Birth control
New research shows babies in midwife-led births have poorer outcomes than those in doctor-led care.
She’s a former midwife in training to become a doctor. A few days ago, she published public health research that lobbed a grenade into the historically war-torn territory of New Zealand’s maternity care. It found babies delivered by midwives are at a higher risk of poorer outcomes than those delivered by doctors. And yet, Ellie Wernham says, she’ll still choose a midwife over an obstetrician if she has another baby. Her decision is a useful summary of the expert reaction to the new data she’s produced – no one is throwing the reform baby out with this lot of bathwater, anyway. Yes, our maternity system could be doing better – some say much better. But it isn’t performing badly enough to suggest that the hard-fought battle of the feminist social movement to overturn doctor-led maternity care has been a costly and harmful failure.
The mid-1990s reforms, championed by Helen Clark when she was Minister of Health, handed midwives control over the delivery of most of the country’s babies. Later funding changes ultimately led to nearly all GPs abandoning obstetrics – it’s estimated fewer than 15 nationally still deliver babies. Wernham’s report, which examines the outcomes of more than 240,000 births between 2008 and 2012, is the first detailed information on the effect of that change.
The College of Midwives, according to the minutes of its national meeting in August, initially dismissed the research as “flawed”, “poor” and “based on wrong assumptions”. It now accepts the work was done “in a very robust way” and, publicly at least, has taken no issue with the methodology.
The worrying results show babies delivered by midwives are 55% more likely to have oxygen deprivation during delivery, have a 39% higher chance of neonatal encephalopathy (abnormal brain function commonly caused by asphyxia at birth) and have a 48% higher chance of a low Apgar score, which indicates a baby is unwell.
Wernham, who began the research as a master of public health student at the University of Otago,
Wellington, agrees the findings are
Babies delivered by midwives are 55% more likely to have oxygen deprivation during delivery.
alarming, but acknowledges that adverse outcomes are rare. The study says although the results may raise questions about some aspects of the safety of midwife-led care, research also supported the “many positive aspects”, including patient satisfaction and lower intervention rates. She believes, though, that our maternity system would be better if GPs were still involved, “because it provides another avenue of choice for mothers”.
She agrees with calls for new midwives to spend their first year of practice in a hospital – a choice she opted for – rather than being able to work independently in the community.
In her hospital year, she says, she saw far more complications and births that went wrong than she had in her training. “It was mentioned during training that one of the things that might occur if we worked in a hospital was that we’d be more set in a medicalised model of birth. The midwifery philosophy has always focused on the normal, and that care should be provided in a primary care setting. My personal experience and belief around the first new graduate year are that in order to really understand that something is normal, you have to have a good understanding of what is not normal.”
Her co-author and former supervisor, health epidemiologist Professor Diana Sarfati, says the researchers are trying “not to inflame anyone”, but adds the paper raises important issues that need to be addressed.
“WE HAVE A SAFE MATERNITY SERVICE”
Ministry of Health advisers told the Listener they were surprised and concerned by the findings, but say the paper reinforces the view that, overall, our maternity system is a good one. They’ve asked the National Maternity Monitoring Group to recommend what further research needs to be done to understand what the disparities in outcome actually mean, and more importantly, why they are happening. They expect the group to report back in November.
The monitoring group chairman, Wellington obstetrician John Tait, acknowledges that although opponents of the 1990s reforms could use the paper to say, “I told you so”, “we can still reassure the women of New Zealand that we have a safe maternity service. Our perinatal mortality statistics are comparable with similar countries, and your expectation would be that if we were running a bad service, we would have bad mortality.”
There’s little point in anyone trying to use the research to push for a return to doctor-led care, because practice has changed so much in the past two decades, he says. Whereas in the past, the relationship between obstetricians and midwives was fraught, in the past eight to 10 years, it has been “very good”.
“The old-fashioned fight over the end of the bed, which did probably have an adverse impact on both mothers and babies, has now disappeared.” In delivery suites, obstetricians and midwives work very well together, he says.
“Depending on how this is presented, there is a possibility there will be huge disquiet in the community and we may be scaring people into changing the model – and there would not be a model for them to change to. If someone says, right, we’re all going to go and see an obstetrician, well, there aren’t enough. Could you ever get GPs back into doing maternity care? I think probably not. Are you going to get a whole lot more private obstetricians? Probably not. What we have to do is make the system we have work as well as possible. But this really gives us an opportunity to get some decent research into why this has happened.”
A key area for investigation, he believes, is whether there are differences between the doctor and midwife groups in the time taken to deliver the baby after the carer recognises a problem has arisen. In the doctor-led group, the person who can intervene is also the one making the decisions.
