Experts believe breastfeeding is good for mother and baby
Drugs commonly used to treat bleeding after birth may hamper a woman's ability to breastfeed her baby, research suggests.
The study, which appears in the journal BJOG, suggests the drugs may impede milk production.
The Swansea University team also confirmed high doses of painkilling drugs have a similar effect.
The findings may help to explain the limited success of efforts to increase breastfeeding rates in the UK.
But an expert warned no firm conclusions could be drawn from the study.
The Department of Health recommends children are exclusively breastfed for the first six months, because of the health benefits it can provide for both mother and baby.
It has set a target of increasing breastfeeding rates by 2% each year.
But in the UK only 45% of babies are exclusively breastfeed just one week after their birth. One in four receive only formula milk from birth.
The Swansea team analysed data on more than 48,000 women who gave birth in South Wales.
They found use of the drugs oxytocin or ergometrine to cut the risk of haemorrhage was associated with an overall 7% decline in the proportion who started breastfeeding within 48 hours of giving birth.
Among women who were not given the drugs, two-thirds (65.5%) started breastfeeding within 48 hours of giving birth.
But among those given a shot of oxytocin the breastfeeding rate was 59.1%, and among those who were given an additional injection of ergometrine the rate fell to just 56.4%.
The researchers believe the drugs may hamper a woman's ability to produce milk.
This could mean that their initial attempts to breastfeed may end in frustration, and that many give up rather than persevering.
The link between painkilling drugs and lower rates of breastfeeding has already been established, and has led to revised NHS guidance on the use of epidurals in labour.
Lead researcher Dr Sue Jordan said more research was required, but based on the study's findings use of the clotting drugs could lead to up to 50,000 fewer British babies being breastfed every year.
She calculated that this could mean an extra 1,000 children becoming clinically obese, and 3,000 developing childhood asthma.
Dr Jordan said the study highlighted a pressing need for follow-up research.
She said: "The potentially life-saving treatments to prevent bleeding after birth must not be compromised on the basis of this study but further studies are required to establish ways to minimise any effects on breastfeeding rates.
"In the meantime, what we would like to see would be provision of extra help for new mothers trying to establish breastfeeding by making sure to allow enough time for the effect of drugs given in labour to subside."
Patrick O'Brien, a consultant obstetrician and spokesperson for the Royal College of Obstetricians and Gynaecologists, said the study was interesting, but more tightly-controlled follow ups were needed.
For instance, he said it was possible that women who refused drug therapy might also be more likely to breastfeed.
He said: "This small possible effect on breastfeeding has to be balanced against the beneficial effects of these drugs, which are known to reduce the risk of heavy bleeding after birth from 18% to 6%.
"At the moment, I would say the benefits of using the drugs outweigh the theoretical downsides."
Rosemary Dodds, policy research officer for the National Childbirth Trust, said women needed more support to start breastfeeding.
She said: "A lot of women are not given enough information about the medications that might be given to them during childbirth, and women at low risk of bleeding may not need to take these drugs.
"It is important that women understand the risks and can give their informed consent before they go into labour."