It appears that you are viewing this site with something other than a standards compliant graphical browser. This site will work and look better in a browser that supports web standards, but it is accessible to any browser or Internet device.
Basic Design   |  Skip to content  |  Site Map   |  Search   -   Print page
South Australian Government logo. Link to Government Ministers' web site. Department of Health logo. Link to web site.
Module Border Module Border
  What's New 

Module Border Module Border
  33f Breastfeeding 

General principles | Breastfeeding and tobacco | Breastfeeding and nicotine replacement therapy | Breastfeeding and alcohol | Breastfeeding and opioids | Breastfeeding and benzodiazepines | Breastfeeding and psychostimulants | Breastfeeding and cannabis | Breastfeeding and blood-borne viruses | Lactation advice | Last reviewed
Show as multiple pages

“This section on Breastfeeding draws on National clinical guidelines for the management of drug use during pregnancy, birth and the early development years of the newborn (2006) published by the Australian Government, and is used with permission. The text has been revised to take into account South Australian circumstances, and the new text may not necessarily reflect the views of the Australian Government".


Refer to the South Australian Perinatal Practice Guideline “Breastfeeding guidelines for women with alcohol, tobacco or other drug dependencies” for South Australian recommendations.

General principles

§         Most drugs diffuse into breast milk

§         The “dose” received by the baby is usually very low

§         Most drugs diffuse back from the milk to the mother’s bloodstream as serum levels fall and this should be considered in relation to the timing of breastfeeding


Mothers who are drug dependent should be encouraged to breastfeed with appropriate support and precautions.  In addition, it is now recognised that skin-to-skin contact is important regardless of feeding choice and needs to be actively encouraged for the mother who is fully conscious and aware and able to respond to her baby’s needs.

Level of evidence: Consensus


Comment: Breastfeeding is recognised as the best nutrition for the infant. It is also inexpensive and easier to prepare and deliver than other options. As with all mothers of newborns, breastfeeding is recommended, where possible, for drug-dependent mothers, with the cautions described in the following statements.


A harm minimisation approach to breastfeeding is recommended in these guidelines. Encouraging breastfeeding is preferred to avoiding breastfeeding, provided that:

  • The woman is informed about the likely effects on the infant of the drugs she is using (or may use) and
  • The woman is assisted to plan minimum exposure of the infant to the effects of these drugs.

Level of evidence: Consensus


Comment: In these guidelines, a ‘harm minimisation approach’ does not mean that the woman should be advised against breastfeeding.  In advising drug-dependent women with regard to breastfeeding, the specific potential risks in each woman’s individual circumstances should be weighed against the benefits of breastfeeding, and she should be informed of them.

Level of evidence: Consensus


Mothers who present with an ongoing unstable pattern of drug use should not be encouraged to breast feed. Ongoing frequent intravenous substance use, polysubstance use, intoxicated presentations, and failure to engage in treatment plans may compromise the safety of the infant. A review regarding child protection concerns may also be indicated.


Comment: There is very little evidence about the effects of most drugs, prescription and licit as well as illicit, when administered to an infant through breastfeeding.  Much more evidence is needed.  Hale (2004) provides an unofficial USA rating system for categorising the risk posed by drugs when administered to an infant through breastfeeding (See Appendix 7: Categorisation of drug risks in pregnancy and breastfeeding).   Hale’s recommendations may contradict Australian practice e.g. Cannabis receives a L5 rating (breastfeeding contraindicated).  As with all breastfeeding women, drug-dependent women should not wean rapidly.

Level of evidence: Consensus


Comment: The level of methadone in breastmilk is low when the mother is on a methadone maintenance program, and does not affect the infant’s blood level of methadone.



Breastfeeding and tobacco

  • Breastfeeding first, and then smoking, reduces the dose of nicotine in the breast milk

Minimal amounts of nicotine are excreted into breast milk and absorption of nicotine through the infant’s gut is minimal, but tobacco smoking can have other effects on breastfeeding that might indirectly affect the baby.  Women should be informed that:

  • milk production may be reduced by as much as 250 mL per day in mothers who smoke
  • mothers who smoke are less likely to start breastfeeding than non-smokers
  • that mothers who smoke tend to breastfeed for a shorter time.

Level of evidence: Consensus in British Medical Association 2004.


Comment: This information must be given to the woman in the context of discussing the substantial benefits to both the infant and mother of breastfeeding.  There may be broader psychosocial issues affecting the woman’s ability to breastfeed, and it would be helpful to assist the woman to identify and address these.


Breastfeeding and nicotine replacement therapy (NRT)

Women who wish to breastfeed while continuing to use nicotine replacement therapy should be advised to breastfeed first, then, as soon as possible after feeding, use one of the intermittent delivery methods of NRT (inhaler, gum, lozenge or sublingual tablet). This will maximise the time between use of NRT and the next feed, and reduce the baby’s exposure to nicotine.

