Royal Society of Medicine Forum: Breastfeeding – State of the Art.
This is a report of a meeting of the Forum on Maternity and the Newborn of the Royal Society of Medicine, held on Tuesday 24th February 2004. The meeting was chaired by Dr. Elvidina Adamson-Macedo, Perinatal psychologist, University of Wolverhampton.

The report has been published in The Midwives Journal of the Royal College of Midwives. It is reproduced here with their consent and our thanks.

Enabling women to breastfeed - the evidence from systematic reviews
Professor Mary Renfrew, Midwife, Director of the Mother and Infant Research Unit, Leeds Infirmary

Breastfeeding is a fundamentally important public health issue, and an important one for mothers, babies and families, and recent policy changes in the UK and internationally support work towards increasing breastfeeding initiation rates, and encouraging exclusive breastfeeding for six months. The government's very difficult target is to increase breastfeeding by 2% annually among women from the deprived groups, where rates are low. The challenge to mothers and healthcare practitioners is very hard to meet.

This paper will summarise the results of a series of systematic reviews of clinical and health promotion interventions, examining ways of increasing breastfeeding initiation and duration, especially among families from low-income groups. Suggestions for coordinated action at national and local levels will be made, based on this evidence.

It is clear that breastfeeding is an important and effective intervention with multiple benefits: less mortality in premature babies; less gastroenteritis, atopic disease, respiratory disease, urinary tract infection, otitis media, and hospitalisation in infancy. Other risks are reduced: diabetes and obesity in childhood; adult disease, including cardiovascular disease and Crohn's disease. IQ is increased, and there are benefits for maternal health and the mother-baby relationship.

Much research in this area is flawed, though improving recently. Nevertheless if breastfeeding were a drug the drug companies would be climbing over themselves to get it to the market.

The problem is that in the UK, breastfeeding rates are low, and lowest among families living in socially and materially deprived circumstances. There are many reasons for this, involving psychosocial factors, continuation of practices in hospital that make breastfeeding difficult, and a lack of basic education among health professionals working in hospital and community settings. Lack of public support is a major issue. Though easy physiologically breastfeeding is culturally difficult, more so in lower socio-economic groups. Less than half the babies living in the most deprived conditions and most likely to experience the poorest health status are likely to breast feed at all. Less than a quarter of those breastfed are so fed for more than four months, and supplements are introduced early.

To address inequality the number of women breastfeeding in social classes four and five needs to be doubled, for initiation and for continuation at four months, to achieve the same rates as a higher social class groups and to close the gap.

Inappropriate and harmful hospital practices and routines make breastfeeding difficult. So do lack of confidence and experience in mothers, health professionals and families, with consequent lack of support. Early return to employment is the norm, with little support in the workplace. Society is not breastfeeding friendly; there is a bottle-feeding culture among some groups.

As part of a larger study – Looking at Infant Feeding Today - focus groups with pregnant women and men living in socially and materially deprived circumstances expressed their perceptions of the disadvantages of breastfeeding and of related physical problems such as pain; few were aware of the disadvantages of artificial feeding, and breast-feeding in public was a predominant issue; there was an underlying lack of self-confidence. However, research among low income groups is very limited. There are problems of literacy, language and culture, and poor response to postal questionnaires; these are groups which move house frequently.

The partner of a pregnant woman said "I just think that breastfeeding gives you breast cancer; no one ever told us the good points, so they haven't said it's good."

"I think a lot of young mothers won't breastfeed because the reaction is 'You're lazy, you don't want to be getting up and making bottles at night, it's easier to grab the baby and put the baby on your breast than to go downstairs and warm the bottle up'".

Breastfeeding in public is a huge concern for women; they see this as providing voyeuristic pleasure for men were not their partners, and many men have confirmed this. It is particularly difficult for women travelling on public transport, but it can also be a problem for women breastfeeding in front of others in their own homes.

What follows is evidence-based information on initiation and maintenance of breastfeeding.

