• Call us now: 0300 303 0444
  • Call us now: 0300 303 0444
Evidence Based Midwifery

You are here

The promotion of breastfeeding among low-income women: midwives’ knowledge and attitudes following a WHO/UNICEF breastfeeding ma

7 January, 2009

The promotion of breastfeeding among low-income women: midwives’ knowledge and attitudes following a WHO/UNICEF breastfeeding ma

Low-income women in the UK have been identified as having especially low breastfeeding rates and in particular need of targeted interventions. The World Health Organization/UNICEF Baby Friendly Initiative (BFI) course in breastfeeding management provides a framework for best practice, but does not specifically address the needs of low-income families.
EBM: March 2007

Francesca Entwistle1 MSc, RM, RN. Sally Kendall2 PhD, BSc, RGN, RHV. Marianne Mead3 PhD, BA, RM, RN.

1 Senior lecturer in midwifery, University of Hertfordshire, College Lane, Hatfield AL10 9AB England. Email: f.entwistle@herts.ac.uk
2 Professor of nursing and director of the Centre for Research in Primary and Community Care, University of Hertfordshire, College Lane, Hatfield AL10 9AB England.
Email: s.kendall@herts.ac.uk
3 Reader in midwifery, University of Hertfordshire, College Lane, Hatfield AL10 9AB England. Email: m.m.p.mead@herts.ac.uk


Background. Low-income women in the UK have been identified as having especially low breastfeeding rates and in particular need of targeted interventions. The World Health Organization/UNICEF Baby Friendly Initiative (BFI) course in breastfeeding management provides a framework for best practice, but does not specifically address the needs of low-income families.
Aim. To determine whether attending a BFI course in breastfeeding management would improve midwives’ knowledge of and attitudes toward breastfeeding generally with respect to BFI guidelines, and specifically in relation to the promotion of breastfeeding among low-income women.
Method. A quasi-experimental approach was used to compare 23 midwives who attended the BFI course with 37 who did not, administering questionnaires based on the BFI guidelines before the course and two weeks and four months after it.
Results. At pre-intervention, 46% (19 out of 41) of the midwives reported feeling unprepared to support women to breastfeed. Attendance was associated with some changes in knowledge and attitudes at two weeks and one at four months. Midwives in the intervention group also felt more optimistic about the number of women who wish to breastfeed, whether or not they actually chose to do so. There were no differences in agreement with statements based on BFI guidelines between the intervention or the comparison group or any significant changes during the period of the study.
Conclusions. This study identified only a limited effect of BFI course attendance on the midwives’ ability to promote breastfeeding among low-income women. The BFI ‘Ten Steps to Successful Breastfeeding’ need to be amended to address the needs of low-income women in overcoming the psychosocial and cultural barriers to breastfeeding.

Key words
: Breastfeeding, Baby Friendly Initiative, World Health Organization, low income, midwives, midwifery


Breastfeeding is a key determinant of public health for women, babies and their families (Department of Health, 2002a). The World Health Organization (WHO) (1998) recommends that all infants should be breastfed exclusively from birth until six months of age. The Innocenti declaration (World Health Organization/UNICEF, 1990) and the WHO/UNICEF Baby Friendly Initiative (BFI) provide a framework for best practice to support breastfeeding in maternity units and other healthcare facilities. The BFI course in breastfeeding management, incorporating the ‘Ten Steps to Successful Breastfeeding’ (Ten Steps) (UNICEF, 1998), aims to prepare midwives and other healthcare professionals to support the initiation and continuation of breastfeeding (Dinwoodie et al, 2000). BFI implementation has been shown to help improve breastfeeding practices worldwide (Cattaneo and Buzzetti, 2001; Gau, 2004).