Level of Evidence: Consensus


Comment: Nicotine is both water and lipid soluble and distributes rapidly to and from breast milk, but little is likely to be absorbed by the infant. As maternal plasma nicotine concentration rises and falls, the same occurs in breast milk. The mean elimination half-life of nicotine in breast milk is 95 minutes. Even if the mother is using a high level of NRT, the infant’s daily exposure (normalised for the weight of the infant) is less than 2 per cent of the exposure of the mother. It is unlikely that such low levels of exposure are harmful to the infant. In contrast, there is good evidence that exposure to environmental tobacco smoke is harmful to the infant. Therefore, providing NRT to the mother, if this results in her not smoking, is of great potential benefit to the baby.  The formulation of NRT used may affect the level of nicotine in breast milk. The nicotine transdermal patch provides a steady level of nicotine in plasma and therefore in breast milk and the mother has no control over the level of nicotine in the milk. Mothers who use intermittent delivery systems of NRT may be able to minimise the nicotine in their milk by prolonging the duration between nicotine administration and breastfeeding.

(Level of Evidence to support Comment:  III-2 Dempsey and Benowitz 2001)


Breastfeeding and alcohol

The Australian Alcohol Guidelines recommend a prudent approach to breastfeeding if alcohol is consumed  Women who are breastfeeding are advised not to exceed the levels of drinking recommended during pregnancy, and may consider not drinking at all.  If a breastfeeding mother wants to drink alcohol, it is suggested that she breastfeed before drinking alcohol, then wait a minimum of three to four hours after the last drink before breastfeeding again. In the event that the woman exceeds the recommended levels of drinking, it is suggested that she wait approximately three hours per standard drink consumed before breastfeeding again. She may consider expressing and storing breastmilk prior to drinking.


Comment: Metabolism of alcohol varies with individual differences, such as weight and liver function, making it difficult to be prescriptive about the amount of time needed for the mother’s blood alcohol to return to zero.  Alcohol does not remain in breastmilk, but diffuses back into the mother’s circulation as her blood level drops.

Consequently there is no need for her to express and discard milk as long as she waits until her blood alcohol returns to zero to breastfeed her baby again. Although there is very little research evidence about the effect of alcohol on the infant, there are reports that even low levels may reduce the supply of milk and cause poor feeding with irritability and sleep disturbance in the infant. See also Appendix 8: Australian Alcohol Guidelines:  pregnancy and breastfeeding for more information.


Breastfeeding and opioids

  • Opioid analgesics include buprenorphine, hydromorphone, pethidine, codeine, methadone, dextropropoxyphene, morphine, fentanyl, oxycodone
  • Opiates, part of the opioid  group, are a class of substances with morphine like effects that can be reversed by the specific antagonist naloxone
  • Opioids have a depressant effect on the maternal central nervous system and produce drowsiness, mood changes and mental clouding

Mothers who are stable on methadone treatment programs should be supported if they choose to breastfeed.

Level of evidence: Consensus


Comment: Women who are stable on methadone, but may occasionally use heroin, in a ‘one-off’ pattern, should be advised to express and discard breastmilk for a 24 hour period afterwards, then return to breastfeeding.  This is not an indication for ceasing breastfeeding. The mother should consider having a ‘safety plan’ for the infant on such an occasion. This may include expressing and storing breastmilk in advance, or preparing formula, as well as having a responsible adult care for the infant.  Mothers who are unstable, continuing to use short acting opioids such as heroin, or using multiple drugs, should be encouraged not to breastfeed, and attention should be paid to assisting them to stabilise their lifestyle.

Level of evidence: Consensus


Comment: An unstable pattern of drug use may raise child protection concerns (see Child protection issues).  The safety of buprenorphine is not yet established for breastfeeding. Women who choose to breastfeed while taking buprenorphine, and can make an informed decision, should be informed of the risks and supported in their decision. The amount of buprenorphine in breastmilk is small and considered to be clinically insignificant.

Level of evidence: Consensus


Codeine is available as codeine phosphate and is present in panadeine forte.  Codeine is widely used for postpartum pain.  A minority of mothers may rapidly metabolise codeine to morphine.  As a result, in a very small number of breastfed babies, central nervous system (CNS) depression, apnoea, and even death may occur (


If the mother shows signs of CNS depression (e.g., somnolent, groggy) after codeine use, the baby should be examined by a physician for signs of CNS depression.  Constipation in a breastfed baby may be a symptom of rapid metabolism of maternally administered codeine.