Peer support programmes in the antenatal and postnatal periods, combined with local media campaigns, are especially useful for women on low incomes who have expressed a desire to breast feed, particularly when delivered by a trained counsellor who is herself a locally resident mother who has breastfed. Also effective are mother to mother support groups led by a trained counsellor, sometimes with a health visitor, in women's homes or a community centre. Small group informal health education sessions increase initiation rates among women from all income groups and from different ethnic groups. Modest payments in cash or in kind (for example provision of breast pads or bus fares), and involvement of women's partners, sisters or mothers in health education activities increase participation in group sessions (Renfrew et al. 2000).

Continuous support in labour (Hodnett 2004), early close contact and - which is most effective, and not specific to women on low incomes - the baby rooming in, promote maintenance of breastfeeding. Vital are accurate advice and care in regard to positioning, unrestricted contact and problems, and here the good advice of health professionals is crucial; unfortunately many are not well trained to provide such advice. Face to face support, lay or professional is best, though telephone support can also be effective. (Sikorski et al. 2004; Fairbank et al. 2000) Harmful practices such as supplementation, discharge packs (Donnelly et al. 2004), and restricted contact must be avoided.

A multifaceted approach, otherwise the Baby Friendly initiative, works - peer support in combination with health education, media programmes and/or changes to the healthcare sector, including training of health professionals and changes in government and hospital policies; all these were well done in Scandinavia.

What doesn't work? Literature or leaflets alone. Staff training, unless it is part of a multifaceted approach as above. National media campaigns increase breastfeeding levels among women on high incomes with no effect on those with lower incomes.

Desirable policy developments are government guidance on good practice as incorporated in the National Institute for Clinical Excellence (NICE), the NHS National Service Framework (NSF), and the Commission for Healthcare Audit and Inspection (CHAI). Agreed targets, with appropriate routine monitoring as with immunisation, are helpful; so too are Healthy Start, where support for breastfeeding is expected to be an improvement on the present welfare food scheme, and the development of community based peer and professional resources, Sure Start and public facilities for breastfeeding. Nationally we need to implement rooming in fully, to remove harmful practices, implement evidence-based care in hospitals, and to support the Baby Friendly initiative in hospital and community (Protheroe et al. 2003). Breastfeeding training for all health professionals must be reviewed, ensuring that it is evidence-based and appropriate for all groups. Pre-registration education and mandatory updating, currently absent from the curriculum, are badly needed.

Ideas for action locally include the national linking of breastfeeding co-ordinators to Primary Care Trusts (PCTs)and hospitals, and inequalities initiatives such as Sure Start and the Health Action Zone initiative (HAZ); the recruitment and training of local mothers as paid counsellors; incentives for mothers to support participation in groups, with pumps (Snowden et al. 2004) and breast pads; reviews of health promotion; parent activities to ensure appropriate delivery of appropriate information, and support for local staff training .

The objective is successful breastfeeding for all, enabling mothers and babies to enjoy their breastfeeding relationships without the traumas to which I have alluded. They and their families need access to activities throughout pregnancy, birth and after, involving policy makers and educators.


Donnelly A, Snowden HM, Renfrew MJ, Woolridge MW Commercial hospital discharge packs for breastfeeding women. Cochrane Database of Systematic Reviews, The Cochrane Library. Oxford: Update Software.

Fairbank L, Renfrew MJ, Woolridge MW, Sowden AJS, O'Meara S (2000) Systematic review to evaluate the effectiveness of interventions to promote the uptake of breastfeeding. Health Technology Assessment (HTA) programme of the NHS R&D programme: Monograph 4(25). Can be ordered direct from web (free)

Hamlyn B, Brooker S, Oleinikova K, Wands S (2002) Infant feeding 2000. London: The Stationery Office. Available on

Hodnett ED. Caregiver support for women during childbirth. Cochrane Database of Systematic Reviews. The Cochrane Library. Oxford: Update Software.