In the UK in 2000, 69% of new mothers initiated breastfeeding and only 21% of babies were being offered any breastmilk at four to six months, with considerable geographical variation – 81% in the south of England and 61% in the north (La Leche League International Center for Breastfeeding Information, 2003). Rates also varied by social class – 91% in ‘higher occupation’, 59% in ‘lower occupation’ and 52% in ‘never occupation’ women (Hamlyn et al, 2002). The preliminary results of the Infant-feeding survey suggest that increases in breastfeeding tend to be among higher-income and older women (Bolling, 2006). Women in low-income areas often live within a ‘bottle-feeding culture’ and have negative feelings towards breastfeeding, with few role models and higher expectations of breastfeeding failure (Hawkins and Heard, 2001; Scott and Mostyn, 2003). Women are also less likely to breastfeed if they have left school by 16 years and if they are under 20 years of age when they have their infants (Hamlyn et al, 2002). An independent inquiry identified breastfeeding as an issue where inequity can be correlated with inequality in health outcomes, and recommended the development of policies that would increase its prevalence (Acheson, 1998), and the Department of Health (DH) has prioritised the reduction in infant morbidity and mortality in addressing health inequalities (Department of Health, 2001). An NHS target has been set to reduce the gap in infant mortality rates between manual groups and the total population by at least 10%, starting with children aged under one year (Milburn, 2001).

Midwives and other healthcare professionals have been asked to deliver an increase of two percentage points per year in breastfeeding initiation rates, focusing especially on women from disadvantaged groups (Department of Health, 2002b), and the second of the Ten Steps stipulates that all healthcare staff should be trained in the necessary skills to implement a breastfeeding policy based on them. However, there is little or no evidence on the effectiveness of training on midwives’ ability to promote breastfeeding initiation or duration among low-income women. The breastfeeding needs of this group are challenging and require a focused approach (Hoddinott and Pill, 1999; Locklin, 1995). Milligan and Pugh (2000: 250) suggest that: ‘The midwife must work to become culturally competent with low-income women. As primary care providers, midwives can provide consistent, persistent advice during the prenatal period on breastfeeding and promote optimism about the mother’s success.’

A systematic review of the effectiveness of public health interventions to promote the initiation of breastfeeding highlighted the need for evaluative research to assess the training of healthcare professionals and their impact on implementing the Ten Steps (Fairbank et al, 2000; Protheroe et al, 2003).

Several studies have been carried out worldwide to explore the impact of breastfeeding training for healthcare professionals. Those that explored midwives’ knowledge, attitude and/or breastfeeding outcomes generally found that breastfeeding rates improved on initiation and at discharge from hospital, but were not sustained over time (Becker, 1992; Labarere et al, 2003; Vittoz et al, 2004). In some cases this was due to inadequate follow up (Gau, 2004), however further research to evaluate the duration of breastfeeding has been recommended.

Other studies have explored healthcare professionals’ changes in knowledge and/or attitudes following a training programme to support the promotion of breastfeeding practices. Although the healthcare professionals were more knowledgeable and had a more positive attitude to breastfeeding, the application of knowledge to practice in terms of breastfeeding initiation and continuation rates was not measured (Dinwoodie et al, 2000; Westphal et al, 1995; Wissett et al, 2000).

Midwives traditionally view helping women to breastfeed as integral to their role, and want to feel that their knowledge base is supported by up-to-date training and practical experiences (Shaw et al, 2004). But evidence suggests that midwives overestimate their knowledge and practical abilities to support women to breastfeed successfully (Cantrill et al, 2003).

A theory-practice gap exists as some midwives believe in offering practical help and advice to mothers whether or not the forms of help and advice are based on research evidence (Stein et al, 2000). Moran et al (2000) used a Breastfeeding Support Skills Tool (BeSST) to assess midwives’ practical skills – though the results were positive following training at two weeks, this was not related to breastfeeding outcomes. Many healthcare professionals draw on their own experiences to support breastfeeding, and while some find that this assists their practice, others find it causes difficulties (Battersby, 2001; Freed et al, 1995). The attitudes of midwives and other healthcare professionals are influential (Gau, 2004; Henderson et al, 2000), and where women report neutral or negative breastfeeding messages, breastfeeding duration can be affected (DiGirolamo et al, 2003).

