"Psychosocial factors and duration of breastfeeding among women in"
Psychosocial factors and duration of breastfeeding among women in Denmark Annotated Bibliography CIS 495 Lucy Marcil 11/18/05 Table of Contents Data set 3 General breastfeeding 3 Postpartum depression 6 Psychosocial factors 9 Biological factors 14 Alcohol 16 Smoking 19 Medications 20 Breastfeeding policies 23 2 Data set Olsen, J., Melbye, M., Olsen, S. F., Sorensen, T. I., Aaby, P., Andersen, A. M., et al. (2001). The Danish National Birth Cohort--its background, structure and aim. Scand J Public Health, 29(4), 300-307. This article describes the conception of the Danish National Birth Cohort (DNBC), a data set designed to allow long-term study of pregnancy and early life morbidity and mortality. This area of study is especially important because of a high correlation between infant health and later life health. The ethics committee granted permission to follow the cohort for 20 years. The authors purposefully designed the cohort to include a large number of dyads (100,000) to allow study of rare conditions. They recruited subjects through general practitioners (GPs). Some regional unions initially recommended that GPs not participate in the study, which led to somewhat biased subject selection. Data collection included four telephone interviews during and after pregnancy, a food frequency questionnaire, blood sampling, and linkage to national disease registers. One shortcoming of the study is that it required women speak Danish well enough to participate in a telephone interview. Thus, most of the immigrant population was excluded. Ten percent of pregnancies occur in immigrants. General breastfeeding Ball, T. M., & Wright, A. L. (1999). Health care costs of formula-feeding in the first year of life. Pediatrics, 103(4 Pt 2), 870-876. These authors assessed the relative risk of lower respiratory tract illness, otitis media, and gastrointestinal illness in respect to breastfeeding status in the first year of life. Subjects were 1588 American and Scottish infants, who were characterized as never, partially, or exclusively breastfed. They found that compared to exclusively breastfed infants, never breastfed ones had excessive office visits, hospitalization and prescriptions. These difference translate into an extra $331-$475/infant during the first year. This study does not include a long-term analysis of costs, nor does it try to quantify indirect costs and benefits. However, it provides moderately strong evidence in favor of policies promoting breastfeeding. Benis, M. M. (2002). Are pacifiers associated with early weaning from breastfeeding? Adv Neonatal Care, 2(5), 259-266. This randomized study assessed the efficacy of a counseling intervention advocating decreased pacifier use. Some theorize that pacifier use may lead to incorrect suckling technique during breastfeeding. Out of 281 women, 140 were advised to avoid pacifiers and given alternative techniques for dealing with fussy infants (experimental group). Both groups were counseling on the benefits of breastfeeding. Researchers did find that women in the experimental group were less likely to give pacifiers and also less likely to wean by 3 months. However, some causal ambiguity does exist. A third intervening variable, such as breastfeeding problems or lack of breastfeeding motivation, may explain both increased pacifier use and early weaning. 3 Birth, Initiation of breastfeeding, and the First Seven Days after Birth. In Facts for Feeding. (2003). Academy for Educational Development. This document provides an easily understandable guide for health care providers on promoting exclusive breastfeeding during labor and delivery, the first few hours after birth, the first few days after birth, and the rest of the first week after birth. This information is critical since early successful breastfeeding is predictive of long-term success. The document includes strategies for both proactive care and interventions when problems arise. This type of paper is a useful example of effective dissemination of information to health care professionals. Information dissemination is necessary for the successful implementation of policies. Grummer-Strawn, L. M., & Mei, Z. (2004). Does breastfeeding protect against pediatric overweight? Analysis of longitudinal data from the Centers for Disease Control and Prevention Pediatric Nutrition Surveillance System. Pediatrics, 113(2), e81-86. This study assesses the relationship between duration of breastfeeding and body mass index (BMI) in a cohort of low-income 4 year olds. Dietary and other lifestyle factors probably differ between low and higher income children, so these results may not be applicable to children in higher socioeconomic groups (as in my study). The authors found that longer durations of breastfeeding were related to lower risk of over- and underweight among Caucasians. This effect was only present for breastfeeding durations of 3 months or longer. Behavioral differences may explain why the relationship was insignificant for Hispanic and Black children. Philipp, B. L., Merewood, A., & O'Brien, S. (2001). Physicians and breastfeeding promotion in the United States: a call for action. Pediatrics, 107(3), 584-587. This article reviews the current breastfeeding situation in the US and provides suggestions for improvement. The authors note that although breastfeeding rates in the US are low in general, they are much lower among WIC recipients. This difference is at least partially due to formula coupons. Many pediatricians, experts with potential to influence maternal behavior, also lack adequate breastfeeding knowledge. This article further points out that the US lags behind other countries in respect to implementing the Baby Friendly Hospital Initiative (WHO sponsored). Only 25 US hospitals are certified Baby Friendly, in comparison to > 16,000 worldwide. The US also has failed to enforce the International Code of Marketing of Breast Milk Substitutes. Finally, the authors provide some suggestions, which mainly focus on the agency of pediatricians to better educate themselves and their patients about breastfeeding and to become politically active on the issue. Li, R., Darling, N., Maurice, E., Barker, L., & Grummer-Strawn, L. M. (2005). Breastfeeding rates in the United States by characteristics of the child, mother, or family: the 2002 National Immunization Survey. Pediatrics, 115(1), e31-37. This article reviews breastfeeding rates, subdivided by demographics, in the US. The data originates from the National Immunization Survey (NIS), which included questions on breastfeeding for the first time in 2001. Although breastfeeding initiation occurred among 71.4% of infants (a relatively high rate close to national goals), exclusive breastfeeding rates at 3 4 and 6 months were 42.5% and 13.3%, respectively. These rates are significantly below stated goals. The authors found that rates were lowest among blacks, those in day care or on WIC, lower socioeconomic groups, and in some geographic regions. Overall, this article provides good evidence for the need for breastfeeding promotion initiatives, especially among risk- groups. Sikorski, J., Renfrew, M. J., Pindoria, S., & Wade, A. (2003). Support for breastfeeding mothers: a systematic review. Paediatr Perinat Epidemiol, 17(4), 407-417. This article provides a comprehensive review of the literature on the effects of lay and professional breastfeeding support on duration of breastfeeding. The authors concluded that support had a larger positive effect on duration of exclusive breastfeeding than on duration of partial breastfeeding. However, when support was divided into professional and lay, professional support only influenced duration of partial breastfeeding while lay support only influenced duration of exclusive breastfeeding. These results indicate that further studies are still needed to fully understand the precise benefits of added support. Overall, though, support seems to be beneficial and should be offered more widely. The National Breastfeeding Committee Denmark. (2004). A Snapshot 2002. Unpublished data compiled by the National Breastfeeding Committee in Copenhagen, Denmark. This document provides graphical data on women’s interactions with health care professionals in respect to breastfeeding. 458 Danish women took part in the survey. . The National Breastfeeding Committee Denmark. (n.d.). [Summary data]. An unpublished report of the breastfeeding situation in Denmark compiled by the National Breastfeeding Committee in Copenhagen, Denmark. This document provides pertinent information on breastfeeding attitudes and rates in Denmark. It also assesses the Baby Friendly Hospital Initiative and other on going policy efforts to improve breastfeeding rates in Denmark. Thus, this paper provides crucial background information for my analyses. WHO. (1998). Evidence for the ten steps to successful breastfeeding, Geneva, Switzerland (No. 98.9). This document presents ten steps supporting breastfeeding that hospitals and other health facilities providing maternity care should follow. These steps are the key elements of the Baby Friendly Hospital Initiative. If hospitals adequately follow these steps, then the WHO will certify them as Baby Friendly. This report carefully presents each of the ten steps, reviews literature supporting the importance of each step, and then draws conclusions offering practical advice on achieving the particular step. These ten steps are an excellent example of practical and realistic policies that can be used to promote breastfeeding. This policy especially focuses on breastfeeding initiation while in the hospital, with the expectation that higher rates of successful initiation will lead to longer duration. 