“The obstetrician says, ‘There’s a problem, bang, I do something.’ The midwife says there’s a problem, but there’s immediately a delay while they find someone to help. But you couldn’t have a system where you have an obstetrician standing outside every
Although opponents of the 1990s reforms could say, “I told you so”, “we can still reassure women that we have a safe maternity service”.
delivery room.”
Because the research categorised the midwife-led and doctor-led groups according to who was first booked as the lead maternity carer, it was impossible to know how many in the midwife-led group had an obstetrician involved in their patients’ care.
“There will be an awful lot of those, because we run at a 30% caesar rate, and add 12% forceps. There would be obstetricians who will see women at an antenatal clinic, or be called in just to have a look at a trace [during labour].”
It’s also possible there might be differences in outcomes for women in urban and rural centres, and for babies delivered in primary rather than secondary or tertiary units.
RESULTS SPARK DISAGREEMENT
The College of Midwives believes the results can be explained by resourcing, with chief executive Karen Guilliland saying the study highlights that private maternity care is better funded and resourced than the public system. “Most of our maternity hospitals are understaffed and often struggle to provide immediate response when midwives request medical input.”
This was why the college had taken legal action to ensure equity of funding for midwifery-led maternity care. The case is currently adjourned for mediation.
Lesley Dixon, the college’s midwifery adviser for practice and research, says she takes several calls each week from midwives worried about potentially harmful delays in timely medical interventions, and the problem seems to be getting worse, particularly in rural areas. She says the college is encouraging and supporting midwives in taking up the issue with district health boards. “Hospitals are short of obstetric and midwifery staff, and women who go in under private care are having their care prioritised above and before the women in the public system. When you’ve got an obstetrician who is providing your care only, they can commandeer the theatres and do a caesarean section.” However, several other experts the Listener spoke
Lawton says the findings didn’t surprise her. “I think it is time for some swift action.”
to did not accept that.
Asked if it is true that women cared for by midwives are getting less timely interventions, former obstetrician Dr Ken Clark, chairman of the national DHB Chief Medical Officers group, says, “No, it’s not. Of course we would always like more resource, whether in obstetrics or any other part of the service, and I’m not saying there’s never a department in New Zealand that from time to time doesn’t have staff shortages, but generally, numbers of obstetricians, house officers and registrars are strong. We would need real evidence to feel that was the case.” He says it’s never been brought to his attention as an issue.
The research doesn’t suggest the 1990s reforms were a mistake, but he believes the system would be better if GPs had remained part of it, either as lead carers or just a part of the care. “We missed a trick, frankly. We got all the good things that were brought by having midwives intimately involved as lead carers, but we did it while excluding an important part of the workforce in GPs. If we could relive history, and we were smart, we would have done it so we had the best in all senses – midwifery-led care, but still with strong involvement of GPs and obstetricians in public and private.”
Dixon also pointed out that although the research separated the women into whether they registered first with a midwife or a doctor, the researchers “were unable to work out who was providing care during labour and birth. We have a very integrated system. Midwives provide care through the antenatal period and have obstetric input during the antenatal period, labour and birth. Consultant obstetricians have a midwife working with them providing care, so it’s very difficult to say this is purely midwife- versus purely medical-led care, because both are involved.”
But Charlotte Paul, emeritus professor of preventive and social medicine at the University of Otago, who reviewed the paper, says the comparison is valid. “They are interested in models of care, so midwife-led care includes a handover or consultation if needed with an obstetrician, and if that handover worked well on both sides, you’ll get lower birth complications.
“I hope these results will lead to serious consideration of what might be the reasons for the difference, but not alarm, because obviously we still have pretty low rates. It is evidence that potentially we could do a bit better and we need to find out where we could do better.”
“TIME FOR SOME SWIFT ACTION”
For advocacy group Action to Improve Maternity, set up to push for improvement in the quality and safety of the maternity system, the college response, along with the research itself, is no surprise.
“We know their knee-jerk reaction was ‘it can’t be true’,” says founder Jenn Hooper. “Now they have discovered it’s not flawed, their second reaction will always be to blame someone else.”
She says Action to Improve Maternity has helped more than 700 families and not one has complained about a delay in medical intervention in hospital after a midwife flagged an issue. “Instead, there has been lack of transfer, of recognising the severity of issues, a lack of communication and communicating the sense of urgency, specifically by independent, self-employed midwives. There are families that these numbers represent, families who have been avoidably hurt, a lot of them permanently.”