The BFI breastfeeding management course (World Health Organization/UNICEF, 1997) was designed to narrow the theory-practice gap (Dinwoodie et al, 2000). However, there is limited information on the impact of healthcare professional breastfeeding training programmes on the implementation of the Ten Steps (Fairbank et al, 2000). Furthermore, there is no evidence to evaluate how these training initiatives affect the initiation of breastfeeding for low-income women.


This study sought to assess potential improvement in midwives’ knowledge of and attitudes toward breastfeeding following attendance of a BFI course in breastfeeding management – generally with respect to the BFI guidelines and specifically in relation to the promotion of breastfeeding among low-income women.


A pre- and post-test quasi-experimental approach (no randomisation) was used. Funding was secured from the DH in the context of its Infant-Feeding Initiative. Ethical approval was granted by two local research ethics committees.


The study was conducted in two NHS Trusts covering similar geographical areas of the south of England where pockets of deprivation could be identified through the use of the Jarman Index (Jarman, 1991). In both Trusts, the midwives worked in teams to provide continuity of care for the women, caring for them both in the hospital and community environment, and care was midwifery led for low-risk women. There were some essential differences between the two Trusts at the time of the study – the intervention Trust (IT) had staff vacancies arising from normal turnover, whereas the comparison Trust (CT) was experiencing a severe shortage of midwives with a large bank of agency staff routinely complementing the stable workforce. This was not unusual in the south of England at the time of the study, where the average shortage of staff was 6% (Davis, 2002). However, this may have had an effect on midwifery care and therefore breastfeeding support, uptake and continuation.


Midwives in the IT were recruited from two teams of community midwives who worked with women from areas identified as deprived by the Jarman Index (Jarman, 1991). A supplementary group of midwives from the labour ward were selected if they cared for women in the study sites when a team midwife was not available. A total of 24 places were available on the course. The same recruitment approach was used in the CT, although numbers of potential participants were not restricted. Midwifery teams working with women who came from less deprived areas were excluded.


The midwives from the CT provided their normal pattern of care, while IT midwives attended a BFI breastfeeding management course facilitated by two BFI team tutors in May 2002. The BFI course was based on the physiology of breastfeeding,common maternal and neonatal problems associated with breastfeeding and their solutions, and some longer-term items such implementation of the Ten Steps. It ran over two consecutive days and one follow-up day two weeks later, with a compulsory practice element between the second and third days. A module handbook with several optional worksheets accompanied the course. No specific breastfeeding training was provided in the CT – midwives relied on their midwifery pre-registration course and any further education they may have chosen to attend since. Prior to the intervention, the two Trusts had adopted similar approaches to in-service breastfeeding education.

Data collection

After critical examination of evaluation tools used by other researchers (Dinwoodie et al, 2000; Wissett et al, 2000), a tool based on studies examining midwives’ and medical students’ attitudes (Dover and Gauge, 1995; Dracup and Sanderson, 1994) was devised. Three questionnaires based on the Ten Steps were developed to be used at the pre-intervention stage and at two weeks’ and four months’ post-intervention.

The pre-intervention questionnaire collected data on demographic details, personal experience of breastfeeding and understanding and implementation of BFI guidelines, followed by a problem-based scenario. As there is an increased likelihood for women who were breastfed to breastfeed their own children (Foster, 1997), the personal experience section explored whether midwives had been breastfed as infants and, if they had children whether they had breastfed them. Of 33 questions relating to BFI guidelines, 19 dealt specifically with ‘knowledge and attitudes’ and 14 with ‘policy and management’. The scenario presented the case of a 20-year old firsttime single mother who had left school at 16 years of age and came from a bottle-feeding environment, and used nine questions to explore midwives’ attitudes further. A mixture of positive and negative statements were used to reduce bias (Denscombe, 2003). Likert scales – ‘strongly agree’ to ‘strongly disagree’ with a ‘don’t know’ option – were used, although the data relating to the BFI guidelines were later recoded into dichotomous responses.

