIC enrollment, the second question asked “Were you enrolled in WIC at any time during your pregnancy with this child?” Respondents who answered “no” were also instructed to stop the survey.

Women who answered affirmatively to these questions proceeded with the remaining 51 multiple-choice and 2 open-ended questions. Instructions to participants included a definition of breastfeeding as “…your baby is getting breast milk from either the breast or the bottle.” This definition of breastfeeding is consistent with the National Immunization Survey's definition of breastfeeding (Centers for Disease Control and Prevention, 2011). However, the definition does not distinguish among exclusive, almost exclusive, partial, or token breastfeeding (Labbok & Krasovec, 1990). To determine whether respondents directly fed from the breast, they were asked “Was this child ever breastfed or fed breast milk even if only once?” A response of “yes” was considered evidence that the respondent had breastfed.

A total of 6,795 surveys were voluntarily completed by WIC enrollees. After the DSHS eliminated surveys from respondents who were not biological mothers of the children receiving services and women who were not enrolled in WIC during any time of their pregnancy, the final sample included 5,427 women (Texas Department of State Health Services, 2009). Women who had previously breastfed children or had fed pumped breast milk to any of their babies were excluded as these women could have greater self-efficacy for breastfeeding than women who had not previously breastfed (Dennis, 2006). Therefore, the analysis was restricted to women who had not breastfed before. Respondents were asked if “…you ever breastfed or fed pumped breast milk to any of your other babies.” A response of “no” qualified them for inclusion in the study.

Respondents were also queried about whether or not they had met with WIC peer counselors “during pregnancy,” “in the hospital,” and “after the baby was born”, defined as in the hospital or after discharge (Texas Department of State Health Services, 2009). Data were coded “1 for yes and 0 for no” based on respondents' answers to whether they had met with WIC peer counselors “during pregnancy,” “in the hospital,” or “after my baby was born.” Many respondents received WIC peer counselor contact at more than one time. Responses of “I don't know” were coded as missing.

The WIC IFPS did not ask respondents to isolate specific interventions (e.g., education) that could have occurred during the times when the WIC recipients met with the peer counselors. Questions related to specific interventions were framed from the perspective of what the WIC staff members covered and included “encouraged me to breastfeed,” “told me about the benefits of the food package for breastfeeding mothers,” “breast pumps are available at WIC,” “told me about the benefits of breastfeeding,” and “when to begin cereal and other foods.”

Analytic strategy

WIC peer counselor contact times and the likelihood of breastfeeding for primiparas and women with previous births and no breastfeeding experience were analyzed using odds ratios, corresponding confidence intervals, and p values. Further analysis was conducted based on the fact that (1) those who were primiparas and (2) those who were not primiparas but did not breastfeed after previous pregnancies were two distinctly different groups and different factors may have influenced their decision to breastfeed for this pregnancy.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

Among 3,070 women, 64% (n = 1,969) breastfed, and 36% (n = 1,101) did not breastfeed. Women who had more than a high school education (n = 621, 69.5%) and Hispanic women (n = 1,360, 69.3%) were significantly more likely to breastfeed than other women, p < .001(see Table 1). Mean ages for women who breastfed and those who did not breastfeed were 24.6 years (5.6) and 24.4 years (5.7), respectively. Women who had peer counselor contact during pregnancy (OR 1.50, 95% CI [1.24, 1.82], p < .001) or after delivery (OR 1.93, 95% CI [1.60, 2.32], p < .001) were significantly more likely to initiate breastfeeding than women who did not have a peer counselor (see Table 2). Women who received peer counselor contacts during pregnancy and hospitalization had the highest percentage of breastfeeding initiation (75.1%), OR 1.95, 95% CI [1.43, 2.65], p < .001 of all the women who had WIC peer counselor contact. Interestingly, the number of women who had WIC peer counselor contact in the hospital was lower (n = 281) than the number who had contact at other times (see Table 2).

