Identifying Best-Practice for Increasing Breastfeeding Initiation Rates Among Adolescent Mothers
Ellen Hollander Sande, Eliana Roshel, Sarika Downing and Maria Mendez, New York University
Research in Nursing NURSE-GN.2303
Rona F. Levin, PhD, RN, and Nancy E. Kline, PhD, RN, CPNP, FAAN
Breastfeeding rates among American adolescent mothers are consistently lower than those observed within the adult population. Adolescent mothers are a unique population who face challenges specific to their age group regarding initiation of breastfeeding. Interventions aimed at increasing breastfeeding rates in this population, specifically targeted to their developmental stage, are vital though rarely studied. Through a strategic search, three studies were identified for evaluating the effects of various interventions on breastfeeding initiation rates. The interventions examined in each study were specifically designed for this population, including guidance from intrapartum nurses, educational programs, and peer support. Two of the three studies found significant increases in breastfeeding initiation in their treatment groups; a third initially showed an increase in initiation rates but this result became insignificant once the data were adjusted for covariates. Synthesis of the evidence supports educational and supportive interventions tailored to the needs of the adolescent mother to increase rates of breastfeeding initiation in this population. Implications for practice include recommendations for utilization of the Breastfeeding Educated and Supported Teen (BEST) Club within high schools, prenatal clinics, and hospitals; training of intrapartum nurses in the Supportive Needs of Adolescents during Childbirth (SNAC) program; and improved support of breastfeeding in high schools.
The purpose of this paper is to determine what interventions are supported by the evidence as best practice for increasing breastfeeding initiation rates among American adolescent mothers. We reviewed and critiqued three trials, each of which tested interventions aimed at improving breastfeeding initiation rates in this challenging population through educational programs, nursing interventions, and/or peer support. Results of the three studies were synthesized and presented in narrative form and in an overall table of evidence. Taken together, the results showed that interventions tailored to adolescent mothers do significantly increase breastfeeding initiation rates and can be implemented with success in a variety of settings and cultures.
The benefits of breastfeeding to women and their children have been well established in the literature. Evidence has shown that women who breastfeed have lower rates of type II diabetes mellitus (Ip et al., 2007), breast and ovarian cancers (Ip et al., 2007; Labbok 2001), and endometrial cancers (Newcomb & Trentham-Dietz, 2000). Systematic reviews have shown that breastfeeding lowers an infant’s risk of acute otitis media, atopic dermatitis, gastroenteritis, lower respiratory tract infections, asthma of early childhood, obesity, diabetes mellitus, childhood leukemia, sudden infant death syndrome, and necrotizing enterocolitis (Ip et al., 2007). Though many of these benefits may be dose dependent (Ip et al., 2007), initiation alone is of great benefit; for infants in the United States, risk of death within the postneonatal period (28 days to 12 months) is reduced by 21% for those who were ever breastfed (Chen & Rogan, 2004).
Though the World Health Organization (2012), United Nations Children’s Fund (2012), the American Academy of Pediatrics (2012), and the American Academy of Family Physicians (2012) recommend that infants should be fed breast milk exclusively for the first six months of life (with continued breastfeeding along with complementary feeding thereafter), breastfeeding is initiated with 75%of infants in the U.S., and only 44% are receiving any breast milk at 6 months (Save the Children, 2012). For adolescent mothers in the United States, breastfeeding rates are significantly lower. In 2007, only 59.7% of mothers under age 20 initiated breastfeeding, with only 22.2% breastfeeding at six months postpartum (Centers for Disease Control, 2012). Though there is work to be done to improve breastfeeding duration, the work of Chen and Rogan (2004) clearly shows that improving rates of breastfeeding initiation in American mothers would have profoundly beneficial effects for the health of their infants. Given the lower rates of breastfeeding among adolescent mothers (CDC, 2012) and the developmental and cultural differences between adolescent and adult mothers (Drake, 1996; Maehr, Lizarraga, Wingard & Felice, 1993), there is a need for clinicians working with this population to use interventions tailored to the unique needs of the adolescent mother (Drake, 1996). Though a search of the literature provides few examples of such population-specific interventions, the three studies reviewed for this paper are examples of how these interventions can be developed and utilized with success.