The post-intervention questionnaires repeated the sections relating to BFI guidelines for midwives in both Trusts. IT midwives were also asked to evaluate the content and delivery of the course, and to rate their personal achievement of aims and learning outcomes.

The questionnaires were piloted with practising midwives in one of the Trusts associated with the university but not involved in the study and critically analysed by a research methodologist. No major flaws in the design of the questionnaire were found. This process increased face validity and reliability of the data collection tool.


The data were entered into SPSS for Windows (version 11.0). The answers to questions relating to BFI guidelines were recoded into dichotomous ‘1=correct’ or ‘0=incorrect’ variables, according to the Ten Steps criteria. A ‘don’t know’ response was recoded as ‘incorrect’, since it could not be shown to demonstrate support for breastfeeding. The total score of the ‘correct’ answers were calculated for each midwife at the three stages of the study, for the IT and for the CT.

Although the small sample meant that statistical analysis was undertaken using descriptive frequencies, chi-squares and the Fisher’s Exact test for categorical variables for comparison of the IT and CT, and paired t-tests and ANOVA for the comparison of the scores of IT midwives at the three stages of the study. The significance level was set at p<0.05.


A total of 23 midwives were recruited to the study from the IT and 37 from the CT. Two midwives from the IT and four from the CT were sick or left between the first and second questionnaires. The denominators given throughout the findings refer to the number of midwives who answered a specific question. The response rate was higher in the IT – 20/23 (87%), 15/21 (71%) and 12/21 (57%) respectively for the three questionnaires. The equivalent rates in the CT were 23/37 (62%), 17/33 (52%) and 17/33 (52%). The number of midwives who responded to all three questionnaires were 10/21 (48%) in the IT and 13/33 (29%) in the CT.

Although midwives from both Trusts cared for low-income women, those in the IT cared for women from areas that were less deprived according to the Jarman Index.

The midwives from both Trusts were comparable in age (median age group 34 to 45 years), years of experience (60% with ten years’) and academic background (certificate 42%, diploma 34%, degree 24%). Half of them worked exclusively in hospital, but more of those who did not worked exclusively in the community in the IT (7/19), and in both hospital and the community in the CT (9/21). The midwives from the IT worked in two teams: one assured continuity of care during delivery in hospital, whereas the other only did so when the team midwives were on-call.

The majority of midwives had attended additional breastfeeding training sessions since qualifying as a midwife – 15/20 in the IT and 20/21 in the CT. Approximately half the midwives thought that these courses had prepared them to help women breastfeed – 9/20 in the IT and 9/21 in the CT.

Of respondents from both Trusts, 25 had a total of 67 children. Only three of these children were exclusively artificially fed, and all of the midwives who had children had breastfed. There were significant differences between the two Trusts in whether respondents had been breastfed – 15/18 midwives in the IT and 7/22 in the CT (chi-square=9.079, df=1, p=0.003).

At the onset of the course, 19/41 (46%) of midwives from both Trusts reported feeling unprepared to support women to breastfeed, and 34/40 (85%) stated that a skilled trained maternity care assistant could provide women with breastfeeding support as effectively as a trained midwife. Without any identifiable cause, the midwives in the CT gave significantly more positive answers from the IT midwives on the following BFI principles:

■ Antenatal checklists were beneficial to women and were used by midwives
■ Use of bottle-feeding demonstration sessions during antenatal classes
■ Breastfeeding policy was displayed
■ Breastfeeding within one hour of delivery increased the success rate
■ Labour wards are usually too busy to encourage skin-to skin contact
■ No extra fluids should be given if babies underwent phototherapy
■ The extended role of the midwife does not allow them to spend as much time as they would like with the mother and her baby to help with breastfeeding.