Kramer MS, Kakuma R. Optimal duration of exclusive breastfeeding. Cochrane Database of Systematic Reviews, The Cochrane Library. Oxford: Update Software

O'Meara S, Sowden A, Lister-Sharp D, Mather L (NHS Centre for Reviews and Dissemination) and Fairbank L, Woolridge MW, Renfrew MJ (Mother and Infant Research Unit) (2000) Promoting the initiation of breastfeeding. Effective Health Care Bulletin. NHS Centre for Reviews and Dissemination, University of York, 6(2). Subscriptions Department, Royal Society of Medicine Press, PO Box 9002, London, W1A 0ZA. Copies available free from the web -

Protheroe L, Dyson L, Renfrew MJ, Bull J, Mulvihill C (2003) The effectiveness of public health interventions to promote the initiation of breastfeeding: Evidence briefing. Health Development Agency 1st Edition June 2003

Renfrew MJ, Woolridge MW, Ross McGill H (2000) Enabling women to breastfeed. A review of practices which promote or inhibit breastfeeding – with evidence-based guidance for practice. Norwich: The Stationery Office. TSO, PO Box 29, St Crispins Duke Street, Norwich NR3 1GN. Can be ordered directly from the web -

Renfrew MJ, Fisher C, Arms S (2000) The new Bestfeeding: getting breastfeeding right for you. Celestial Arts, PO Box 7123, Berkeley, California 94707 USA. ISBN 0890879559. Can be ordered directly from website,, or from Amazon. Third edition to be published 2004.

Sikorski J, Renfrew MJ, Pindoria S, Wade A Support for breastfeeding mothers. Cochrane Database of Systematic Reviews, The Cochrane Library. Oxford: Update Software.

Snowden HM, Renfrew MJ, Woolridge MW Treatments to relieve breast engorgement during lactation. Cochrane Database of Systematic Reviews, The Cochrane Library. Oxford: Update Software.

WHO (1998). Evidence for the ten steps to successful breastfeeding. WHO Geneva

WHO Global strategy for infant and young child feeding. (2002) WHO Geneva. Available through WHO website,

Meier P. 2003. Supporting lactation in mothers with very low birthweight infants. Pediatric Annals 32:5 317-326

Meier P. 2001 Breastfeeding in the special care nursery: prematures and infants with medical problems. Pediatric clinics of North America 48:2 425-442.

The Rush Mothers' Milk Club: Breastfeeding interventions for mothers with very low birth weight infants. Meier P. et al. J Obstet Gynecol Neonatal Nurs.2004; 33: 164-174.

Guidelines for breastmilk feeding of babies in special care units. The Rush Mothers' Milk Club, Department of Maternal-Child Nursing and Section of Neonatology, Rush-Presbyterian-St Luke's Medical Centre, Chicago.

Developing local breastfeeding strategies for Primary Care Trusts: use of research evidence
Professor Louise Wallace, Professor of Psychology & Health and Director of Health Services Research Centre, Coventry University

The importance of breastfeeding to achieving a wide range of public health benefits to children and mothers has been long established. The new NHS target for PCTS to increase breastfeeding initiation by 2% p.a. in those least likely to breastfeed has put the spotlight of NHS management on breastfeeding research, and on those practitioners whose skills are needed to turn evidence into practice. Results of research on midwifery support for breastfeeding mothers are used here to highlight areas for action. Given the importance of the skills and knowledge of staff and lay supporters who assist mothers and babies, development of new approaches to training and assessment are described. The approach to consultancy with local PCTs is described. Using a social marketing approach, key stakeholders' views can be harnessed for effective local action.

So we have the 2% per annum increase in breastfeeding target for PCTs from April 2004, but no national breastfeeding strategy, though one is expected for England. There are huge known public health benefits, and breastfeeding sits within the government's inequalities priorities, since those least likely to breastfeed are also the most disadvantaged in other ways. National survey data shows that mixed feeding within days or weeks is the norm, and less than 10% are breastfeeding exclusively at six months as recommended by the World Health Organisation. Social class and education differences favour the most advantaged.