Multiple databases were searched with the bulk of relevant data noted within the Cochrane Library, CINAHL Plus with Full Text, and Proquest. Experimentation with each database was necessary to determine the best search terminology; for example, CINAHL preferred breastfeeding as two words, and the Cochrane library yielded the most specific results when using lactation versus breastfeeding. The Cochrane Central Register of Controlled Trials and Systematic Reviews was searched with an additional limiter to exclude research published before 2007. ProQuest searches were limited to peer reviewed studies only, and similarly CINAHL results were narrowed by allowing only research articles and those with age-specific relevance. This research strategy generated about 100 articles. Three studies were selected for high levels of evidence and relevance: one randomized controlled trial and two quasi-experimental controlled trials. All three studies measured the effects of adolescent-specific interventions on breastfeeding initiation rates in American adolescent mothers (see Table 1).
Presentation of Evidence
Grassley and Sauls (2011) evaluated the effectiveness of a nursing intervention at a Texas-based tertiary hospital on childbirth satisfaction and initiation of breastfeeding in adolescent mothers in a separate samples post-test quasi-experimental study. The Supportive Needs of Adolescents during Childbirth (SNAC) program was used to target adolescent mothers during the intrapartum phase. This intervention, based on the adolescent support model and labor support model, emphasizes the need for age-appropriate care during the intrapartum phase of pregnancy for improvement on childbirth satisfaction, which in turn has proved to increase breastfeeding initiation rates among adolescent mothers (Shealy, Li, Benton-Davis, & Grummer-Strawn, 2005). Although randomization within the same time frame was not possible in this setting, two phases of the study were done for control group and intervention group comparison. In the first phase, adolescent mothers enrolled in the study received standard care, and data collection was performed, including rates of breastfeeding initiation. Once data were collected, intrapartum nurses were educated on providing age-specific care to adolescents by focusing on the psychosocial needs of this age group. In the second phase of the study, the interventions performed by the nurses included teaching with emphasis on skin-to-skin contact between mother and child, assistance with appropriate positioning and latching within the first hour of giving birth, and age-appropriate support during the intrapartum period. All participants received follow up for data collection during the postpartum phase prior to discharge, and at six weeks and three months postpartum. A total of 132 participants were recruited with only 106 remaining at discharge and 82 completing the entire study. Five subjects were eliminated by exclusion criteria, 21 were lost through failure to obtain packets containing consent and contact information at discharge, and 24 were lost due to inability to follow-up at the six week and three month mark. The time limitation on the intervention period, given that mothers were transferred to the postpartum unit shortly after giving birth, was a weakness of this study. The outcome measured was rate of initiation within the first hour postpartum; during this short window, the nurses responsible for carrying out the interventions may have had other tasks preventing them from fully focusing on the SNAC intervention. Since the intervention was aimed at participants already in labor, only mothers who wanted to breastfeed, whether exclusively or in conjunction with formula, were included, which may have created significant sample bias. The age range of the participants may pose a limitation on the generalizability; though the selection criteria show that the authors intended to capture adolescents between ages 13 and 20, the youngest participant was 15 years old, and 83% of the sample was between 17 and 20 years old. Results may vary for younger adolescents. The high percentage of participants of Hispanic background, 52.8%, not only brings into question the generalizability of the results across cultures, but may also be a confounding factor: There are higher rates of breastfeeding among Hispanic Americans (Grassley & Sauls, 2012) and, more specifically, among Mexican adolescent mothers (McDowell, Wang, & Kennedy-Stephenson, 2008), which may have been a significant percentage of the Hispanic mothers in this hospital in Texas. However, the sample was more diverse than other studies reviewed for this paper, and the authors state that the ethnic makeup of the sample is representative of the population of adolescent mothers in their area. Significant differences in living arrangements between groups were noted by the authors as possible cofounding factors; participants in the control group were more likely to live with a significant other (48% vs. 37%), whereas participants in the experimental group were more likely to live with their parents (46% vs. 26%). Despite a substantial loss of participants, sample size for those followed up until discharge (106) was sufficient based on a power analysis of .80. Though lack of traditional randomization suggests the possibility of confounding factors related to changes within the hospital from one time period to the next, the authors minimized this possibility by maintaining the same nurses for the two groups. Overall, the study found that participants in the intervention group were significantly more likely to breastfeed within the first hour of giving birth than the control group, 83% and 56% respectively, with a p-value of 0.004. Effect size was moderate at 0.30. Of note, no statistical significance was found on rates of breastfeeding at discharge (p=0.88). These findings suggest that intrapartum nurses are effective in promoting the initiation of breastfeeding within the first hour of birth, though further study should be done on the role of postpartum nurses in continued breastfeeding promotion and assistance.