All of the midwives from both Trusts gave responses supporting the following BFI principles:

■ Women should be informed of the health benefits of breastfeeding
■ It should be mandatory for all midwives to be taught how to teach mothers how to position and attach their babies at the breast
■ The promotion and support of breastfeeding is an essential role of the midwife. Near unanimous agreement was achieved on statements that:
■ The Trust has a written breastfeeding policy (95%)
■ It is essential to express milk when mother and baby are separated for long periods of time (97%)
■ Midwives should work closely with neonatal unit staff to establish breastfeeding
■ Midwives should encourage undecided women to breastfeed
■ Training midwives in the skills of breastfeeding will increase breastfeeding rates (95%). At the pre-intervention stage, some BFI principles were more challenging to the midwives from both Trusts:
■ Antenatal checklists were only reported to be used routinely by 21/41 (51%)
■ 9/39 (23%) agreed that a paediatrician should always be consulted if a baby required artificial formula when breastfed
■ 11/41 (27%) agreed that breastfeeding support groups were only available to a select group of women
■ 9/39 (23%) agreed that lactational amenorrhoea was as safe a method of contraception as the progesterone-only pill
■ 9/41 (22%) reported that increasing demands on their role prevented them from supporting breastfeeding mothers adequately.

Differences in the near-unanimous or low-grade agreement of midwives between the two Trusts were not significant. The differences between pre-intervention, two weeks’ and four months’ post-intervention responses from midwives in both Trusts were also analysed. There were no statistical changes in the items that had attracted unanimous or near-unanimous agreement, either at two weeks or four months within or between the Trusts. Midwives were significantly more likely to report that antenatal checklists were used routinely in the IT at two weeks’ and four month’s post-intervention than in the CT. For unidentified reasons, midwives in the CT were significantly less likely to report this at two weeks and four months than at pre-intervention. Significant differences were measured among responses from IT midwives on the importance of antenatal checklist, hospital breastfeeding policy on display and that extra fluids are unnecessary if a breastfed baby is receivingphototherapy between pre-intervention and two weeks after it. That a paediatrician should be consulted if a breastfed baby requires artificial supplements demonstrated a sustained change at four months among IT midwives (see Table 1).

 As the number of respondents was small, it may be useful to examine variations at two weeks and four months even where statistical significance was not reached. More of the midwives from the IT than the CT reported that an antenatal infantfeeding checklist was used, that a unit breastfeeding policy was displayed in the hospital, that a copy was available for women to take home and read and that the policy was explained to women routinely. More IT midwives also agreed that women should be informed about the harmful effects of teats and dummies, that bottle-feeding demonstrations should not be included in antenatal classes, that it was essential that women should be taught how to express milk by hand, that a paediatrician should be consulted if a breastfed baby required artificial formula, that nipple shields should not be used as an interim measure if the mother had sore nipples, and that lactational amenorrhoea was as safe a method of contraception as the progesterone-only pill. More IT midwives’ responses demonstrated that their policy on neonatal hypoglycaemia followed WHO guidelines.

Respondents were also asked to quantify the percentage of all women they thought would choose to breastfeed. The rates identified by the CT did not vary significantly between the three surveys, but there were significant differences in the answers of the IT midwives between pre-intervention and postintervention surveys. This suggests that the intervention led to a more optimistic perception of the rate of women who wish to breastfeed, whether or not they actually choose to do so.

Total scores from the 33 questions relating on BFI guidelines were compared using paired t-tests. The overall score showed a significant difference at two weeks in the IT and none at four months, and there were no significant differences in the CT. One question dealt specifically with the influence of the women’s social and cultural background compared to the influence of the midwife. At pre-intervention, midwives tended to agree that women’s social and cultural background exercised a greater influence on their choice of infant-feeding method – 14/20 (70%) for the IT and 14/21 (67%) for the CT. However, at two weeks 8/15 midwives (53%) in the IT still agreed with the statement. At four months, the rate had risen to 8/12 (67%). In the CT, the rates at two weeks and four months were 11/17 (65%) and 11/19 (58%) respectively.

In the nine-question scenario, there were no significant differences in the mean scores between midwives from the two Trusts or between the three stages of the study – 5.5, 7.0 and 3.9 in the IT and 6.0, 6.7 and 4.1 in the CT at pre-intervention, two weeks and four months. However, it is important to bear in mind that the power of the study was not sufficient to reach statistical significance.