The information I have on where we are now is based on my experience with four Midlands PCTs, where my role has been as a researcher with data on local services, and with a familiarity with published research which busy workers in the field may have little time to access. I am a non-executive director of a PCT with a lead role in public health, and consultant to the public health director of another PCT.

The definitions of breast-feeding patterns show a marked variation between clinical notes and diaries kept by the mothers. These show a range between exclusive breastfeeding and mixed feeding; precise definitions can provide accurate data, whereas loose definitions are likely to lead to definition drift, tending erroneously to favour exclusive breastfeeding.

Enquiry into the antenatal discussion of breast-feeding showed that 97% of primigravidae attended antenatal checks and 85 per cent attended antenatal classes; however 30% reported that they could recall no discussion of breastfeeding, serious examples of missed opportunities. In a hospital trial of the help received by breastfeeding mothers, 85% reported needing help with the first feed and 75% needed help with later feeds also; 55% always got the help they needed, but the level of help available varied considerably, with 1% receiving no help at all, despite a requirement that the midwives should help all the mothers.

Predictors of breastfeeding at six weeks were found in the speaker's Breastfeeding Best Start (BBS) randomised controlled trial: mothers who had been breastfed themselves, and early feeding after the birth (Inch, Wallace et al. 2003 and 2004). Failure to feed for six weeks was predicted if the baby was even slightly unwell after the birth, or was given other than the mother's milk at the breast in hospital or at home (30.3% the mother's expressed breast milk , 9.5% donated breast milk and 60.2% formula) or if the mother did not receive the help she needed.

Antenatal advice is often lacking. Even in a trial of midwifery care breastfeeding was not always initiated in the delivery suite; help received is often insufficient. Supplementation of 25% of so-called breastfed babies reduces the likelihood of maintained breastfeeding. The training of 104 midwives was assessed before and after four hours of training on hands off positioning and attachment care, and compared with the knowledge of student midwives who did not receive the same training. Although the midwives' training was effective there was great variability in their basic knowledge, which was no better than that of the students.

As first steps in achieving better breastfeeding numbers, turn to the Baby Friendly initiative, improve training and reduce poor hospital and community health practices. The hard to reach mothers need lay, peer and professional support ante- and postnatally; individual, group, buddying and phone support should be available, and occasional radio or newspaper stories about "breast is best" are helpful.

But why isn't doing all this enough?

Breastfeeding is championed by front line staff with limited resources and influence. It is complex, and influenced by social, employment and environmental factors that PCTs cannot control. Big shifts in voluntary behaviour such as breastfeeding require many and consistent messages and behavioural incentives for staff as well as mothers.

I recommend a social marketing approach.

Data is needed to identify the target group: is it all pregnant women and new mothers, or deprived sub-groups? This requires audit and prospective data, with agreed definitions and routine data collection at all key times. Interview women and their partners about support they need, which will differ significantly for example between white British and immigrant groups; assemble new or existing data such as patient surveys, comments collected at antenatal classes, and from feeding diaries and focus groups. Discover the views of staff on their training needs, and make a formal assessment of their skills.

Develop a strategy with senior and expert buy-in. Establish a powerful steering group with representatives from the PCT, public health, Sure Start and employers, for example the Chamber of Commerce. Involve a GP champion, a breastfeeding adviser, and the heads of health visiting and midwifery services; set up connections to young people, their education and early employment, and to mothers and other organisations such as the NCT and the La Leche League; to academics and experts on infant nutrition, public health evidence, and communications.

Mothers need to hear not just breast is best, but that breastfeeding is convenient and aids bonding, based on the views of local cultural groups and upon the aim to achieve pain-free breastfeeding and expression. Staff must be equipped with effective breastfeeding skills, built on a training needs analysis and shaped to the level required. Typically this would be 18 hours for local trainers (known to be effective), 4-hour workshops for midwives and health visitors, workbook and video for support staff, and video for GP receptionists. Ensure that the message is consistent across strategies, for example in teenage pregnancy work, homelessness, and prisons.