Volpe and Bear (2000) hypothesized that breastfeeding education geared towards pregnant adolescents and the use of peer counselors would increase breastfeeding initiation, defined as breastfeeding one or more times daily for 3 days postpartum, among students enrolled in a high school adolescent pregnancy program. Participants of this quasi-experimental study attended a high school that already incorporated a program designed for pregnant adolescents, which allowed the students to continue their high school education while learning parenting skills. Classes offered included childbirth preparation, cardio-pulmonary resuscitation, infant care and parenting. However, prior to this study breastfeeding education within the program was limited. The educational intervention utilized, the Breastfeeding Educated and Supported Teen (BEST) Club, was developed by an International Board Certified Lactation Consultant (IBCLC), who is also a registered nurse clinician. The BEST Club consisted of three weekly one-hour sessions that focused on nutrition, safety, and maternal health. The approach to learning was interactive, non-intimidating and supportive, and prizes were given each week to participants to encourage attendance and participation. The program also sought to help the young mothers overcome three major obstacles to breastfeeding noted within their specific population: a short (two week) maternity leave from school, which put a hindrance on establishing adequate milk supply and bonding; teachers’ assumptions that breastfeeding was being used as an excuse to miss class time; and opposition to breastfeeding among the mothers’ support persons. Mothers who chose to breastfeed worked with the study’s breastfeeding peer counselor, who advised them on ways to maintain milk supply and promote bonding within the constraints of their short maternity leave and school schedule. This aftercare lasted six months, though data on breastfeeding duration were not collected. To address teachers’ resistance, the peer counselor educated teachers and created feeding schedules around breaks and class. To address the issue of opposition from the adolescent mothers’ support systems, partners and family members were encouraged to attend the breastfeeding classes with the mothers. The study sample consisted of 91 pregnant females between the ages of 14-19 with a mean age of 16.2 years; the sample was 63% Caucasian, 26% African American and 11% Hispanic. Due to the close contact between students in the program, participants were not randomized into treatment and control groups within the same year. Data were collected on the control group (n=48), who received the standard breastfeeding preparation offered in the school’s program, during the 1995-1996 school year. The experimental group (n=43) received the same standard breastfeeding preparation as the control group as well as the three additional BEST classes and peer counselor support in 1996-1997. A major weakness in the study was lack of data regarding what stage of pregnancy each adolescent was in at the start of the intervention; differences in outcome depending on stage of pregnancy at initiation of the intervention may have been significant. The authors of the study noted as a weakness that use of a traditional approach to randomization would have removed possible inconsistencies between groups, including teacher changes or updates to program teaching materials. Repeating the study within the same year and splitting the control and experimental groups into two separate classes could eliminate these factors as possible covariates. Although this study was intended only to measure initiation, data analyzing the effects of the intervention on breastfeeding duration would have been valuable. Seven of the 48 participants in the control group (14.6%) and 28 of the 43 in the experimental group (65.1%) initiated breastfeeding, a significant difference between groups (p<.001). Logistic regression analysis was run to control for age and race, and neither were found to have any significance for the outcomes. These results indicate that the BEST program and use of peer counselors can be utilized to increase rates of breastfeeding initiation among adolescent mothers across multiple cultures.
