Intended for healthcare professionals

Clinical Review

Breast feeding

BMJ 2008; 336 doi: (Published 17 April 2008) Cite this as: BMJ 2008;336:881
  1. Pat Hoddinott, senior clinical research fellow and general practitioner 1,
  2. David Tappin, clinical senior lecturer in child health2,
  3. Charlotte Wright, professor of community child health2
  1. 1Centre for Rural Health, University of Aberdeen, Inverness IV2 3BL
  2. 2Paediatric Epidemiology and Community Health Unit, Section of Child Health, University of Glasgow, Royal Hospital for Sick Children, Glasgow G3 8SJ
  1. Correspondence to: P Hoddinott p.hoddinott{at}
  • Accepted 10 March 2008

Clinicians know that breast feeding is crucial to infant health in developing countries, but they may be less aware of the potential longer term health benefits for mothers and babies in developed countries, particularly in relation to obesity, blood pressure, cholesterol, and cancer. The World Health Organization (WHO) recommends exclusive breast feeding (breast milk only, with no water, other fluids, or solids) for six months, with supplemental breast feeding continuing for two years and beyond. Governments in the United Kingdom have adopted this recommendation, but it presents an enormous challenge for countries like the UK and the United States, where breast feeding rates have been low for decades and can seem remarkably resistant to change. In this review, we will focus mainly on developed countries, with reference to the global context. We will summarise the evidence for the beneficial effects of breastfeeding on health, discuss the epidemiology, and provide practical guidance for managing problems associated with breast feeding. We highlight new developments in infant growth charts and current controversies around HIV and donor breast milk.

Sources and selection criteria

We searched several databases—including Medline and Embase—using the keywords “breastfeeding”, “breast-feeding”, “breast feeding”, and “infant feeding”. We also searched Issue 4 2007 of the Cochrane Database of Systematic Reviews, National Institute for Health and Clinical Excellence guidelines, World Health Organization systematic reviews, Clinical Evidence, and personal reference archives.

How does breast milk differ from formula milk?

Formula milk is just a food, whereas breast milk is a complex living nutritional fluid that contains antibodies, enzymes, and hormones, all of which have health benefits. In addition, some methods of delivering formula milk expose the baby to serious risks of infection. Early intake of colostrum, which is rich in antibodies, is especially important in developing countries, and the small volume of colostrum helps to prevent renal overload when the newborn baby is adjusting its fluid balance.

What are the health benefits of breast feeding?

Tables 1 and 2 summarise the short term and long term health benefits for the infant and mother (taken from two evidence based reviews).1 2 Caution is needed when assessing evidence from observational studies in high income countries, as these are prone to bias and confounding by educational and socioeconomic factors. For low birthweight infants (below 2500 g), evidence from systematic reviews shows that breast milk reduces mortality and morbidity and has a beneficial effect on neurodevelopment and growth.3 4

Table 1

 Short term and long term health benefits of breast feeding for the child in developed countries3 4

View this table:
Table 2

Short term and long term health benefits of breast feeding for mothers in developed countries3 4

View this table:

In the developing world, low immunisation rates, contaminated drinking water, and reduced immunity as a result of malnutrition make breast feeding crucial to reducing life threatening infections. A review of interventions in 42 developing countries estimated that exclusive breast feeding for six months, with partial breast feeding continuing to 12 months, could prevent 1.3 million (13%) deaths each year in children under 5.6 In comparison, Haemophilus influenzae type b vaccine could prevent 4% of all child deaths and measles vaccine 1% of such deaths. Breast feeding also suppresses ovulation, so that women who are still breast feeding are less likely to become pregnant than those who are not breast feeding.

In the UK millennium cohort survey of 15 890 infants, six months of exclusive breast feeding was associated with a 53% decrease in hospital admissions for diarrhoea and a 27% decrease in respiratory tract infections each month; partial breast feeding was associated with 31% and 25% decreases, respectively.7 The results of this study suggested that the protective effects wore off soon after breast feeding ceased, contrary to smaller cohorts, which have reported benefits for up to seven years.8

How have breastfeeding rates changed?