5 Yngve, A., & Sjostrom, M. (2001). Breastfeeding in countries of the European Union and EFTA: current and proposed recommendations, rationale, prevalence, duration and trends. Public Health Nutr, 4(2B), 631-645. This article is a useful EU specific source that delineates trends and differences in breastfeeding among European countries. Notably, the authors point out that most studies have focused on any breastfeeding, while evidence indicates that exclusive breastfeeding is the most beneficial type. Generally, they emphasize the need for more comprehensive and comparable studies on a wide array of factors influencing breastfeeding. They also call for a European breastfeeding conference to initiate action. The article does include some Denmark specific information and statistics. Postpartum depression Berle, J. O., Aarre, T. F., Mykletun, A., Dahl, A. A., & Holsten, F. (2003). Screening for postnatal depression. Validation of the Norwegian version of the Edinburgh Postnatal Depression Scale, and assessment of risk factors for postnatal depression. J Affect Disord, 76(1-3), 151-156. The authors of this study assessed the usefulness of the Edinburgh Postnatal Depression Scale (EPDS) in Norway in a sample of 411 women. The information contained in this paper was very useful to me when I was calculating an appropriate cut point for my depressive symptoms scale. Here, the authors found the scale was valid (through both specificity and sensitivity), which is also important information for my study. Since my scale was very close to the EPDS, if the EPDS was not a valid measure in Scandinavian populations, my scale might not be either. Ten percent of this study population qualified as having major and minor depression, which corroborates other findings that the prevalence of postpartum depression in Western populations is 10-15%. Carter, F. A., Carter, J. D., Luty, S. E., Wilson, D. A., Frampton, C. M., & Joyce, P. R. (2005). Screening and treatment for depression during pregnancy: a cautionary note. Aust N Z J Psychiatry, 39(4), 255-261. This New Zealand study used the EPDS to screen pregnant women for depression. Researchers used a cut point very similar to that used in the Berle et al study, which further validates the appropriateness of the general cut off (while reminding that slight variations do occur between countries). The authors found that most women were willing to be screened with the EPDS, but most were unwilling to participate in further assessments or treatments. These findings emphasize the difficulties associated with identifying and treating women with postpartum depression. Stigma may be a major obstacle here. Thus, this article indicates that personalized screening through a trusted health professional may be necessary to adequately intervene in postpartum depression. This finding should help guide policy. 6 Eberhard-Gran, M., Eskild, A., Tambs, K., Schei, B., & Opjordsmoen, S. (2001). The Edinburgh Postnatal Depression Scale: validation in a Norwegian community sample. Nord J Psychiatry, 55(2), 113-117. Like the Berle et al study, this one also sought to validate the EPDS in Norway. The authors concluded that a cut point of ≥ 10 was appropriate. This cut off is one point lower than in the Berle et al study. However, researchers here were utilizing the scale in a non-clinical setting, for which lower cut points are usually appropriate. The authors also point out that EPDS is superior to regular depression scales when screening for postpartum depression because new mothers often experience fatigue, lack of sleep, and changes in appetite. All of these symptoms are characterized as signs of regular depression, but can lead to misdiagnosis in new mothers. Edhborg, M., Friberg, M., Lundh, W., & Widstrom, A. M. (2005). "Struggling with life": narratives from women with signs of postpartum depression. Scand J Public Health, 33(4), 261-267. This Swedish study provides a qualitative analysis of the problems women with postpartum depression encounter. Since most of my references utilize quantitative statistical analyses, this study is very useful to me. It provides insight into the thoughts and feelings of new mothers. This research is doubly useful since it occurred in a Scandinavian population. The authors interviewed 22 women with signs of postpartum depression, which was identified using the EPDS. Many women reported breastfeeding problems, which they indicated made them feel like failures. Breastfeeding is very much a norm in Scandinavia so women there likely experience much greater social pressure to breastfeed than do women in the US. Also, this paper provides evidence that stigma does exist in Scandinavia in respect to postpartum depression. Falceto, O. G., Giugliani, E. R., & Fernandes, C. L. (2004). Influence of parental mental health on early termination of breast-feeding: a case-control study. J Am Board Fam Pract, 17(3), 173-183. This study occurred in Brazil, which weakens its applicability to my study in Denmark. However, the authors did find that mental health disorders in the first month postpartum were correlated with early termination of breastfeeding. Depression was only one of the disorders included in the variable “mental health disorder,” and the researchers screen for depression using the Self-Reporting Questionnaire, not EPDS. Georgiopoulos, A. M., Bryan, T. L., Wollan, P., & Yawn, B. P. (2001). Routine screening for postpartum depression. J Fam Pract, 50(2), 117-122. This study evaluated the long-term usefulness of the EPDS as a screening tool in a US community. Other studies cited have not looked at the scale’s long-term usefulness, so this study adds important information. The authors found that postpartum depression was diagnosed at a rate of 3.7% before the screening tool intervention and 10.7% afterwards. Given the evidence on the prevalence of postpartum depression, the latter percentage is probably much closer to the true population rate. Of women who qualified as having depressive symptoms based on their responses to the EPDS, 35% were clinically diagnosed with depression. Only 5% of those who 7 did not qualify through EPDS were diagnosed with depression. Thus, high EPDS scores predicted depression well. Overall, EPDS seems to be a useful tool in this community, which implies that it should be considered as an effective policy tool in other communities as well. Henderson, J. J., Evans, S. F., Straton, J. A., Priest, S. R., & Hagan, R. (2003). Impact of postnatal depression on breastfeeding duration. Birth, 30(3), 175-180. This study followed 1745 women for 12 months postpartum. The focus of this study is the same as my main focus so their findings are extremely pertinent. The authors did differentiate between partial and exclusive breastfeeding. They found that the onset of postpartum depression occurred before two months postpartum for 63% of the women who experienced it. This finding implies that my sample may have problems with recall bias since my subjects were asked at six months postpartum if they had experienced symptoms anytime during the past six months. The authors also found that postpartum depression and early breastfeeding cessation were significantly correlated. Mezzacappa, E. S. (2004). Breastfeeding and maternal stress response and health. Nutr Rev, 62(7 Pt 1), 261-268. This review article focuses on the physiological effects of breastfeeding on the mother. Evidence suggests that breastfeeding may have a positive effect on the neuroendocrine system and actually reduce the experience of depressive symptoms. Thus, the relationship between breastfeeding and postpartum depression may be circular. This finding is an important consideration when interpreting the results of my associations. Nishizono-Maher, A., Kishimoto, J., Yoshida, H., Urayama, K., Miyato, M., Otsuka, Y., et al. (2004). The role of self-report questionnaire in the screening of postnatal depression- a community sample survey in central Tokyo. Soc Psychiatry Psychiatr Epidemiol, 39(3), 185-190. The authors attempt to establish the usefulness of the EPDS as a screening tool in Japan. Because of significant cultural differences, their findings may not be directly applicable for Denmark. However, they did find that 13.9% of new mothers were high scorers on the scale, which compares to the rates seen in Scandinavian studies. This article especially focuses on the issue of stigma, which creates a need for a reliable, standard screening device. Otherwise, postpartum depression will continue to be under diagnosed, which will propagate stigma in a cyclical manner. Robertson, E., Grace, S., Wallington, T., & Stewart, D. E. (2004). Antenatal risk factors for postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry, 26(4), 289- 295. This article does not directly address breastfeeding but instead reviews the literature on factors leading to postpartum depression. This subject is important since decreasing the prevalence of postpartum depression in the first place would decrease its opportunity to negatively affect duration of breastfeeding. In a meta-analysis of past data, the authors found that past 8 experiences with depression, anxiety or stress during pregnancy, and low levels of social support all predicted postpartum depression. The authors also note that more research is needed on the appropriateness of different screening tools in different populations. Clearly, this topic is important since postpartum depression cannot be treated if it is not diagnosed. Health professionals must make sure that they are using a tool that is useful within their cultural context. Seimyr, L., Edhborg, M., Lundh, W., & Sjogren, B. (2004). In the shadow of maternal depressed mood: experiences of parenthood during the first year after childbirth. J Psychosom Obstet Gynaecol, 25(1), 23-34. This Swedish study utilized the EPDS to assess postpartum