The study by Wambach, Aaronson, Breedlove, Domain, Rojjanasrirat & Yeh (2011) was a randomized controlled trial evaluating the effect of a mixed intervention, utilizing established educational programing and peer support, on breastfeeding initiation rates and breastfeeding duration among adolescent mothers. A peer counselor with experience as a breastfeeding adolescent mother and a registered nurse lactation consultant (IBCLC) provided the interventions to the treatment group, utilizing the BEST curriculum (Volpe & Bear, 2000). Intervention began in the second trimester and continued through the fourth week postpartum. A research biostatistician randomized the subjects into a treatment group, an attention control group, and a usual care control. Various data were collected regarding demographics, perceptions on behavioral control, intentions, knowledge and subjective norms to ensure balance between the experimental, control and attention control groups. Participants between the ages of 15 and18 were recruited from seven prenatal clinics, two public health clinics, four high schools and five hospitals. The study excluded younger and older adolescents, as the researchers feared developmental and cognitive differences between age groups would confound outcomes. Despite the wide recruiting base, the sample was predominantly African-American and homogeneous in terms of socio-economic stratification. Baseline data were collected from 315 participants; of those, 289 provided data on initiation rates and 201 subjects provided data on duration. Initiation rates were measured prior to hospital discharge, and follow-up data on duration were collected at three and six weeks and at two, three, four, five, and six months postpartum. Attrition within the study was detrimentally high, ranging from 24% to 30% for the three groups. All subjects were accounted for based on exclusionary criteria (e.g., preterm labor) or loss-to-follow-up. The high attrition rates in conjunction with unplanned covariates caused under-powering within the study despite the authors’ efforts to provide sufficient sample size based on initial power calculations. Though the authors found significant differences between groups in breastfeeding initiation rates (p<.03), this difference was no longer significant after adjustments were made for covariates. Of note, the experimental group had a significantly greater proportion of participants intending to return to school after giving birth. No definitive conclusion can be drawn from this study regarding breastfeeding initiation. Though the initial results showed a significant increase, those results were invalidated due to under-powering and covariates, which may be suggestive of a type II error. Increased breastfeeding duration in the treatment group, however, was consistently significant throughout the six-month course of follow-up.
Summary and Synthesis
Two of the studies reviewed showed significant increase in breastfeeding initiation in their respective treatment groups. Volpe and Bear (2000) found an increase in breastfeeding initiation within the first three days post-partum in the treatment group with a p < 0.001, and the treatment group in the study of Grassley and Sauls (2011) increased initiation rates within the first hour postpartum with a p-value of 0.004. Wambach et al. (2010) state their intervention did not significantly increase breastfeeding initiation. However, their initial results did show a significant (p<.03) increase in initiation rates, and this difference lost significance only after adjustment was made for covariates. The authors note that their study lacked the power necessary to identify group effect after the addition of the covariates, specifically the identification of a greater proportion of participants intending to return to school soon after giving birth – given the rationale, the possibility of a type II error cannot be ruled out. A table of evidence is presented in Table 2 summarizing the three studies.
The three studies altogether cover a variety of settings (hospitals, high schools and prenatal clinics) and regions (Texas, Florida, Kansas and Missouri). The predominant ethnicities of participants varied across studies as well. Looking at the three samples as parts of one whole, participants of African American, European American and Hispanic American descent are all well represented. This variation between the studies strengthens the generalizability to the greater population of American adolescent mothers, and implies applicability and feasibility across various settings. Of note, two of the three studies indicated that return to school shortly after giving birth and management of breastfeeding within the school setting posed challenges for participants.
Together, the studies looked at 486 participants, with 200 in the experimental groups and 286 in the control groups. Synthesizing the data, 80% of participants in the experimental groups initiated breastfeeding, compared with 54% of participants in the control groups. We performed a chi square on the raw data and found the results to be significant (p<.001). Overall the results support the use of adolescent-specific programs providing education, support, and peer counseling to increase rates of breastfeeding initiation among American adolescent mothers.
Recommendations for Practice
The importance of breastfeeding for the health of women and infants has been well established, and the low rates of breastfeeding among adolescent mothers coupled with the specific needs of this population indicate the necessity for adolescent-specific interventions. Based on the evidence, recommendations for practice include training of intra-partum nurses in the Supportive Needs of Adolescents during Childbirth (SNAC) program, and more widespread use of peer counselors and the Breastfeeding Educated and Supported Teen (BEST) Club in high schools, prenatal clinics, and hospitals. Furthermore, substantial maternity leave from high school coupled with short-term distance learning and increased support for breastfeeding during the school day would support adolescent mothers in overcoming a significant obstacle to breastfeeding. Further research in this area would be valuable to confirm results, refine interventions, and explore ways to continue to improve duration of breastfeeding once initiated.