Modified cows’ milk was first manufactured at the end of the 19th century and subsequently breastfeeding rates started to fall, reaching an all time low in developed countries in the 1960s. Worldwide, exclusive breast feeding until 4 months of age (fig 1) seemed to rise from 48% to 52% during the 1990s. The low quality of comparable robust data worldwide is a problem, however. In 2005, the prevalence of exclusive breast feeding until 4 months was 7% in the UK,9 in contrast to 64% in Norway, a comparable developed country. The proportion of babies who are breast fed initially (even for just one feed) has increased steadily since 1990. Older, better educated mothers, who do not smoke, and who have higher socioeconomic status are more likely to breast feed, as are mothers who have previously breast fed or who were breast fed themselves.9


Fig 1 Exclusive breastfeeding at 4 months in 1995-2000 (from Unicef data). Data come from different years and different sources, not all of which are comparable

Of concern, the biggest decline in breast feeding occurs during the first four days after birth, when 12% of women in the UK stop, with 22% stopping by two weeks and 37% by six weeks (fig 2).9 Early skilled help is extremely important, as nine out of 10 mothers say they would like to have breast fed for longer.


Fig 2 Prevalence of breast feeding up to the age of 9 months in 20059

Breastfeeding practices vary across different cultures—for example, around 50 cultures withhold colostrum from babies in the first 48 hours.10 Second and subsequent generations of immigrants are beginning to adopt UK customs, with a consequent decline in the number of women who start breast feeding and the duration of breast feeding.11

What interventions might increase breastfeeding rates?

Three Cochrane reviews of randomised controlled trials of interventions to promote and support breast feeding and a National Institute for Health and Clinical Excellence (NICE) review cover this topic.12 13 14 15 Interventions tailored to particular cultural or socioeconomic groups and multifaceted interventions seem to be most effective.12 15 However, the overall quality of trials is poor, health system and cultural contexts are often not comparable, and interventions are heterogeneous.

Summary points

  • The best option is exclusive breast feeding for six months, with no solids or other fluids, and supplemental breast feeding for two years or more

  • Breast feeding has important health benefits, including reduced risk of infection in babies and reduced risk of breast and ovarian cancer in mothers

  • Breast feeding has potential long term health benefits in children—reduced blood pressure, cholesterol concentrations, and obesity

  • Early assessment of breast feeding and skilled help is the key to preventing problems

  • New WHO growth charts will establish the breastfed infant as the biological norm with which all children should be compared and will be applicable to all ethnic groups

During pregnancy

Results of five studies from the US (582 women) indicate that education during pregnancy can increase the numbers of women on low income who start breast feeding,12 but overall, evidence for effective interventions is lacking. A Cochrane systematic review looking at the effect of interventions during pregnancy on the duration of breast feeding is in progress.

In hospital

In hospital, early skin to skin contact between mothers and babies (30 trials, 1925 participants),16 frequent and unrestricted breast feeding to ensure continued production of milk (three old trials, new trials considered unethical),15 and help with positioning and attaching the baby (one trial, 160 women)15 increase the chances of breast feeding being successful (table 3). The NICE guidelines on postnatal care recommend the Unicef “baby friendly hospital initiative” is implemented as a minimum standard. This initiative is supported by evidence from several studies, including trials, and has an important health professional training component.10 15 Two trials (1431 women) have found that giving birth in a home-like environment increases the number of women who start breast feeding and continue breast feeding for six to eight weeks.17 Women spend less time in hospital after birth these days, and NICE has concluded that this does not affect the duration of breast feeding.10 18 Caution is needed, however, as breast feeding was a secondary outcome measure. Nine trials (3730 women) have provided convincing evidence that giving mothers commercial discharge packs containing formula or promotional material for formula milk reduces the number of women who exclusively breast feed until 10 weeks.19

Table 3

Prevention and management of breast feeding problems

View this table:

After birth

A Cochrane review of 34 trials (29 385 women) found that additional professional or lay support increases the duration of any breast feeding to six months, with a greater effect for exclusive breast feeding.13 Exclusive breast feeding was prolonged by WHO and Unicef professional training for health professionals (meta-analysis of six trials). Unicef UK has extended the baby friendly initiative to community healthcare settings; the effectiveness of this policy is still to be evaluated.