depression in first year mothers. The authors used a cut point of 10, which is identical to that used in the Norwegian Eberhard-Gran et al study and one off from the Norwegian Berle et al study. Thus, the EPDS seems to be a rather consistent tool within Scandinavian populations. This information allows me to more confidently extend it to a Danish population. The authors did find that maternal depressive symptoms decreased duration of breastfeeding, which corroborates my findings. Taj, R., & Sikander, K. S. (2003). Effects of maternal depression on breast-feeding. J Pak Med Assoc, 53(1), 8-11. This study of a Pakistani population asserts that breastfeeding may have positive effects on mental health. The applicability of the findings to a Danish population is questionable because of significant cultural differences. The authors did find that women with depressive symptoms stopped breastfeeding sooner. However, they note that the symptoms proceeded cessation of breastfeeding, which indicates that depression probably led to ceasing breastfeeding, not vice versa. Terp, I. M., Engholm, G., Moller, H., & Mortensen, P. B. (1999). A follow-up study of postpartum psychoses: prognosis and risk factors for readmission. Acta Psychiatr Scand, 100(1), 40-46. This study does not directly examine breastfeeding but rather examines the negative effects of and risk factors for postpartum psychoses, a more serious form of postpartum depression. The information in this paper is useful because understanding risk factors for mental health disorders can help decrease their prevalence, which could help increase the prevalence of breastfeeding in the maternal population. Also, the subjects in this study are Danish, which makes the results very applicable to my study population, especially in respect to important cultural factors affecting mental health. Psychosocial factors Cernadas, J. M., Noceda, G., Barrera, L., Martinez, A. M., & Garsd, A. (2003). Maternal and perinatal factors influencing the duration of exclusive breastfeeding during the first 6 months of life. J Hum Lact, 19(2), 136-144. 9 The authors conducted this mid-sized study (537 women) in Argentina. The non-Western location of this study limits its applicability to my study in Denmark. This study also focused on only the first 6 months postpartum whereas the DNBC collected data through 18 months. However, exclusive breastfeeding, their outcome, only occurs for a maximum of 6 months. Thus, the duration of their study was adequate for their purpose. The authors found that positive maternal attitudes towards breastfeeding, adequate family support, good mother-infant bonding, appropriate suckling technique and the absence of nipple problems all led to longer duration of exclusive breastfeeding. They did control for biological and demographic variables, which strengths the validity of these findings. DiGirolamo, A., Thompson, N., Martorell, R., Fein, S., & Grummer-Strawn, L. (2005). Intention or experience? Predictors of continued breastfeeding. Health Educ Behav, 32(2), 208- 226. The authors focused on the relationship between prenatal intention to breastfeed and actual duration of breastfeeding among 1665 US women. Although it seems intuitive that women who intend to breastfeed would be more likely to actually succeed, it is still important to scientifically establish that a relationship does exist. The authors did find the expected relationship but also found that initial breastfeeding experiences modify the relationship. This factor is just one of many that contributes to duration of breastfeeding. Kirkland, V. L., & Fein, S. B. (2003). Characterizing reasons for breastfeeding cessation throughout the first year postpartum using the construct of thriving. J Hum Lact, 19(3), 278-285. The authors examined 758 mothers from the US Food and Drug Administration’s Infant Feeding Practices Study (IFPS). Their chosen focus is based on Orem’s theory that nutritional, psychological, and lifestyle demands can “overwhelm a mother’s physical and/or mental capacity to care for herself and her dependents” (p. 279). They found that most women ceased breastfeeding because of lifestyle factors (wanting to leave infant or wanting to share feeding duties) and personal perceptions (milk was insufficient nutritionally, infant was too old for breastfeeding). This study provided interesting information on the mother’s direct perception of her reasons for breastfeeding cessation. These perceptions may not always include all factors that led to cessation. However, the authors failed to differentiate between exclusive (only breast milk plus water) and partial (breast milk plus any other substances) breastfeeding as outcomes. Kloeblen-Tarver, A. S., Thompson, N. J., & Miner, K. R. (2002). Intent to breast-feed: the impact of attitudes, norms, parity, and experience. Am J Health Behav, 26(3), 182-187. This study examined a low-income cohort of 963 women. The focus on low-income women makes the findings less directly applicable to the women in the Danish National Birth Cohort (DNBC), most of whom are of middle to upper SES. Researchers found that attitudes were most predictive of breastfeeding intent. Previous experience was also important for multiparous women (those with previous children). This study focuses on intention to breastfeed rather than duration (my focus). Theoretically, different variables contribute to intent and duration, but some overlapping causative factors probably exist. 10 Kronborg, H., & Vaeth, M. (2004). The influence of psychosocial factors on the duration of breastfeeding. Scand J Public Health, 32(3), 210-216. The authors studied a cohort of 471 Danish women for four months. Because the subjects are Danish, this study is especially relevant to my research. The researchers used survival analysis to assess the significance of their variables because most (59%) of women were still exclusively breastfeeding at the conclusion of the study. They did not evaluate partial breastfeeding as an outcome. Analyses showed that increased maternal self-efficacy, confidence, knowledge, intent, and prior experience in respect to breastfeeding all lead to increased duration of breastfeeding. Knowledge was only significant among primiparous (first time) mothers. Increased maternal education also had a positive effect on breastfeeding duration. Importantly, over half (51%) of women who ceased breastfeeding did so during the first five weeks postpartum. Researchers and clinicians should be especially cognizant of this increased risk in the early postpartum period. Lande, B., Andersen, L. F., Baerug, A., Trygg, K. U., Lund-Larsen, K., Veierod, M. B., et al. (2003). Infant feeding practices and associated factors in the first six months of life: the Norwegian infant nutrition survey. Acta Paediatr, 92(2), 152-161. This study assessed a range of psychosocial factors associated with duration of breastfeeding in a cohort of 2383 Norwegian infants. These results are directly pertinent to my study since the authors analyze many of the same variables and use a Scandinavian population. The authors do differentiate between partial and exclusive breastfeeding. They found breastfeeding duration was positively associated with maternal education, maternal age, infant gender, and martial status. Rempel, L. A. (2004). Factors influencing the breastfeeding decisions of long-term breastfeeders. J Hum Lact, 20(3), 306-318. This small study used the theory of planned behavior to evaluate planned and actual breastfeeding durations among 80 women in the US. Based on this theory, Rempel assessed the effects of maternal attitudes, social approval, and perceived control on breastfeeding behavior. The author did not differentiate between partial and exclusive breastfeeding. Overall, she found that perceived approval of breastfeeding decreased as the infant aged and that this approval had a strong effect on intended duration at 9 months postpartum. A mother’s perceived control over breastfeeding affected intended duration at both the prenatal and 9-month assessments. The two reasons women most cited for weaning between 9 and 12 months were perceived readiness of the infant to wean and the belief that the infant had received enough or all breastfeeding benefits by this time point. Interesting, women’s reasons for weaning were most related to perceived control rather than perceived social approval. Scott, J. A., & Binns, C. W. (1999). Factors associated with the initiation and duration of breastfeeding: a review of the literature. Breastfeed Rev, 7(1), 5-16. This literature review provides a comprehensive summary and comparison of recent findings on factors affecting breastfeeding decisions. The authors limit their review to findings on Western 11 women and to studies that utilized multivariate analyses, a more informative method than univariate analyses. Overall, they conclude that a wide range of variables contributes to breastfeeding decisions, but many findings in the literature are inconsistent. Demographic factors seem to more consistently affect these decisions than do biomedical ones. Scott and Binns did find a constant inverse relationship between maternal smoking and breastfeeding duration. Studies on psychosocial variables consistently indicate that fathers have an important influence on maternal decisions in relation to breastfeeding and that intended duration strongly predicts actual duration. This later finding is rather intuitive. This article serves as a good reminder that a complex interaction of multiple variables influences breastfeeding decisions. Scott, J. A., Landers, M. C., Hughes, R. M., & Binns, C. W. (2001). Psychosocial factors associated with the abandonment of breastfeeding prior to hospital discharge. J Hum Lact, 17(1), 24-30. This Australian study of 1059 women assessed cessation