Table 1. Study Demographics and Breastfeeding of WIC-Enrolled Women Sample N = 3,070
FactorLevelBreastfed this childp value
Yes n (%)No n (%)
Note
  1. Variations in totals are a result of women selecting “don't know” or missing data.

Education<High school education583 (65.4)308 (34.6)<.001
High school education688 (58.1)496 (41.9)
>High school education621 (69.5)273 (30.5)
RaceWhite368 (59.1)255 (40.9)<.001
Black180 (45.9)212 (54.1)
Hispanic1,360 (69.3)601 (30.7)
Asian19 (63.3)11 (36.7)
Native American9 (64.3)5 (35.7)
Other10 (83.3)2 (16.7)
Table 2. Peer Counselors Contact and Breastfeeding Initiation of Primiparas and Women Who Had Not Previously Breastfed Sample N = 3,070
Peer counselor contactMothers (n)Breastfed (%)OR95% CIp value
Note
  1. CI = confidence interval. For each category, the yes and no responses do not add up to the total study sample size. Variations in the category totals are due to missing data.

During pregnancy
Yes2,06866.01.50[1.24, 1.82]<.001
No53856.3
During hospitalization
Yes28174.01.89[1.43, 2.50]<.001
No2,28960.2
After the baby was born
Yes1,88566.61.93[1.60, 2.32]<.001
No61450.8
During pregnancy, hospital, after the baby was born
Yes22074.61.91[1.39, 2.62]<.001
No1,98660.5
During pregnancy, hospital
Yes23775.11.95[1.43, 2.65]<.001
No2,11060.8
During pregnancy, after the baby was born
Yes1,62666.91.68[1.41, 2.01]<.001
No73054.5
Hospital, after the baby was born
Yes23974.51.95[1.44, 2.64]<.001
No2,07659.9

Results from logistic regression showed that type of mother, that is, primiparas or mothers who did not breastfeed after previous pregnancies, did not modify the effect of peer counselor contact (see Table 3). Peer counselor contacts (see Table 3) increased the likelihood of breastfeeding, OR = 1.36, 2.06, 1.85, whether during pregnancy, in the hospital, and after delivery, with p = .0024, <.0001, <.0001, respectively. First time pregnant women were more likely to breastfeed than those who did not breastfeed after previous pregnancies (by a factor of three, with p < .0001).

Table 3. Adjusted Odds Ratio and 95% Confidence Interval for Women Who Had Peer Counselor Contact and Breastfed Sample N = 3,070
Peer counselor contact and women who breastfedOR95% CIEffect of primiparas versus multiparas who had not breastfed previous children
OR95% CI
Note
  1. Variations in totals are a result of missing data. DP = during pregnancy; IH = in hospital; AD = after delivery.

During pregnancy1.361.12, 1.663.142.62, 3.77
In hospital2.061.54, 2.753.462.89, 4.14
After delivery1.851.53, 2.243.162.63, 3.80
DP + IH2.071.51, 2.843.492.88, 4.22
DP + AD1.561.30, 1.883.162.61, 3.81
IH + AD2.071.51, 2.833.412.82, 4.12
DP + IH + AD2.001.44, 2.763.422.82, 4.16

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

This study found that primiparas and women who did not breastfeed with previous pregnancies were more likely to initiate breastfeeding when they received WIC peer counselor contact. The survey did not ask respondents to identify which of three focal strategies (education, appraisal, and emotional support) was most commonly used by counselors. Nonetheless, the significance of contact with a peer counselor cannot be underestimated. This finding is consistent with the findings of previous studies (Arlotti et al., 1998; Dennis et al., 2002; Gill, 2009; Gill et al., 2007a; Humphreys et al., 1998; Meier et al., 2007; Mitra et al., 2004; Olsen et al., 2010; Schmied et al., 2011).