American Academy of Family Physicians. (2012). Breastfeeding (Policy Statement). Retrieved from http://www.aafp.org/online/en/home/policy/policies/b/breastfeedingpolic…
American Academy of Pediatrics. (2012). Policy statement: Breastfeeding and the Use of Human Milk. Retrieved from http://pediatrics.aappublications.org/content/129/3/e827.short
Centers for Disease Control and Prevention (2012, June 8). Provisional Breastfeeding Rates by Socio-demographic Factors, Among Children Born in 2007. Retrieved from http://www.cdc.gov/breastfeeding/data/NIS_data/2007/socio-demographic_a…
Chen, A., Rogan, W. J. (2004). Breastfeeding and the risk of postneonatal death in the United States. Pediatrics, 113(5), e435-9. Retrieved from http://pediatrics.aappublications.org/content/113/5/e435.full.pdf
Drake, P. (1996). Addressing developmental needs of pregnant adolescents. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 25(6), 518-24. doi: 10.1111/j.1552-6909.1996.tb01473.x
Grassley, J. S. & Sauls, D. J. (2012). Evaluation of the supportive needs of adolescents during childbirth intrapartum nursing intervention on adolescents’ childbirth satisfaction and breastfeeding rates. Journal of Obstetric, Gynecologic, & Neonatal Nursing, 41, 33-44. doi: 10.1111/j.1552-6909.2011.01310.x
Ip S., Chung M., Raman G., Chew P., Magula N., DeVine D., Trikalinos T., & Lau J. (2007). Breastfeeding and maternal and infant health outcomes in developed countries. AHRQ Publication No. 07-E007. Retrieved from http://www.ahrq.gov/downloads/pub/evidence/pdf/brfout/brfout.pdf
Labbok, M. H. (2001). Effects of breastfeeding on the mother. The Pediatric clinics of North America, 48(1). 143-58.
Levin, R. F. (2011). Edifices of evidence: the proliferation of pyramids. Research & Theory For Nursing Practice, 25(1), 8-10. doi:10.1891/0889-722.214.171.124
Maehr, J., Lizarraga, J., Wingard, D. & Felice, M. (1993). A comparative study of adolescent and adult mothers who intend to breastfeed. Journal of Adolescent Health, 14(6). 453-57. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8241202
McDowell, M., Wang, C., & Kennedy-Stephenson, J. (2008). Breastfeeding in the United States: Findings from the National Health and Nutrition Examination Survey, 1999-2006. NCHS Data Brief ,5. Retrieved from http://www.cdc.gov/nchs/data/databriefs/db05.pdf
Newcomb, P. A., & Trentham-Dietz, A. (2000). Breast feeding practices in relation to endometrial cancer risk, USA. Cancer Causes & Control, 11(7), 663-7. Retrieved from https://ezproxy.library.nyu.edu/login?url=http://search.proquest.com/do…
Save the Children (2012). Nutrition in the First 1,000 Days: State of the World’s Mothers 2012. Retrieved from http://www.savethechildren.org/atf/cf/%7B9def2ebe-10ae-432c-9bd0-df91d2…
Shealy, K. R., Li, R., Benton-Davis, S., & Grummer-Strawn, L. M. (2005). The CDC guide to breastfeeding interventions. Retrieved from http://www.cdc.gov/breastfeeding/pdf/breastfeeding_interventions.pdf
United Nations Children’s Fund, (n.d.) Infant and Young Child Feeding. Retrieved June 10, 2012 from http://www.unicef.org/nutrition/index_breastfeeding.html
Volpe, E. M. & Bear, M. (2000). Enhancing breastfeeding initiation in adolescent mothers through the Breastfeeding Educated and Supported Teen (BEST) Club. Journal of Human Lactation, 16(3), 196-200. doi: 10.1177/089033440001600303
Wambach, K. A., Aaronson, L., Breedlove, G., Domain, E. W., Rojjanasrirat, W., & Yeh, H. (2011). A randomized controlled trial of breastfeeding support and education for adolescent mothers. Western Journal of Nursing Research, 33(4), 486-505. doi: 10.1177/0193945910380408
World Health Organization, (2012). Infant Nutrition. Retrieved from http://www.who.int/topics/infant_nutrition/en/