of breastfeeding within the first week postpartum. Their measurement of breastfeeding included both exclusive and partial. Psychosocial variables affecting the decision to cease breastfeeding may be very different in the first week and at six months, which makes their finding less relevant for me. The researchers found that low commitment levels (deciding to breastfeed after becoming pregnant), lack of social support (by the father), and lack of prior exposure to breastfeeding (through the mother’s mother) were all risk factors for abandoning breastfeeding while still in the hospital. Hospital discharge times varied, which may have confounded these results. However, researchers did follow up with the women for up to six months. The authors found that most women who ceased breastfeeding did so in the first six weeks postpartum, which is similar to what the Kronborg study (see above) found. Shaker, I., Scott, J. A., & Reid, M. (2004). Infant feeding attitudes of expectant parents: breastfeeding and formula feeding. J Adv Nurs, 45(3), 260-268. This Scottish study compared prenatal breastfeeding attitudes of both mother and father to infant feeding method at discharge from the hospital. The inclusion of the father in data collection is very important since previous research has indicated that his opinions and behaviors strongly influence the mother’s breastfeeding decisions (Scott & Binns, 1999). The 108 couples consist of a convenience sample; this method of data collection may have introduced some bias into the results. The authors did not differentiate between exclusive and partial breastfeeding. Parents with breastfed infants at discharge had more positive attitudes toward and greater knowledge of the benefits of breastfeeding. Regardless of feeding method, fathers were more disapproving than mothers of breastfeeding in public. Mothers of formula fed infants were more disapproving of drinking alcohol while breastfeeding. This last finding is important for my study since I found that women drinking alcohol actually breastfed longer. Sullivan, M. L., Leathers, S. J., & Kelley, M. A. (2004). Family characteristics associated with duration of breastfeeding during early infancy among primiparas. J Hum Lact, 20(2), 196-205. 12 This study investigated the effect of gender roles on breastfeeding duration (either exclusive or partial). Unlike the Danish National Birth Cohort subjects, these 115 American women were ethnically diverse. Using survival analyses, the authors found that women with higher levels of relationship distress ceased breastfeeding sooner. Surprisingly, they found that women with higher levels of responsibility for infant care and household tasks continued breastfeeding longer. The authors postulate that this correlation may exist because some tasks are more convenient for breastfeeding women to complete, which naturally leads to a higher level of responsibility for these tasks. This study is one of few to provide evidence that comprehensive aspects of the couple’s relationship (not just paternal support of breastfeeding) influence duration of breastfeeding. Taveras, E. M., Capra, A. M., Braveman, P. A., Jensvold, N. G., Escobar, G. J., & Lieu, T. A. (2003). Clinician support and psychosocial risk factors associated with breastfeeding discontinuation. Pediatrics, 112(1 Pt 1), 108-115. This US study followed 1163 mother-infant dyads for 12 weeks. Eighty-seven percent of women in this cohort initiated breastfeeding, which is much higher than the average rate in the US (64%). This dichotomy indicates that the study subjects were of higher socioeconomic status (SES) than the average American women. Demographic statistics support this inference. Because these women are not representative of the typical American, they are more similar to Danish women and particularly the women in the Danish National Birth Cohort. Researchers found that lower SES was a risk factor for breastfeeding cessation. They also found that maternal depressive symptoms were correlated with increased breastfeeding discontinuation. This finding is very pertinent to my study. Furthermore, the authors found that clinician support is very important to breastfeeding success. Returning to work or school at 12 weeks was one of the strongest indicators of discontinuation, which indicates the importance of maternity leave. These findings identify several areas for policy changes. Taveras, E. M., Li, R., Grummer-Strawn, L., Richardson, M., Marshall, R., Rego, V. H., et al. (2004). Opinions and practices of clinicians associated with continuation of exclusive breastfeeding. Pediatrics, 113(4), e283-290. This American study of 288 mothers provides strong evidence for the importance of clinicians in respect to breastfeeding decisions. The authors investigate only exclusive breastfeeding. Thus, the applicability of these findings to partial breastfeeding is uncertain. Also, the researchers selected women who were breastfeeding at four weeks to participate in the study. Again, this limitation makes it difficult to conclude what the clinician’s role may be in the first month postpartum. Overall, they found that mothers with biomedical problems and those whose clinicians recommended formula use were more likely to cease breastfeeding. Clinicians with confidence in the importance of their advice had a greater positive effect on duration of breastfeeding among their patients. Clinicians also reported having little time to preemptively discuss breastfeeding problems. These findings indicate that policy should advocate increased clinician knowledge about and time to promote breastfeeding. 13 Yang, Q., Wen, S. W., Dubois, L., Chen, Y., Walker, M. C., & Krewski, D. (2004). Determinants of breast-feeding and weaning in Alberta, Canada. J Obstet Gynaecol Can, 26(11), 975-981. This Canadian study assessed an array of variables affecting duration of breastfeeding in 1113 women. The authors did not differentiate between partial and exclusive breastfeeding. They found that marital status, education, maternal smoking, and family income were all important indicators of breastfeeding duration. This study simply provides me with more data supporting the notion that a large number of factors influence breastfeeding duration and that this statement is true worldwide. Biological factors Anderson, G. C., Moore, E., Hepworth, J., & Bergman, N. (2003). Early skin-to-skin contact for mothers and their healthy newborn infants. Cochrane Database Syst Rev(2), CD003519. This review summarizes known information on the relationship between early mother-infant contact and breastfeeding. Most studies have found a positive relationship between the two. This finding establishes the importance of yet another factor in relation to breastfeeding duration. It also provides guidance on design of effective hospital policies that aim to increase breastfeeding rates. Baker, J. L., Michaelsen, K. F., Rasmussen, K. M., & Sorensen, T. I. (2004). Maternal prepregnant body mass index, duration of breastfeeding, and timing of complementary food introduction are associated with infant weight gain. Am J Clin Nutr, 80(6), 1579- 1588. This large study (n=3768) is especially useful because its subjects are also from the Danish National Birth Cohort (DNBC). These researchers utilized a prospective study design, which eliminated problems of retrospective bias. The authors found that decreased durations of breastfeeding, as well as increased maternal body mass index (BMI) and earlier complementary food introduction, led to greater infant weight gain in the first year of life. However, timing of complementary food introduction was only significant when combined with breastfeeding duration of < 20 weeks. Complementary food refers to infant foods other than breast milk or formula. The authors lacked sufficient information to differentiate between exclusive and partial breastfeeding, so they analyzed “any” breastfeeding. Their findings that maternal BMI and duration of breastfeeding both have an independent effect on infant weight gain is especially important given that the Hilson et al study (see below) found that increased maternal BMI led to shorter breastfeeding duration. Thus, infants with mothers with high BMI are likely at increased risk of greater weight gain through both mechanisms (maternal BMI and breastfeeding duration). Dewey, K. G., Nommsen-Rivers, L. A., Heinig, M. J., & Cohen, R. J. (2003). Risk factors for suboptimal infant breastfeeding behavior, delayed onset of lactation, and excess neonatal weight loss. Pediatrics, 112(3 Pt 1), 607-619. 