Messages to women should reduce barriers, highlight benefits, build up self-efficacy and confidence to seek support, and build skills to achieve competence in positioning, attachment, expression, and demand feeding, based on psychological social cognition theories. The message needs also to inform health promotion and personal and social education in schools and school nursing, ante- and postnatal care, health visiting, and support groups. By the use of peers and high-profile champions in the media social approval is boosted. Standards have to be achieved, practices reviewed, training needs established and training and ongoing development provided in all the affected services.

Regular audit should be instituted on data quality and definitions, and on manuals on supplementation, Skin 2 Skin, and policy. General and breastfeeding leaflets have to be updated and new materials produced: for example policy guides and health service leaflets on breastfeeding mothers at work. Target employers: for example the PCT through the Department of Health's Improving Working Lives initiative, and employers' forums, with media coverage.

Adjustments can be made in data quality, training needs, and messages to mothers using the feeding diaries of new mothers; data from health visitor records; the suggestions board in the Baby Café; the views of the Patients Council, and those of maternity liaison committees.

Skills can be evaluated using the methods of project management and advice from research departments. Examples are the Coventry University Breastfeeding Assessment (CUBA, under development) and assessment of the impact of media campaigns by coverage.

Breastfeeding strategies should not be documents, but live plans developed by key players involving power brokers in the NHS and the local community at all stages. Existing information sources must be used to develop unique tailored approaches which build on and contribute to research evidence of what works. Now is the time to speak to PCTs, who, having targets to achieve, are very grateful for advice from front line workers.


Inch S, Law S and Wallace L.M. (2003) Hands Off! The Breastfeeding Best Start Project (1). The Practising Midwife 6(10): 17-19.

Inch S, Law S and Wallace L.M. (2003) Hands Off! The Breastfeeding Best Start Project (2). The Practising Midwife 6(11): 23-25.

Alder, E.M., Wallace L.M., and Dunn O.M. (2003) Infant feeding choices: a psychological explanation? Journal of Reproductive and Infant Psychology 21: 239-240

Shaw R, Wallace L, Cook M, Phillips A (2004) Perceptions of the Breastfeeding Best Start Project. The Practising Midwife 7 (1) 20-24.

Biological nurturing - a new approach to breastfeeding
Philipa Parrett, Breastfeeding counsellor, trainee tutor and lactation consultant, The National Childbirth Trust, and Suzanne Colson, lactation midwife, East Kent Hospitals NHS Trust.

“The established approach in our society is to think of birthing as something done while lying down. We try to give a birthing woman freedom to find the right position for her own needs and comfort” (Odent, 1984).

British mothers are some of the least likely in Europe to sustain breastfeeding, despite almost twenty years of public health strategies to promote the biological choice. The established approach suggests a fixed system of verbal instruction, with midwives getting mothers to position themselves (back upright at right angles to lap) to attach the baby onto the breast correctly. Knowledgeable support may be crucial in overcoming common problems leading to unintentional early weaning.

Biological nurturing is designed to facilitate breastfeeding, with the aim of avoiding difficulties, but also as a problem solving strategy. Parents are shown a video and given a booklet illustrating positions where newborn babies attach themselves to the breast; mothers are encouraged to adopt any position where they feel comfortable. Photographs illustrate how to place the baby prone against a body contour. Midwives are trained to assess milk transfer using their observations of hormonal complexion, nutritional physiology and counselling techniques.

Research has shown that biological nurturing can support mothers who might otherwise give up breastfeeding in the first two postnatal weeks, and this applies importantly to preterm and small for gestational age (SGA) babies.

The theoretical framework, introducing freedom of maternal positions, a range of feeding states, applied anatomy and nutritional physiology, comprises kangaroo mother care (KMC), Skin 2 Skin (Anderson 1999), infant behaviours (Widstrom 1987; Righard & Alade 1990) and anatomy and physiology (Howie 1985; Woolridge 1986; Nissen at al. 1996; Uvnas-Moberg 1996; Odent 1999).