As this study recruited all the midwives involved in the care of low-income women in the two Trusts, there can be confidence that the results are an accurate reflection of a specific group of midwives’ knowledge before and after the introduction of the breastfeeding management course. The small sample size restricts the study’s generalisability and the weight that may be given to any measured significance in differences between responses. The collection of data using Likert scales and recoding to dichotomous values may also have affected the findings. The impossibility of allocating two NHS Trusts and the midwives working in them randomly to either group of participants meant that differences between them could be expected to be found, and some were also identified at the onset of the study. The BFI course did not deal specifically with the needs of low-income women, and as the questionnaire was based on the content of the course, it was limited in the amount of information it could explore on midwives’ knowledge of and attitudes toward low-income women. The differences between the two Trusts in staff vacancies and the use of bank staff may have affected the type of midwifery care available to women, and was reflected in some of the responses of the midwives.

That the majority of midwives stated that a skilled trained maternity care assistant could provide women with breastfeeding support as well as a trained midwife might reflect confidence in these roles. In the UK, the role of maternity care assistant has been developed with the stated aim of supporting midwives and widening access to pre-registration midwifery programmes (Department of Health, 2005; McKenna and Hasson, 2003). However, there has been little evaluation of breastfeeding education programmes for maternity care assistants, and further research is needed to investigate whether or not such programmes would support the role of the midwife and improve the quality of breastfeeding outcomes for women, and for low-income women in particular.

Despite previous training and a high level of professional experience, the fact that approximately half of the midwives reported feeling unprepared to support breastfeeding among women upon qualification or even after further preparation suggests that changes in the delivery of pre- and post-registration breastfeeding education are required. Since this study was conducted, UNICEF UK BFI launched new education standards to improve breastfeeding training for midwives and health visitors (UNICEF UK Baby Friendly Initiative, 2002).

Although it has limitations, this study confirms previous findings that personal experience of breastfeeding is high among midwives (McMulkin and Malone, 1994). It has been argued that embodied knowledge of breastfeeding should enable midwives to review how their positive and negative experiences of it enables them to better understand and support the women they care for (Battersby, 2000).

This study did not demonstrate sustained changes in knowledge and attitude between pre-intervention and four months’ post-intervention, raising questions about the impact of a costly and time-consuming course. According to the National Institute for Health and Clinical Excellence (NICE), all maternity care providers should implement an externally evaluated, structured programme that encourages breastfeeding, using BFI guidelines as a minimum standard (National Institute for Health and Clinical Excellence, 2006). In developed countries, major reasons for participation in continuing professional development include compliance with employers’ requirements, and the knowledge and skills gained are often insufficiently applied in daily practice (Aiga and Banta, 2003). To fulfil the NICE recommendations, educationalists need to ensure that training programmes are designed to equip healthcare professionals with the skills necessary to empower low-income women to breastfeed successfully. Educationalists need to consider course delivery, as interactive workshops can result in moderately large changes in professional practice, whereas didactic sessions alone are unlikely to change practice (Davis et al, 1999).

While there is no documented cost analysis for the BFI course, a costing by NICE states that the net saving achieved through the implementation of BFI guidelines could vary between £1.9million and £5.6million in individual NHS Trusts as occurrences of childhood illnesses such as gastroenteritis, asthma and otitis media are significantly reduced through an increase in breastfeeding (National Institute for Health and Clinical Excellence, 2006).

The Ten Steps does not address the needs of specific social groups, which may be inevitable given wide variations in breastfeeding rates across the world (UNICEF, 2001) and potentially different needs of socially disadvantaged women in different countries.

Conclusion and recommendations

This study has identified only a limited effect of attendance of the BFI course on midwives’ ability to promote breastfeeding among low-income women. Midwives need to be competent in supporting individual women with their psychosocial and cultural as well as physical needs. The Ten Steps do not address the needs of specific social groups, but the BFI breastfeeding management course follows them. To ensure that all education programmes meet the needs of women from low-income groups, the researchers recommend that an amendment to the Ten Steps be considered – to state that maternity services should help mothers to overcome the psychosocial and cultural barriers to breastfeeding, focusing on the needs of socially disadvantaged women.