14 This study focuses on a range of biological factors that lead to decreased rates of breastfeeding. The authors examine a mother’s success maintaining exclusive breastfeeding as their outcome. In a cohort of 280 women, they find that maternal body mass index, primiparity (being a first time mother), and delivery through cesarean section all increased the risk for breastfeeding cessation. These findings are useful because they provide further evidence that a variety of factors contribute to breastfeeding success and indicate that biology is at least one important aspect of the process. Hilson, J. A., Rasmussen, K. M., & Kjolhede, C. L. (2004). High prepregnant body mass index is associated with poor lactation outcomes among white, rural women independent of psychosocial and demographic correlates. J Hum Lact, 20(1), 18-29. This study, conducted in upstate New York, evaluated obesity as an indicator for problems initiating and continuing breastfeeding among 151 women. This study contained a relatively large percentage (~ 40%) of women of low socioeconomic status. The researchers found that obese women had fewer vaginal deliveries and entered lactogenesis II, the onset of high volume milk production, later than normal. Elevated levels of progesterone, produced by excess adipose tissue, may be the mechanism for this change. Notably, they found an even stronger relationship between primiparity (first time mothers) and the late onset of lactogenesis II. This study provides good evidence for the need for lactation support programs oriented toward first time and obese mothers. Obesity as a risk factor is more relevant to the US than to Denmark, which has lower rates of overweight and obesity in its population. Nielsen, G. A., Thomsen, B. L., & Michaelsen, K. F. (1998). Influence of breastfeeding and complementary food on growth between 5 and 10 months. Acta Paediatr, 87(9), 911-917. This Danish study followed 339 infants through their first 10 months of life. Researchers found that infants breastfed for ≥ 7 months gained less weight and length. These infants also received smaller amounts of cow’s milk, meat, and sweets (all complementary foods). These findings reinforce the findings of Baker et al (described above). The authors emphasize that these results should not be used to discourage long-term breastfeeding as the long-term effects of infant weight gain and growth in the first year of life are not fully understood. Evidence indicates that lower weight gain and growth may actually be a positive outcome. Infants with the highest protein intake (indirectly, infants consuming the most meat) grew the most in the 10 months. These researchers do not overtly differentiate between partial and exclusive breastfeeding, but since they measure complementary food intake, they do assess this difference indirectly. Patel, R. R., Liebling, R. E., & Murphy, D. J. (2003). Effect of operative delivery in the second stage of labor on breastfeeding success. Birth, 30(4), 255-260. Researchers debate the effect of delivery by cesarean section on duration of breastfeeding. The authors analyze this specific question, which hopefully adds important information to help resolve the debate. They assess the effects of mode of delivery in 393 ]women from the UK and use duration of exclusive breastfeeding as their outcome. Overall, the authors find no difference in duration of breastfeeding between women who delivered vaginally and by cesarean section. They do find that women who delivered by c-section and stayed in the hospital longer were more 15 successful breastfeeders. This finding makes sense; presumably these women received extra support and instruction from hospital staff. Thus, this paper is useful because it indicates that hospital actions and support may be more important than mode of delivery. Alcohol Albertsen, K., Andersen, A. M., Olsen, J., & Gronbaek, M. (2004). Alcohol consumption during pregnancy and the risk of preterm delivery. Am J Epidemiol, 159(2), 155-161. This study utilized the Danish National Birth Cohort, and thus was mainly useful to me when I was making decisions on coding variables. The authors characterized alcohol consumption in the following categories: nondrinkers, one-half drink/wk, 1-1.5 drinks/wk, 2-3.5 drinks/wk, 4-6.5 drinks/wk, ≥ 7 drinks/wk. I utilized these categorizations initially, but used a different definition of alcohol consumption (yes or no) in my final analyses. I did base my categorization of type of alcohol on their definition: “preference for one type of alcohol was defined as beer, wine, or spirits if the intake of this type exceeded 50 percent of the woman’s total alcohol intake” (p.156). The authors found that wine drinkers were less likely to smoke and more likely to hold professional level occupations, both of which corroborate my findings. Goodwin, D. W., Gabrielli, W. F., Jr., Penick, E. C., Nickel, E. J., Chhibber, S., Knop, J., et al. (1999). Breast-feeding and alcoholism: the Trotter hypothesis. Am J Psychiatry, 156(4), 650-652. The authors evaluate the Trotter hypothesis, developed in the early 1800s, which posited that early weaning led to alcoholism later in life. Subjects are 200 Danish infants from circa 1960, who were followed for 30 years. Overall, they found that early weaning is a risk factor for alcoholism in adulthood. Some evidence suggests that this correlation is related to the formation of dopaminergenic mechanisms, which are important in addiction development. Unfortunately, they lacked information on the mothers’ alcohol consumption status during pregnancy. This study is useful because it focuses on a Danish population and provides further evidence about the detrimental effects of attenuated breastfeeding duration. Gronbaek, M. (2001). Factors influencing the relation between alcohol and mortality--with focus on wine. J Intern Med, 250(4), 291-308. This article reviews the literature on alcohol to establish more clearly its effects on health. Generally, most studies have found a J-shaped relationship between alcohol and mortality, which suggests beneficial effects of alcohol in small doses. Some evidence indicates that wine confers added benefits not found in other forms of alcohol. This article provides useful background evidence for the evaluation of maternal consumption of alcohol during breastfeeding within my study population, which has relatively high SES and predominately consumes wine. Gronbaek, M., Di Castelnuovo, A., Iacoviello, L., Furman, K., Donati, M. B., de Gaetano, G., et al. (2004). Wine, alcohol and cardiovascular risk: open issue. J Thromb Haemost, 2(11), 2041-2048. 16 This article argues that alcohol does have a protective effect against cardiovascular disease but that the strength of this effect may vary by age, gender, and method of consumption. Furthermore, the authors present evidence that wine may be the most beneficial type of alcohol and discuss some possible mechanisms for this relationship. Some confounding factors include lifestyle (including diet) and genetics. As in the other Gronbaek article, this article is useful as background information for evaluating the implications of alcohol consumption in breastfeeding women. Gunzerath, L., Faden, V., Zakhari, S., & Warren, K. (2004). National Institute on Alcohol Abuse and Alcoholism report on moderate drinking. Alcohol Clin Exp Res, 28(6), 829-847. This report presents evidence on the risks and benefits associated with alcohol consumption in different areas, one of which is breastfeeding. Much of the literature they review on alcohol and breastfeeding come from Mennella. I discuss several of her specific studies below. Generally, the authors here conclude that alcohol does not aid breastfeeding, as commonly thought, and that mothers can reduce infant exposure to alcohol by waiting for a couple hours after alcohol consumption to breastfeed. This source provides a useful summary of the literature in respect to alcohol and breastfeeding. Ho, E., Collantes, A., Kapur, B. M., Moretti, M., & Koren, G. (2001). Alcohol and breast feeding: calculation of time to zero level in milk. Biol Neonate, 80(3), 219-222. This paper calculates rates of alcohol elimination in women of different body weights to help breastfeeding women avoid exposing their infants to alcohol. Practical information such as elimination rates is important since infants take twice as long as adults to eliminate alcohol and the long-term effects of alcohol exposure in infants are unknown. Information presented in this paper could be used to help develop practical policies and recommendations on alcohol consumption in breastfeeding women. Little, R. E., Northstone, K., & Golding, J. (2002). Alcohol, breastfeeding, and development at 18 months. Pediatrics, 109(5), E72-72. This study of 915 one year olds in the UK provides some evidence on short-term effects of alcohol exposure on infant development. The authors did not find any developmental delays in the study population. This negative finding may make sense if alcohol has small effects on development or if these effects do not appear until later in life. The authors were trying to replicate another study that did find motor deficits in one year olds exposed to alcohol. Thus, the exact mechanisms and effects of alcohol exposure are still poorly understood. This paper provides evidence that more research is needed before developing meaningful recommendations for breastfeeding women. Mennella, J. (2001). Alcohol's effect on lactation. Alcohol Res Health, 25(3), 230-234. In this review Mennella asserts that contrary to popular belief, alcohol consumption may reduce milk production. She presents evidence that infants exposed to alcohol may experience adverse sleep effects and delays in gross motor development. Exposed infants may also develop early 17 positive emotional associations with alcohol. All of these findings indicate that breastfeeding women should be cautioned against alcohol consumption. However, especially in the context of my research, these findings must be balanced with social norms. Mennella, J. A. (2001). Regulation of milk intake after exposure to alcohol in mothers' milk. Alcohol Clin Exp Res, 25(4), 590-593. This study compared infant milk consumption after the mother consumed alcohol-containing orange juice and regular orange juice in 12 dyads. After maternal alcohol consumption, infant milk consumption decreased 20%. However, the infants made up for this deficit over the next 8- 12 hours. Outcomes for chronically alcohol-exposed infants are uncertain although some evidence suggests that weight differences do not exist. This study presents important evidence that alcohol consumption might be disruptive for breastfeeding. More evidence is needed for conclusive recommendations. Mennella, J. A., & Gerrish, C. J. (1998). Effects of exposure to alcohol in mother's milk on infant sleep. Pediatrics, 101(5), E2. In this study of 13 dyads, infant’s behaviors were monitored in respect to their exposure to alcohol through breast milk. Infants were tested two different days. Each