Policy interventions

Breastfeeding targets have recently been set in England and Northern Ireland, but their effect is yet to be evaluated. In Scotland, although rates increased, no health board achieved a 1994 target of 50% of babies being breast fed at six weeks by 2005. In 2006, when targets were no longer in place, breastfeeding rates declined in Scotland for the first time in 10 years. No trials have investigated support for breast feeding in the workplace. In Britain, only one in seven working mothers had the facilities to express milk or to breast feed at work.26 In 2005, the Breastfeeding (Scotland) Act made it an offence to prevent or stop a mother breast feeding a child under 2 years in public. In the same year a UK survey of 7186 mothers found that Scottish residents had the most positive experiences of breast feeding in public.9

What clinical problems arise when breast feeding?

Correct positioning and attachment of the baby at the breast (fig 3) are crucial to establishing and sustaining effective breast feeding. When a mother and baby are learning to breast feed, good practice is for a trained person to observe feeds and provide skilled help and support. Getting the first few feeds right can prevent problems like breast or nipple pain, poor milk supply, and early infant weight loss. Table 3 gives details of associated clinical problems.


Fig 3 Positioning and attachment of the baby on the breast. Adapted, with permission, from the Unicef UK baby friendly initiative

How should doctors prescribe for breastfeeding mothers?

Doctors tend to be overcautious when prescribing for breastfeeding mothers, and specific advice or subtle cues can undermine breast feeding.27 Careful use of expert resources (box), however, can usually enable breast feeding to continue. Each prescribing decision needs to take account of the risks and benefits to the individual mother and baby, including the indication for treatment, the pharmacokinetic properties of the drug, the age of the baby, the volume of feeds, and the frequency of feeds. Unfortunately, standard adult reference texts like the British National Formulary may be unhelpful. Drug manufacturers are not required to license drugs for use by breastfeeding mothers, and they tend to be cautious and recommend against use. Most published data on safety rely on case studies or small samples of fewer than 20 mothers. However, if a drug is licensed for infants, then the small amounts present in breast milk are likely to be safe, so the British National Formulary for Children is a better guide to maternal prescribing.

Guidance for prescribing in breastfeeding mothers

  • Most common conditions can be prescribed for safely using the information in this box—mothers need stop breast feeding only rarely

  • Prescribe drugs in the British National Formulary for Children that are licensed for use under age 2 years

  • Use drugs with a relative infant dose <10% of the maternal dose

  • Avoid newly developed drugs with little information available

  • Choose drugs that bind to plasma proteins because only small amounts of these drugs are transferred into milk

Drugs that should be used with caution and monitored
  • Some antiepileptics

  • Some antipsychotics

  • Central nervous system sedatives

  • Combined oral contraceptives (not advised for mothers with babies <3 months old)

  • Lithium

  • Diuretics

Drugs that should be avoided
  • Chemotherapy

Reference sources

Current hot topics

New growth charts based on breastfed babies

Until 2006, growth charts were based on children with mixed feeding patterns, predominantly bottle fed, but evidence from various studies suggested that exclusively breastfed infants gained weight differently. Concerns were that misinterpretation of growth charts could lead to breastfed babies being given unnecessary supplements of formula. This led WHO to develop new charts using data collected from six centres worldwide over 15 years. These are intended to be standards of optimum growth, rather than average growth. All data were from children born to non-smoking mothers in non-deprived circumstances who had been breast fed for a year, exclusively for four months, with complementary solids started by 6 months of age. The resulting data show an extraordinary similarity of linear growth between populations and confirm that breastfed infants show a lower weight trajectory from 6 months onwards.28