Clinical evidence includes the CATCH Department of Health practice development project, and observations of some healthy preterm and SGA babies being exclusively breastfed from birth (De Rooy and Hawdon 2002; Colson, De Rooy and Hawdon 2003). Oxytocin modifies sexual, maternal and social behaviours; it has an antistress effect, each suckling episode being followed by a fall in blood pressure. Breastfeeding mothers are calmer, correlating with oxytocin concentrations (Uvnas-Moberg 1998). Maternal concentrations of oxytocin are higher immediately after birth than at any time during labour (Uvnas-Moberg 1999). An increase in oxytocin release on the second day postpartum is associated with longer duration of breastfeeding (Nissen et al. 1996). Prolactin directs maternal love toward the baby (Odent 1999); its blood levels peak within 30 to 45 minutes of the start of a breastfeed (Howie 1985). Mothers delivered by caesarian section lack a significant rise in prolactin levels at 20 to 30 minutes from the start of a breastfeed (Nissen et al. 1996).

Oxytocin has been known as the love hormone for some time; it promotes sexual and maternal behaviours, aside from its mechanical actions in the birth process. Under the influence of oxytocin mothers become contented, disconnected from their environment, develop a facial flush, and the nipples become erect. A mother in this state can easily be disturbed by inappropriate speech, as by a midwife coming on duty and introducing herself. Of course, the state can be recovered. (A series of video examples demonstrated ideal biological nurturing positions, as when a mother lies supine, with her upper back and head supported by a pillow; a baby is seen to latch onto the breast although asleep. Inefficient or painful latching on can be improved by drawing the baby's lower lip down slightly).
A simple definition of nurturing is the way mothers show their love for their babies. In biological nurturing we invite mothers to place their babies in as much skin to skin contact as they wish, following the contours of their bodies. They are encouraged to cuddle their babies for as often and as long as they want, particularly in the first three days of life. The mothers and babies can be lightly dressed; fathers too can take part. An important component of biological nurturing is the midwife's hands off assessment. The baby is given the breast and latches on; she may fall asleep while latched on, and remains in a drowsy state while nonetheless feeding. Why would anyone hold a sleeping baby to the breast? Because they latch on and feed, and when they come off spontaneously, the breast is no longer engorged.
Breastfeeding promotes relaxation, rather than the other way about. A mother in this oxytocic state on the second day after her baby's birth is more likely than not to continue breastfeeding.

Contacting Suzanne Colson: email


Anderson G.C. (1999) Kangaroo care of the premature infant in E. Goldson (Ed) Nurturing the premature infant: Developmental interventions in the neonatal intensive care nursery 131-160 New York: Oxford University Press

Colson S., DeRooy L., Hawdon J. (2003) Biological Nurturing increases duration of breastfeeding for a vulnerable cohort. MIDIRS Midwifery Digest 13:1 92-97

DeRooy L. and Hawdon J. (2002) Nutritional factors that affect the postnatal metabolic adaptation of full term small and large for gestational age infants. Pediatrics 109(3) 1-8

Howie P.W., (1985) Breastfeeding - a new understanding. Midwives Chronicle and Nursing Notes July 184-192

Nissen E., Uvnas-Moberg K., Svensson K., Stock S., Widstrom A.M. & Winberg J., (1996) Different patterns of oxytocin, prolactin but not cortisol release during breastfeeding in women delivered by caesarean section or by the vaginal route. Early Hum Dev 45:103-118.

Odent M., (1984) Birth Reborn. New York: Pantheon Books (first American edition)

Odent M., (1999) The Scientification of Love. London: Free Association Books

Righard L. and Alade O.M., (1990) Effects of delivery room routines on success of first feed. The Lancet 336:1105-07

Uvnas-Moberg K., (1996) Neuroendocrinology of the mother-child interaction. TEM 7 4:126-131

Uvnas-Moberg K., (1998) Antistress pattern induced by oxytocin. New Physiological Science 13:22-26

Widstrom A.M., Ransjo-Arvidson A.B., Matthiesen A.S., Winberg J., & Uvnas-Moberg K., (1987) Gastric suction in healthy newborn infants. Acta Paediat Scand 76:566-572

Woolridge M.W., (1986) Aetiology of sore nipples. Midwifery 2:172-76


Q: Promoting instinctive behaviour must be a good thing, but is there a place for instruction when there are problems with attachment? A: First encourage the biological approach. Teaching will interfere with the cognitive process and there will be a loss of oxytocin. 80% of the time getting mothers to hold their babies will lead to breastfeeding. Of course some mothers will need help; it may be as simple as bringing down the lower lip.