Acheson D. (1998) Independent inquiry into inequalities in health. HMSO: London.

Aiga H, Banta JE. (2003) Emerging crisis of continuing professional education: becoming a dead letter between external pressure and personal incentive? International Journal of Health Planning Management 18(3): 267-71.

Battersby S. (2000) Breastfeeding and bullying: who’s putting the pressure on? The Practising Midwife 3(8): 36-8.

Battersby S. (2001) The Worldly Wise project: a different approach to breastfeeding support. The Practising Midwife 4(6): 30-1.

Becker GE. (1992) Breastfeeding knowledge of hospital staff in rural maternity units in Ireland. Journal of Human Lactation 8(3): 137-42.

Bolling K. (2006) Infant-feeding survey 2005: early results. The Information Centre: London.

Cantrill RM, Creedy DK, Cooke M. (2003) An Australian study of midwives’ breastfeeding knowledge. Midwifery 19(4): 310-7.

Cattaneo A, Buzzetti R. (2001) Effect on rates of breastfeeding of training for the baby friendly hospital initiative. BMJ 323(7325): 1358-62.

Davis D, O’Brien MA, Freemantle N, Wolf FM, Mazmanian P, Taylor-Vaisey A. (1999) Impact of formal continuing medical education: do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or healthcare outcomes? JAMA 282(9): 867-74.

Davis DK. (2002) Renewing midwifery and fulfilling our potential. RCM Midwives Journal 5(5): 162-5.

Denscombe M. (2003) The good research guide: for small-scale social research projects. Open University Press: Maidenhead.

Department of Health. (2001) Tackling inequalities. HMSO: London.

Department of Health. (2002a) Improving the nation’s health: government to produce food and health action plan. See: www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4026013&amp;chk=BHuH46 (accessed November 2006).

Department of Health. (2002b) The NHS plan: the next three years: planning priorities and framework 2003 to 2006. HMSO: London.

Department of Health. (2005) Improving working lives in midwifery: a guide for midwives and managers. HMSO: London.

DiGirolamo AM, Grummer-Strawn LM, Fein SB. (2003) Do perceived attitudes of physicians and hospital staff affect breastfeeding decisions? Birth 30(2): 94-100.

Dinwoodie K, Bramwell R, Dykes F, Lang S, Runcieman A. (2000) The Baby Friendly breastfeeding management course. BJ Midwifery 8(6): 362-6.

Dover S, Gauge S. (1995) Fetal monitoring: midwifery attitudes. Midwifery 11(1): 18-27.

Dracup C, Sanderson E. (1994) Health visitors’ perceptions of breastfeeding mothers. Health Visitor 67(5): 158-60.

Fairbank L, O’Meara S, Renfrew MJ, Woolridge M, Sowden AJ, Lister-Sharp D. (2000) A systematic review to evaluate the effectiveness of interventions to promote the initiation of breastfeeding. Health Tech Assment 4(25): 1-171.

Foster K. (1997) Infant-feeding 1995: a survey of infant feeding practices in the UK. HMSO: London.

Freed GL, Clark SJ, Sorenson J, Lohr JA, Cefalo R, Curtis P. (1995) National assessment of physicians’ breastfeeding knowledge, attitudes, training and
experience. JAMA 273(6): 472-6.

Gau ML. (2004) Evaluation of a lactation intervention program to encourage breastfeeding: a longitudinal study. Intl J Nursing Studies 41(4): 425-35.

Hamlyn B, Brooker S, Oleinikova K, Wands S. (2002) Infant-feeding 2000. HMSO: London.

Hawkins A, Heard S. (2001) An exploration of the factors which may affect the duration of breastfeeding by first-time mothers on low incomes: a multiple case study. MIDIRS Midwifery Digest 11(4): 521-6.