The Department of Health has recently recommended that the WHO charts be adopted for all children from 2 weeks to 2 years, with a planned launch by early 2009. This allows the UK to keep its valuable birth weight for gestation charts. The new charts will establish breastfed infants as the biological norm—with whom all children should be compared—and they will be applicable to all ethnic groups. After these charts are adopted fewer infants will be defined as underweight or weight faltering, whereas the proportion who are overweight will increase.29 A supporting educational programme will therefore be essential.

Tips for non-specialists

  • When opportunities arise, inform pregnant women and breastfeeding mothers of the health benefits of breast feeding their infant for six months

  • Encourage mothers and boost their confidence in their ability to breast feed

  • Prevention and early help with breastfeeding problems are crucial. Ensure that pregnant women and breastfeeding mothers know where they can get skilled professional or lay help 24 hours a day, seven days a week

  • If you do not have the skills to assess whether breast feeding is effective, refer the woman to someone who does have the skills and also has the time to observe breast feeds

HIV and breast feeding

WHO consensus guidelines for HIV positive women vary according to context, place, and the individual. Exclusive breast feeding for six months is recommended where no culturally acceptable, feasible, affordable, safe, and sustainable nutritional substitutes for breast milk are available. Otherwise, breast feeding should be avoided in an attempt to prevent new perinatal HIV infections. A Cochrane review of this subject is in progress.

Donor breast milk banks

Throughout history donor milk has been the choice of some parents, and it is currently recommended as second choice if the mother’s own milk is not available.3 However, the risk of possible transmission of HIV, cytomegalovirus, and Creutzfeldt-Jakob disease has recently caused concern about regulation of the 17 UK donor milk banks. Evidence on donor milk is limited and of poor quality.30 The extent to which pasteurised donor breast milk retains the biological properties of mother’s milk is uncertain. Further evidence about the health benefits and economics is needed to develop evidence based guidance.


A varied and balanced diet is recommended to sustain breast feeding, which requires about 2.09 MJ of extra energy a day. NICE guidance to improve the nutrition of pregnant and breastfeeding mothers is now available.31

Low vitamin D concentrations in residents of the northern hemisphere are a concern and recommendations about supplements vary between countries. Little vitamin D is secreted into breast milk, and NICE recommends supplements for all pregnant and breastfeeding mothers.31 A Cochrane review is in progress.

Additional educational resources for patients

Telephone helplines

The following organisations are a good source of information and they offer telephone helplines staffed by highly trained breastfeeding specialists; some offer chat rooms and support groups:

Good sources of information and video clips
Other good sources of information

Additional educational resources for health professionals

Where do we need to go now?

All health professionals should actively support breast feeding as an important way to improve child health. Better implementation of existing evidence—particularly the baby friendly initiative—is needed, as are improvements in the education of healthcare professionals. Adherence to WHO’s International Code of Marketing of Breast-milk Substitutes is also important in both developing and developed countries. New approaches are required at policy and individual level to deal with health inequalities, consider incentives to breast feed, facilitate breast feeding outside the home, and to find the most effective ways of teaching and learning breastfeeding skills. Meanwhile, the early days after birth are crucial and everyone in health care should chip away at the complex psychological, social, cultural, and health service organisation factors that undermine breast feeding.


  • Contributors: PH drafted the review. All authors helped collect data and write the paper. All authors are guarantors. Thanks to Jane Britten, Magda Sachs, Wendy Jones, and Linda Wolfson for their helpful comments on drafts of this review.

  • Competing interests: None declared.

  • Provenance and peer review: Commissioned; externally peer reviewed.


View Abstract