Q: Why is data collection in England so far behind? In Scotland the Baby Friendly initiative has taken off well. A: Computer systems are not properly geared up for collection of data on breastfeeding. The expense of Baby Friendly is a deterrent to action, and it is given low priority in midwifery departments, even when PCTs have taken it on board. There is a failure to follow through after midwife training.

Q: What is the evidence that breastfeeding aids bonding? The claim reflects adversely on those who choose to bottle-feed. If we are trying to promote breastfeeding our messages must be based on hard evidence. A: We know that the all too common situation of distressed breastfeeding disturbs mother-baby attachment, which is likely to be improved by a close physical relationship between the two. Some American research seems to show that breastfed babies are less likely than the bottle-fed to be the victims of physical abuse by their mothers. This is believed to be an oxytocin effect and it is thought that breastfed babies develop more oxytocin receptors in their brains and so are more likely to breastfeed successfully themselves, thus resulting in a cycle of better parenting. Of course one could not propose this as an argument in favour of breastfeeding.

Holding and cuddling stimulates the release of oxytocin; the idea is to get the baby into the mother's arms, and let the chemistry work itself out.

Q: How does biological nurturing help the woman who has inverted nipples and is producing no colostrum? A: There is no magic solution. Different ways of holding the breast may help, but holding the baby will turn on the oxytocin and the flow of milk, which the baby will lick.

It is disappointing to see how few midwives encourage natural feeding positions for mothers and babies: the mothers need their knees and hips to be flexed and/or to be lying on their sides. A: The best feeding position is likely to be the one a mother and baby find for themselves.

The other side of the coin is the need for government to clamp down on the advertising of artificial feeding. A: It comes down to power and control; the midwives can control breastfeeding, but the baby food companies control formula feeding. All the information on formula feed is supplied by the company; there is no independent source. It is little known how variable are the constituents of formula: it may include vegetable or fish oils as well as cow's milk, and the formula may be changed over time. The Food Standards Agency and the Department of Health need to intervene.

We are up against great market forces, and I do not expect much if any help from the Food Standards Agency or the Department of Health. A cultural change in the professions and the public is needed.

The only baby milk is mother's milk; the others should be called Baby Drink.

National evidence on the influence of maternity leave on the initiation and duration of breastfeeding is lacking; there is a lack of evidence based policy. Adequate maternity leave encourages mothers to initiate breastfeeding, since it relieves them of anxiety about when they must leave their babies and when to start supplementing.

A short maternity leave undermines women's confidence: they fear, often without good reason, that they will be unable to express enough breast milk to satisfy their babies.

Whereas the positive benefits of breastfeeding are commonly publicised, the same is not done to emphasise the negative effects of formula feeding. Many poor white people in this country lack the role models whom they can see breastfeeding successfully. There is a difficulty in effectively and tactfully conveying the information that breastfed babies develop a potential IQ which is superior to that achieved by formula fed babies. A: Since many women have no choice but to bottle-feed it behoves us to be very careful in discussing this issue. Great distress can be caused even among health professionals, who may themselves have been bottle-fed or have had to bottle-feed their own babies, when the point is raised.

The news of enterobacter contamination of powder formula may lead to big changes in feeding practice, and is an important political issue: our government is reluctant to differ from the American stance, which is to avoid criticising formula feeding, despite its known disadvantages to health such as obesity.

The multiple benefits of continuous support in labour, notably boosted self esteem, have a beneficial effect in promoting breastfeeding.

The chair ended the discussion with a plea for work to be done in helping fathers to support rather than oppose breastfeeding as is too often the case.