received regular breast milk once and alcohol-containing breast milk once. This study is unusual in that it uses experimental design to test effects of alcohol on infants. Ethical implications make it difficult carry out a study design such as this one, but experimental studies often produce clearer findings. The authors found that alcohol exposure decreases active sleep in the short term but that infants make up for this deficit in the long run. Effects of chronic alcohol exposure are uncertain. These findings provide a possible explanation for delays in gross motor development among infants routinely exposed to alcohol (active sleep may be necessary for motor development). This study provides cautionary evidence against the consumption of alcohol in breastfeeding women. Mukherjee, R. A., Hollins, S., Abou-Saleh, M. T., & Turk, J. (2005). Low level alcohol consumption and the fetus. Bmj, 330(7488), 375-376. This article argues that the only safe message in respect to alcohol consumption during pregnancy is one of abstinence since the exact safe dose of ethanol is unknown. The authors point out that differences in individual alcohol metabolism probably exist. This position is very pertinent to Denmark, given that approximately half of Danish women consume alcohol during breastfeeding. Alcohol seems to have more detrimental effects on the fetus than on infants. Nielsen, N. R., Schnohr, P., Jensen, G., & Gronbaek, M. (2004). Is the relationship between type of alcohol and mortality influenced by socio-economic status? J Intern Med, 255(2), 280- 288. This Danish study examines mortality in relation to alcohol consumption in a cohort of 14223 individuals. The authors find that individuals consuming wine have higher levels of education and income, which corroborates demographic findings in my study and provides added evidence that alcohol consumption acts as an aggregate SES measure. Furthermore, they find that wine 18 has the strongest protective effect in lower SES groups. This finding may have important implications for the interpretation of my results, especially given that my subjects are of higher SES than is representative of the general Danish population. Thackray, H., & Tifft, C. (2001). Fetal alcohol syndrome. Pediatr Rev, 22(2), 47-55. This article provides good basic background information on the causes and characteristics of fetal alcohol syndrome. It also briefly discusses the effects of alcohol on breastfeeding. Interestingly, the authors note that alcohol inhibits prolactin and oxytocin release during breastfeeding. Both these hormones produce a relaxing effect and are crucial to the success of breastfeeding. Alcohol is also thought to be relaxing and thus conducive to breastfeeding. However, this hormonal evidence suggests otherwise. This article provides good evidence against alcohol consumption during pregnancy and questions the wisdom of it during breastfeeding. Smoking Agostoni, C., Marangoni, F., Grandi, F., Lammardo, A. M., Giovannini, M., Riva, E., et al. (2003). Earlier smoking habits are associated with higher serum lipids and lower milk fat and polyunsaturated fatty acid content in the first 6 months of lactation. Eur J Clin Nutr, 57(11), 1466-1472. This study analyzed the composition of breast milk among 92 Italian women who smoke. The authors found that smoking changes fat content of breast milk. Notably, DHA levels were lower in the milk from smoking mothers. DHA is an important fatty acid integral in brain development. These results hold even if the smoking occurred earlier in pregnancy. This research provides important evidence against smoking during pregnancy and breastfeeding and implies that this behavior may have doubly negative effects on breastfeeding: changing composition of milk on top of decreasing duration of breastfeeding (evidenced from my study and others). Batstra, L., Neeleman, J., & Hadders-Algra, M. (2003). Can breast feeding modify the adverse effects of smoking during pregnancy on the child's cognitive development? J Epidemiol Community Health, 57(6), 403-404. This study presents an interesting dichotomy: even though smoking is detrimental to composition of breast milk and duration of breastfeeding, breast milk may decrease other negative effects of smoking that the infant experiences (such as brain development). The authors studied 3162 children from the Netherlands and assessed their cognitive development through their performance on reading, spelling, and math tests given at 9 years of age. Maternal smoking only had a negative effect on children who had been bottle-fed. Genetics may play a confounding role. This study provides encouraging evidence that breastfeeding can protect against negative effects of maternal smoking. However, these findings may be difficult to translate into practical interventions since women who smoke are less likely to successfully breastfeed. 19 Dahlstrom, A., Ebersjo, C., & Lundell, B. (2004). Nicotine exposure in breastfed infants. Acta Paediatr, 93(6), 810-816. This Swedish study examined levels of nicotine exposure, through breast milk and infant urine samples, in 40 infants. All infants were exclusively breastfed. The results of this study are particularly relevant to my study since the subjects are Scandinavian. Infants whose mothers smoked, took snuff, or were around second hand smoke had nicotine in their systems. The authors make the pertinent observation that evidence is lacking on what a ‘safe’ dose of nicotine is for infants and on how they metabolize nicotine. These findings could lead to conflicting advice. From one perspective, breastfeeding may increase infants’ exposure to nicotine (through breast milk and air). However, Batstra et al find that breastfeeding protects against negative long-term effects from nicotine exposure. Clearly, the safest recommendation is for mothers to avoid smoking, or other forms of nicotine, completely. Ilett, K. F., Hale, T. W., Page-Sharp, M., Kristensen, J. H., Kohan, R., & Hackett, L. P. (2003). Use of nicotine patches in breast-feeding mothers: transfer of nicotine and cotinine into human milk. Clin Pharmacol Ther, 74(6), 516-524. This study offers a practical solution for smoking, breastfeeding mothers who desire to reduce their infants’ exposure to nicotine. In a population of 15 women, the authors found that 14 mg/d or 7 mg/d patches significantly reduced nicotine levels in their milk (70% lower in those using 7 mg/d patches). Thus, this study provides practical evidence that can lead to the development of realistic recommendations to improve maternal and infant health. Recommending nicotine patches is probably more effective than is advocating total smoking cessation. Sondergaard, C., Henriksen, T. B., Obel, C., & Wisborg, K. (2001). Smoking during pregnancy and infantile colic. Pediatrics, 108(2), 342-346. This Danish study of 1820 mothers is directly pertinent to my study population. The authors found that smoking during pregnancy led to higher risk of infant colic. Breastfeeding did not modify this relationship. Thus, this research provides evidence on the detrimental effects of smoking but fails to add to evidence on the benefits of breastfeeding. Also, this study would be more useful for me if the authors had evaluated smoking during breastfeeding instead of during pregnancy. Medications Aljazaf, K., Hale, T. W., Ilett, K. F., Hartmann, P. E., Mitoulas, L. R., Kristensen, J. H., et al. (2003). Pseudoephedrine: effects on milk production in women and estimation of infant exposure via breastmilk. Br J Clin Pharmacol, 56(1), 18-24. This study used a small sample of women (n = 8) to determine detrimental effects of pseudoephedrine (cold medication) on breast milk. The authors found that milk production decreased 24% after giving the mothers the medications. They propose that the mechanism for the change may be decreased prolactin (hormone) production. Infants were exposed to 4.3% of 20 the maternal dose through breast milk. This study is useful because it helps elucidate the specific effects of one medication, which in turn can help guide policy and practice. Cold medication is quite commonly used, so it is important to understand its effects on both mother and infant in respect to duration of breastfeeding. Chaves, R. G., & Lamounier, J. A. (2004). [Breastfeeding and maternal medications]. J Pediatr (Rio J), 80(5 Suppl), S189-198. This review provides a helpful summary of the studies done of various medications and breastfeeding from 1993 to 2004. Generally, the effects of medications on breastfeeding are poorly understood, especially because it is unethical to do controlled studies in this area. The paper provides useful tables that list specific drugs and their compatibility levels with breastfeeding. In respect to mental health drugs, many of the effects on either breast milk production or the infant are unknown but potentially concerning. This information is especially pertinent because of the high prevalence of postpartum depression. Ideally, physicians can simultaneously treat postpartum depression (with medications, if necessary) and encourage breastfeeding without compromising the health of either mother or child. If they must choose either breastfeeding or medicating for postpartum depression, this scientific evidence is important to help physicians and mothers understand the costs and benefits of each. Hallberg, P., & Sjoblom, V. (2005). The use of selective serotonin reuptake inhibitors during pregnancy and breast-feeding: a review and clinical aspects. J Clin Psychopharmacol, 25(1), 59-73. This Swedish study provides evidence for a postpartum depression prevalence of 10-16% in Scandinavian populations. Only limited evidence is available on the safety of serotonion reuptake inhibitors (SSRIs) during pregnancy and breastfeeding, so this paper reviews both animal and human studies to provide a comprehensive view of the evidence. The authors provide a useful overview table of all the results. Overall, they conclude that risks associated with SSRIs are low and possibly lower than those associated with tricyclic antidepressants. However, they point out that generally more studies and information are needed. As noted in the annotation for Chaves and Lamounier, researchers must strive to understand the effects of depression medications on women, infants, and breast milk because of the high prevalence of depression in this population. Hendrick, V., Smith, L. M., Hwang, S., Altshuler, L. L., & Haynes, D. (2003). Weight gain in breastfed infants of mothers taking antidepressant medications. J Clin Psychiatry, 64(4), 410-412. This study of 78 breastfeeding mothers examines whether depression medications affect infant weight. They found that maternal antidepressant use was not related to infant weight but that infants with mothers suffering from long-term depression weighed significantly less. Thus, this study provides an important indication that postpartum depression has negative effects on infant health aside from duration of breastfeeding. The findings also imply that depressed women can safely take antidepressants and breastfeed simultaneously. 