To increase women's contact with peer counselors, strategies have been suggested for increasing collaboration among agencies during the prenatal period (Meier et al., 2007) and in the postpartum period, through granting peer counselors access to women in the hospital to assist with breastfeeding (U.S. Department of Health and Human Services Office of the Surgeon General, 2011). These collaborative efforts could potentially provide support to increase women's breastfeeding confidence, which may lead to increased breastfeeding rates in this vulnerable population of women.

A meta-analysis by Guise et al. (2003) found that women who received a combination of support and breastfeeding education had higher rates of breastfeeding initiation and duration than those who received either support or breastfeeding education alone. Furthermore, the U.S. Preventive Services Task Force has suggested that interventions (such as breastfeeding support and education) during the prenatal and postnatal period are strongly associated with increased breastfeeding rates (Chung, Raman, Tikalinos, Lau, & Ip, 2008). A limitation of the WIC IFPS was the failure to ask respondents which types of support (e.g., education and encouragement) provided by WIC peer counselors were received and which were most helpful. However, as Texas WIC peer counselor procedures promote the use of education and other supportive activities, it is likely that a combination of these procedures was used (Texas Department of State Health Services, 2011). A future qualitative study could ask peer counselors to identify which interventions they find of most value and indicate whether all WIC recipients receive similar interventions in similar doses. Peer counselors could also identify whether and how they decide to vary interventions based on individual characteristics of the WIC recipients.

One limitation of this study is the fact that the WIC IFPS was designed to evaluate breastfeeding beliefs, attitudes, and practices, not to address other questions. Also, there were no published tests of the reliability or validity of the WIC IFPS instrument; and as a secondary data analysis, the study was limited by inability to confirm prior WIC participation and multiple births, a lack of control over survey design and sample collection, and missing data. Furthermore, respondents may have not known whether they met with a WIC peer counselor or WIC staff members as the roles were not explicitly defined in the survey, and this could have created inaccuracies in responses.

Other study limitations included bias created by the cross-sectional design, which made it impossible to establish a temporal relationship between WIC peer counselor contact and breastfeeding initiation. Women who became pregnant may have wanted to do what was best for their baby and breastfed regardless of whether they had contact from WIC staff members (Szklo & Nieto, 2007). Finally, literacy and the length of time required to complete the 53 questions of the WIC IFPS could have affected participant responses and created self-selection sampling bias. Therefore, the results can only be generalized with caution.

Future analyses should compare women with and without previous breastfeeding experience to determine whether WIC peer counselor contact is associated with increased breastfeeding initiation and duration rates and what variables improve these rates.

There is clear evidence that breastfeeding supports the health of infants and older children (Dennis, 2002; Bitler & Currie, 2005); Lee & Mackey-Bilaver, 2007; Gill et al., 2007a; James & Lessen, 2009; Foster et al.,2010; Richards et al., 2011). In this study, WIC peer counselor support (in the form of contact) was associated with improved breastfeeding initiation rates in a highly vulnerable population. WIC peer counselor support may improve breastfeeding initiation rates for women by teaching breastfeeding skills and improving women's confidence in their ability to breastfeed. Increasing peer counselor availability in all WIC agencies could thus increase breastfeeding initiation and duration rates, especially among disadvantaged and vulnerable communities of women.

In the future, a national group of WIC peer counselors and breastfeeding experts could collaborate in the establishment and validation of a national WIC survey, much like the Behavioral Risk Factor Surveillance Survey. Additional questions could be used to capture data to inform program strategies at the state level. Dissemination of survey results and technical assistance based on the findings could result in the expansion of WIC peer counselor programs to vulnerable, underserved women and improved health for their infants and children.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References

The authors would like to thank Texas DSHS, Dr. Gita Mirchandani, Dr. Sara Gill, PhD, RN, IBCLC and Julie Stagg, MSN, RN, IBCLC for their valuable contributions to the development of this study and manuscript, and Dr. Huxian Song for her technical assistance.

References

  1. Top of page
  2. Abstract
  3. Background
  4. Research question
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgments
  9. References