Breastfeeding and benzodiazepines

  • Benzodiazepines belong to the sedative / anxiolytics / hypnotics group of drugs.  Diazepam is also a muscle relaxant
  • They have a maternal CNS depressant effect, which is dose dependent – i.e. as the dose increases there is progression from sedation through hypnosis to stupor
  • Benzodiazepines cause respiratory depression, but this effect is minimal unless other CNS depressants are taken (e.g. alcohol and opioids)
  • Half-lives vary e.g. diazepam 43 hours, lorazapam 12 hours, temazepam 10 hours
  • Stop breastfeeding if the baby appears sedated or reluctant to feed, and seek medical advice

Potential risks should be weighed up against benefits of breastfeeding when the mother is using benzodiazepines. If a woman taking benzodiazepines wishes to breastfeed, she should be advised that she should not stop taking the benzodiazepines abruptly, but should undergo supervised gradual withdrawal if she wishes to cease use.  Women on short-acting benzodiazepines should be advised not to breastfeed immediately after taking a dose because of the dual risk of her falling asleep, potentially smothering the infant, and of the infant receiving a maximum dose and becoming excessively drowsy. If the mother does breastfeed while she is drowsy, she should be sure she is securely seated in a chair (not lying down), with the baby also well supported, so that if she falls asleep the baby will be safe (see Sudden unexpected deaths in infancy (SUDI)).

Level of evidence: Consensus


Comment: The safety of benzodiazepines in breast milk is not known. Ideally, pregnant women will have undergone progressive supervised withdrawal throughout the pregnancy (see Benzodiazepines) and will not be taking benzodiazepines while breastfeeding.


Breastfeeding and psychostimulants

Potential risks should be weighed against the benefits of breastfeeding when the mother is using psychostimulants. A mother who wishes to breastfeed should be supported in that decision, unless she is a regular user and is unstable, in which case she should be advised against breastfeeding. Breastfeeding mothers who use psychostimulants rarely or in binges, must be:

  • informed of the risks
  • educated in how to avoid the harmful effects to the baby, that is:
    • to express and discard the breast milk after psychostimulant use (not to simply stop breastfeeding)
    • to have a supplementary feeding plan ready for such eventualities
    • advised not to breastfeed for 24 hours after the use of amphetamines, ecstasy or cocaine.  Although cocaine has a shorter duration of action than amphetamines, the illicit drug may be mixed with other unknown substances, so a 24-hour delay is recommended.

Level of evidence: Consensus


Comment: Ecstasy is an amphetamine derivative. The half life is likely to be brief, less than eight hours, but dependent on dose. Because the structure is similar to methamphetamines it is likely that it is transmitted via breastmilk. It is not known when it is safe to reinstate breastfeeding after use but 24 hours should be sufficient.


Breastfeeding and cannabis

Potential risks should be weighed up against the benefits of breastfeeding. There is insufficient evidence to make an evidence-based recommendation about cannabis and breastfeeding. There is some evidence that cannabis is excreted in breast milk, but the effects on the infant are unknown.  Cannabis is a long acting drug, so advice to take the drug after breastfeeding (as for alcohol) is not useful.  Current advice given to women ranges from supporting the decision to breastfeed to advising against it. Heavy use of cannabis may pose a greater risk of transmission in breast milk, but this is not known.

Level of evidence: Consensus


Advice to mothers and others should be as for tobacco:  that is, smoke away from the infant, out of the house, and not in the car.

Level of evidence: Consensus


Breastfeeding and blood-borne viruses


Human immunodeficiency virus

Breastfeeding increases the risk of transmission of HIV from mother to infant, particularly during the first 6 months. HIV-positive mothers should completely avoid breastfeeding and use formula milk instead. It is important that women who are not breastfeeding be informed of the benefits to the infant of skin-to-skin contact.

Level of evidence: III-2


Comment: Replacing breastfeeding with formula milk is a safe practice in Australia, where safe water and good quality infant formula are readily available. The role of antiretroviral therapy during breastfeeding is yet to be determined in communities where formula feeding carries a substantial risk.


Hepatitis C virus

There is no evidence that breastfeeding increases the risk of transmission of hepatitis C from mother to infant.  Women should be informed of the theoretical risks and discard breast milk if it may be contaminated with blood, such as by cracked, abraded or bleeding nipples.

Level of evidence: III-2


Comment: While encouraging HCV-positive women to breastfeed, it is essential that the woman make an informed decision. The information that should be provided includes: 

  • that the virus does appear in breastmilk
  • that (in the absence of HIV co-infection, which can increase HCV viral load) the risk of transmission appears to be small
  • that transmission may depend on viral load
  • that transmission is not via the gastrointestinal tract, but is blood-borne.


Hepatitis B virus

There is no evidence that breastfeeding increases the risk of transmission of Hepatitis B from mother to infant.  To protect against transmission it is extremely important that all infants of HBsAg (hepatitis B surface antigen) positive mothers receive active and passive immunisation within 12 hours after birth.

Level of evidence: III-2


Comment: Although HBV DNA and HBsAg have been detected in breast milk, no additional risk with breastfeeding has been demonstrated.


Lactation advice

Advice should be sought from a child and family health nurse, a lactation consultant or a midwife with drug and alcohol experience where there is uncertainty about how to advise the drug-dependent mother with regard to breastfeeding.

Level of evidence: Consensus


First developed:  18 March 2004     Subsequent updates:  merged with national guideline 03 March 2009

Last reviewed: 03 March 2009

South Australian Perinatal Practice Guidelines. Please read our disclaimer.