21 Howard, L., Webb, R., & Abel, K. (2004). Safety of antipsychotic drugs for pregnant and breastfeeding women with non-affective psychosis. Bmj, 329(7472), 933-934. This review does little to add to my knowledge on the effects on psychiatric medications during breastfeeding. The authors do advocate large cohort studies since randomized trials are unethical. They also make the useful point that new drugs are emerging rather quickly, which makes it difficult to adequately assess all of them in respect to breastfeeding. Most infants exposed to antipsychotic medications have not shown adverse effects. Ilett, K. F., Kristensen, J. H., Hackett, L. P., Paech, M., Kohan, R., & Rampono, J. (2002). Distribution of venlafaxine and its O-desmethyl metabolite in human milk and their effects in breastfed infants. Br J Clin Pharmacol, 53(1), 17-22. This study focuses on one drug only, Venlafaxine, a novel bicyclic phenylethylamine antidepressant. This drug seems to avoid side-effects of both tricyclic antidepressants and SSRIs. The authors performed a small study, using only six women, all of whom were taking Venlafaxine. The results indicate that use of the drug while breastfeeding is safe. The authors found no adverse effects in the infants but did recommend careful monitoring of the drug levels in infants, especially very young ones that might metabolize the drug more slowly. They do not assess the effectiveness of Venlafaxine, compared to other antidepressants, in treating postpartum depression. Ito, S., & Lee, A. (2003). Drug excretion into breast milk--overview. Adv Drug Deliv Rev, 55(5), 617-627. This review paper, like Chaves and Lamounier, summarizes knowledge on effects of drugs on breastfeeding. However, the authors here also summarize the benefits of breastfeeding, trends in breastfeeding initiation, and prevalence of medication use in breastfeeding women. They state that 90% of women take some sort of medication in the first week postpartum. Clearly then, understanding the effects of medications on breastfeeding is extremely important. They also point out that many women fear exposing their infants to drugs and as a result will either stop breastfeeding or stop taking their medications. This important point emphasizes the need for maternal education in respect to medication use during breastfeeding. The authors also provide a helpful discussion of drug delivery systems and of drug build up in the body. They provide a small reference table of the effects of various drugs in infants. Weissman, A. M., Levy, B. T., Hartz, A. J., Bentler, S., Donohue, M., Ellingrod, V. L., et al. (2004). Pooled analysis of antidepressant levels in lactating mothers, breast milk, and nursing infants. Am J Psychiatry, 161(6), 1066-1078. This review includes studies from as far back as 1979. Not only do the authors report the results found in the studies, they perform their own aggregate analysis on all of the data from them. This method takes advantage of information that can be gained from large numbers of subjects. They conclude that three specific antidepressants – nortriptyline, paroxetine, and sertraline – are the safest for breastfeeding women. Like the other papers, though, they emphasize the need for careful monitoring of infants exposed to antidepressants. The authors reference the infant’s 22 immature hepatic (liver) function during the first few weeks as a possible cause for concern. These infants may be biologically unable to process the medications. This paper cites 13% as the rate of postpartum depression. Breastfeeding policies Ahluwalia, I. B., Tessaro, I., Grummer-Strawn, L. M., MacGowan, C., & Benton-Davis, S. (2000). Georgia's breastfeeding promotion program for low-income women. Pediatrics, 105(6), E85. The authors of this study assess the efficacy of five breastfeeding promotion strategies implemented in Georgia in 1991. These strategies targeted low-income women (those on WIC) and are “1) enhanced breastfeeding, 2) breast pump loans, 3) hospital-based programs, 4) peer counseling, 5) community coalitions” (p. 85). The paper provides a detailed description of each strategy. The authors use a combination of survey data and focus groups to gather both quantitative and qualitative data. This comprehensive approach enhances the usefulness of their findings. Although non-WIC participants breastfed more, they found that breastfeeding initiation rates increased more (from 1992-1996) for those in the WIC program. Women who were younger, had no college education, unmarried, and black experienced the largest increases in initiation. This finding is positive and indicates that the program is successful in reaching its target group. The demographics of the women in this study are quite different from those in the Danish National Birth Cohort; thus, strategies used here may not be useful in a Danish population. However, this study provides useful comparative information. Arthur, C. R., Saenz, R. B., & Replogle, W. H. (2003). The employment-related breastfeeding decisions of physician mothers. J Miss State Med Assoc, 44(12), 383-387. This study assesses the breastfeeding behavior of female physicians in Mississippi. Generally, other studies have found a positive correlation between duration of maternity leave and breastfeeding. Data on this specific profession is important because physicians are highly educated and probably more likely to understand the benefits of breastfeeding than even other professionals. However, they work in a demanding profession that allows little time for maternity leave. The authors find that like women in other professions, physicians who return to work full time wean their infants early. This finding provides compelling evidence for the need for good maternity leave policies. Cattaneo, A., Borgnolo, G., & Simon, G. (2001). Breastfeeding by objectives. Eur J Public Health, 11(4), 397-401. This Italian study assesses the effectiveness of financial penalties in respect to increasing breastfeeding rates. The authors measured exclusive, predominant and complementary breastfeeding; the range of types of breastfeeding included increases the usefulness of the study. In 1998 and 1999 in northeastern Italy, as a part of programs promoting breastfeeding, local health authorities faced financial penalties for not reaching objectives. Results suggest that these 23 penalties were effective. This study poses a relatively novel method of breastfeeding promotion and one that is very feasible. Thus, it adds important information on effective policy. Coffin, C. J., Labbok, M. H., & Belsey, M. (1997). Breastfeeding definitions. Contraception, 55(6), 323-325. This paper is extremely important as it provides standardized definitions of types of breastfeeding. Without the use of standard definitions, research would not be comparable and thus would be relatively useless. The authors visually depict this categorization, which experts devised at a meeting of the Interagency Group of Action on Breastfeeding in 1988. DiGirolamo, A. M., Grummer-Strawn, L. M., & Fein, S. (2001). Maternity care practices: implications for breastfeeding. Birth, 28(2), 94-100. This study analyzes the efficacy of five of the steps in the Baby Friendly Health Initiative (BFHI) in a population of 1085 US women. The outcome is breastfeeding beyond 6 weeks; breastfeeding included any amount of breast milk. This research is important because the BFHI has received much attention as an important initiative and is currently one of the few large-scale efforts to improve breastfeeding rates. Thus, scientists, physicians, and policy makers need to understand the efficacy of the program as well as variations in efficacy among populations. Here, the authors found that few women experienced all five of the steps assessed. Those who experienced none of the steps were eight times more likely to stop breastfeeding before six weeks than those who experienced all five. This study is not directly relevant to a Danish population but provides useful US data for comparison. Galtry, J. (2003). The impact on breastfeeding of labour market policy and practice in Ireland, Sweden, and the USA. Soc Sci Med, 57(1), 167-177. This paper asserts that policy can enable both breastfeeding and women’s participation in the labor market. Few researchers have focused on this topic so far, so the information provided here is quite valuable. Also, because the authors comparatively analyze three different countries, they can assess cultural influences. Overall, lactation programs seem to be cost-effective. Each country utilizes different policies and has different problems they need to overcome. Thus, this paper will be helpful for the comparative aspect of my policy discussion. Hector, D., & King, L. (2005). Interventions to encourage and support breastfeeding. N S W Public Health Bull, 16(3-4), 56-61. This article is quite useful because it analyzes breastfeeding interventions at multiple levels – from individual to societal. Based on a conceptual framework, the authors include individual knowledge, attitudes, and skills, hospital and health services environment, home/family/friends environments, work environment, community environment, and societal norms and expectations. The authors emphasize the complexity of factors that contribute to breastfeeding success. As a result, complex, multi-level interventions are needed. 