Henderson AM, Pincombe J, Stamp G. (2000) Assisting women to establish breastfeeding: exploring midwives’ practices. Breastfeed Review 8(3): 11-7.

Hoddinott P, Pill R. (1999) Qualitative study of decisions about infant-feeding among women in east end of London. BMJ 318(7175): 30-4.

Jarman B. (1991) Jarman index. BMJ 302(6775): 527.

Labarere J, Castell M, Fourny M, Durand M, Pons J-C. (2003) A training program on exclusive breastfeeding in maternity wards. Intl J Gyn Obst 83(1): 77-84.

La Leche League International Center for Breastfeeding Information. (2003) Breastfeeding statistics. See: www.lalecheleague.org/cbi/bfstats03.html
(accessed November 2006).

Locklin MP. (1995) Telling the world: low-income women and their breastfeeding experiences. Journal of Human Lactation 11(4): 285-91.

McKenna H, Hasson F. (2003) Training needs of midwifery assistants. J Advanced Nursing 44(3): 308-17.

McMulkin S, Malone R. (1994) Breastfeeding: midwives’ personal experiences. Modern Midwife 4(5): 10-2.

Milburn A. (2001) Health secretary announces new plans to improve health in poorest areas. See: www.dh.gov.uk/PublicationsAndStatistics/PressReleases/PressReleasesNotices/fs/en?CONTENT_ID=4014292&amp;chk=4HQNYv (accessed November 2006).

Milligan RA, Pugh LC, Bronner YL, Spatz DL, Brown LP. (2000) Breastfeeding duration among low-income women. J Midwifery and Women’s Health 45(3): 246-52.

Moran VH, Bramwell R, Dykes F, Dinwoodie K. (2000) An evaluation of skills acquisition on the WHO/UNICEF breastfeeding management course using the pre-validated Breastfeeding Support Skills Tool. Midwifery 16(3): 197-203.

National Institute for Health and Clinical Excellence. (2006) Routine postnatal care of women and their babies. National Institute for Health and Clinical Excellence: London.

Protheroe L, Dyson L, Renfrew MJ, Bull J, Mulvihill C. (2003) The effectiveness of public health interventions to promote the initiation of breastfeeding. Mother and Infant Research Unit: Leeds.

Scott J, Mostyn T. (2003) Women’s experiences of breastfeeding in a bottlefeeding culture. Journal of Human Lactation 19(3): 270-7.

Shaw R, Wallace L, Cook M, Phillips A. (2004) Perceptions of the breastfeeding Best Start project. The Practising Midwife 7(1): 20-4.

Stein J, Dykes F, Bramwell R. (2000) Breastfeeding: midwives meeting mothers in the middle. British Journal of Midwifery 8(4): 239-45.

UNICEF. (1998) Implementing the ‘Ten steps to successful breastfeeding’. UNICEF: New York.

UNICEF. (2001) Breastfeeding and complementary feeding. See:
www.childinfo.org/eddb/brfeed/index.htm (accessed November 2006).

UNICEF UK Baby Friendly Initiative. (2002) Introducing the Baby Friendly best practice standards into breastfeeding education for student midwives and health visitors. UK Baby Friendly Initiative: London.

Vittoz JP, Labarere J, Castell M, Durand M, Pons JC. (2004) Effect of a training program for maternity ward professionals on duration of breastfeeding. Birth 31(4): 302-7.

Westphal MF, Taddei JA, Venancio SI, Bogus CM. (1995) Breastfeeding training for health professionals and resultant institutional changes. Bulletin of the World Health Organization 73(4): 461-8.

Wissett L, Dykes F, Bramwell R. (2000) Evaluating the WHO/UNICEF breastfeeding course. BJ Midwifery 8(5): 294-300.

World Health Organization. (1998) Evidence for the ‘Ten steps to successful breastfeeding’. World Health Organization: Geneva.

World Health Organization/UNICEF. (1990) The Innocenti declaration on the protection, promotion and support of breastfeeding. UNICEF: Florence.

Table 1. Summary of the papers included in the review

Printer-friendly version