24 Hofvander, Y. (2003). Why women don't breastfeed: a national survey. Acta Paediatr, 92(11), 1243-1244. Hofvander, a Swedish child health expert, presents several arguments for the lack of breastfeeding on an international level. First, he asserts that the WHO recommendation that women breastfeed exclusively for six months is unrealistic and fails to account for social and economic constraints. He cites year long maternity leaves in Scandinavia as a direct cause of extended breastfeeding durations and then points out that in most countries, maternity leave is a maximum of three months. He also mentions the Baby Friendly Hospital Initiative as an important policy. After its implementation in Sweden, breastfeeding at six months postpartum increased from 50% to 73%. Overall, though, Hofvander emphasizes the multifactorial nature of breastfeeding duration. This paper provides a useful overview of important breastfeeding policy considerations and also includes some Scandinavia specific references. Kovach, A. C. (1996). An assessment tool for evaluating hospital breastfeeding policies and practices. J Hum Lact, 12(1), 41-45. This paper describes a method for evaluating the effectiveness of the Baby Friendly Hospital Initiative in specific hospitals where it has been implemented. Most of the article focuses on describing the actual assessment instrument, which is not directly relevant to my research. However, it is useful to have a background knowledge of how policies are evaluated. Kovach, A. C. (1997). Hospital breastfeeding policies in the Philadelphia area: a comparison with the ten steps to successful breastfeeding. Birth, 24(1), 41-48. This paper analyzes the extent to which BFHI steps have been implemented in 28 hospitals in the Pennsylvania Delaware Valley. Overall, the study provides good evidence that the US needs more aggressive implementation of breastfeeding promotion programs. Most hospitals in the study had only implemented four of ten steps. This paper emphasizes the dramatic difference in average hospital policy between Denmark and the US. Kramer, M. S., Chalmers, B., Hodnett, E. D., Sevkovskaya, Z., Dzikovich, I., Shapiro, S., et al. (2001). Promotion of Breastfeeding Intervention Trial (PROBIT): a randomized trial in the Republic of Belarus. Jama, 285(4), 413-420. This study evaluates the effectiveness of the BFHI in respect to breastfeeding duration and exclusivity as well as infant morbidity. The results may not be directly applicable to a Danish population since the study occurred in an eastern European population. However, the study is unique in its use of a randomized trial design. This randomization of experimental sites leads to results that are more clearly interpretable. Overall, the authors found that infants at Baby Friendly sites had lower morbidity rates and higher breastfeeding rates. Thus, it provides good evidence on the usefulness of the BFHI as a breastfeeding promotion policy. Merewood, A., & Philipp, B. L. (2000). Becoming Baby-Friendly: overcoming the issue of accepting free formula. J Hum Lact, 16(4), 279-282. 25 As part of the BFHI, hospitals must pay fully for formula. This stipulation is based on the economic rationale that hospitals and employees will be more reluctant to offer formula if they have to cover its costs. This paper describes the efforts of Boston Medical Center to comply with this requirement. For hospitals struggling to reject free formula, the authors provide useful information. However, free formula is not as relevant an issue in Denmark since national policy forbids hospitals from accepting it. Thus, free formula is an issue on which Denmark and the US are distinctly different. Ortiz, J., McGilligan, K., & Kelly, P. (2004). Duration of breast milk expression among working mothers enrolled in an employer-sponsored lactation program. Pediatr Nurs, 30(2), 111- 119. In this study, the authors assess the importance of supportive work policies for breastfeeding success. Their subjects are 462 US women employed by five corporations that support lactation through breastfeeding education classes, certified lactation consultants, and equipped private rooms for pumping milk. Overall, employee programs seemed very useful. Women pumped for an average of 6.3 months. Most women were full-time and had had 2.8 months of maternity leave. Hourly-wage women were less likely to take advantage of the program, which indicates that these women may need to be more aggressively targeted. Employee programs are especially important for US populations, where women have very low levels of guaranteed maternity leave (unlike Denmark). Philipp, B. L., Malone, K. L., Cimo, S., & Merewood, A. (2003). Sustained breastfeeding rates at a US baby-friendly hospital. Pediatrics, 112(3 Pt 1), e234-236. This paper is a case study of Boston Medical Center (BMC). Subjects are 200 infants from 2000-2001. The authors found that in the first year of Baby Friendly status, breastfeeding initiation rates increased form 58% to 86.5% and that these rates were sustained for the next two years. A strength of this study is that the authors differentiate between exclusive and partial breastfeeding. Overall, the authors present good evidence that the BFHI can be effective in the US. This data is potentially useful for comparisons with breastfeeding initiation rates in Danish hospitals. Rowe-Murray, H. J., & Fisher, J. R. (2002). Baby friendly hospital practices: cesarean section is a persistent barrier to early initiation of breastfeeding. Birth, 29(2), 124-131. This Australian study (n = 203) specifically focuses on caesarean delivery as an obstacle to breastfeeding. In my study population, mode of delivery did have a significant effect on duration of both partial and exclusive breastfeeding, but the effect size was very small. However, the importance of delivery method may vary by country. The authors found that women who delivered by c-section were delayed in breastfeeding initiation. This outcome, though, was more related to hospital practices than biological effects of surgical delivery. Thus, hospitals must work to get infants to their mothers more quickly after delivery. Syler, G. P., Sarvela, P., Welshimer, K., & Anderson, S. L. (1997). A descriptive study of breastfeeding practices and policies in Missouri hospitals. J Hum Lact, 13(2), 103-107. 26 The authors analyzed the extent to which Missouri hospitals comply with the BFHI. This study objective is very similar to that of Kovach (1997), except in a different state. None of the Missouri hospitals fully complied with the BFHI. Consequently, this study provides more compelling evidence that US hospitals have a serious need for improvement in respect to breastfeeding policies. This situation contrasts starkly with high levels of breastfeeding support in hospitals in Denmark. Tully, M. R. (2005). Working and breastfeeding. AWHONN Lifelines, 9(3), 198-203. This article focuses on providing practical advice to both employees and employers on how to increase the prevalence of breastfeeding. The author advocates pumping as a useful strategy. Although this paper does not provide me with much new information, it does give me an example of the current distribution of information to women and employers. WHO. (1991). Indicators for assessing breastfeeding practices, Geneva, Switzerland (No. 91.14). Like the Coffin et al paper, this report defines different types of breastfeeding. It also summarizes methods that are useful for assessing different types. These breastfeeding definitions largely agree with the ones presented in Coffin et al. The establishment of standardized nomenclature is necessary to make the study of breastfeeding a respected science. Both of these papers guided my definitions of exclusive and partial breastfeeding. Yngve, A., & Sjostrom, M. (2001). Breastfeeding determinants and a suggested framework for action in Europe. Public Health Nutr, 4(2B), 729-739. This article summarizes recent international documents advocating breastfeeding: the Innocenti Declaration, the Baby-Friendly Hospital Initiative, the WHO international Code of Marketing of Breast Milk Substitutes, and Healthy People 2010. They also summarize the positions of the European Union, WHO/UNICEF, the British Paediatric Association, Swedish concerted action, the American Academy of Pediatricians, the American Dietetic Association, and the ILO Maternity Protection Convention. Next, they review the determinants of breastfeeding as they fall into five categories: socio-demographic, psycho-social, health care related, and community- and policy attributes. The authors emphasize the need to continue research on and fully understand these factors. Finally, they propose a common surveillance system that would promote studies from local to supranational levels. The authors also emphasize the need to develop EU-wide breastfeeding promotion programs, especially ones that target risk groups (young, low-income, uneducated mothers). 27