The College of Midwives has no evidence on which to base the contention that district health board under-resourcing is the issue, says Hooper. “Just like I can’t say that it’s perhaps because of inadequate education and training programmes and a lack of internships, or the fact that the midwifery first-year-in-practice programme has never been properly tested or independently reviewed. We both may have an opinion and I would probably go so far as to bet the
farm I’m right.”
The college is adamant she’s not, despite controversial research published last year by Associate Professor Bev Lawton, a GP and director of the University of Otago, Wellington, Women’s Health Research Centre. It showed the risk of a perinatal death was 30% higher when midwives were in their first year of practice, compared with those who had five to nine years’ experience and those who had nursing as well as midwifery training. The data was from births between 2005 and 2009, and the college slammed the results at the time as out of date, laying a Medical Council complaint against Lawton, which wasn’t upheld.
Lawton says although Wernham’s research didn’t examine reasons for the disparity in adverse outcomes, when added to her own findings, “there is quite strong evidence that there are some clinical performance issues”. She says the findings didn’t surprise her. “I think it is time for some action, some swift action.”
She believes obstetric specialists should be on site, rather than on call, at night in all major hospitals, and that for two years after qualifying, midwives should be supervised and in hospital practice. Coroners have also recommended new midwives should first practise in hospital before being able to work independently in the community.
Lawton says there has been a rapid increase over the past 10-15 years in women who are older and have other birth risk factors and there’s now a mismatch between midwifery training and the patient group. “The training at the time was quite rightly designed for more low-risk, natural births.”
Wernham’s research took risk variables into account, adjusting for differences in the midwife-led and medical-led groups of age, body mass index, smoking, blood pressure, diabetes, trimester of registration, deprivation and ethnicity.
“THE STUDY IS ROBUST”
Nonetheless, some have questioned whether other differences between the groups, that weren’t taken into account, might explain the outcomes.
Sarfati believes not. “No epidemiological study based on routine data will ever be perfect, but none of the possible issues they are likely to raise explain the very clear findings. We were extremely careful in our approach.” International experts who had been approached to critique the research “all came back saying the study is robust and the results are concerning”.
A perspective by Dutch midwifery researchers published alongside Wernham’s study in PLOS Medicine said it did not control for place of birth or distance to hospital, and most rural care was provided by midwives.
Sarfati says the Dutch researchers were among those who reviewed the paper and raised that issue in their review. “We countered it, and our counter was accepted by the journal.” While distance to hospital couldn’t be established, they examined how much unmeasured confounding of any type would be needed to explain away the findings. “The amount was completely, unfeasibly large. It would have to be some risk factor, or group of risk factors, which increased the risk of adverse outcomes by 400-500%. It would have to be extremely common in the midwife-led group and extremely rare in the medical-led group and unrelated to all the other factors we adjusted for.”
The coroners have said new midwives should first practise in hospital before being able to work independently.
One of the strongest indicators that the disparity in outcomes is not simply related to differences between the groups of women is the fact that the researchers included a “negative control” in their model – babies who were small for gestational age (SGA).
SGA is strongly associated with other risk factors, such as ethnicity, smoking and deprivation, but largely unaffected by what a carer does. The odds of having an SGA baby was identical in both groups, once these other risk factors were accounted for. Had there been an apparent relationship between the model of care and SGA babies, this would have suggested that the differences between the groups had not been dealt with sufficiently. This was not the case.
“This is an important health issue,” says Sarfati. “We have done it in good faith and this is the best possible research that could be done at this time. We just have to hope there are enough people out there who are prepared to take it that way, rather than either dismiss it or acting defensively; that we will get something positive in the long term.”
“PLEASE DON’T PANIC”
Auckland obstetrician Dr Sue Belgrave, chairwoman of the Perinatal and Maternal Mortality Review Committee, says her message to women is “please don’t panic”. She says the data doesn’t necessarily reflect the complexity of the care and inter-relationships during pregnancy and birth.
Although the perinatal and neonatal mortality rate has not changed over the period reflected in the last PMMRC report (200714), there has been a significant reduction in stillbirths.
“There is a whole lot of perinatal mortality we haven’t been able to do anything about as yet, including things like congenital abnormality and very preterm labour.”
Speaking personally, she feels too much could be read into the results. “The worrying thing is, yes, your readers will look at them and say, ‘I must go and have a private obstetrician because it’s clearly safer. I’d like to be a little more cautious about that.”
But Christchurch GP Lynda Exton, who co-founded Action to Improve Maternity and wrote the 2008 book The Baby Business that linked the 1990s reforms with a raft of worrying trends, describes the findings as a “huge, huge concern”. “My book described the maternity system as scandalous, and it hasn’t improved.” She says it’s been “completely and comprehensively overwhelmed by the philosophy of birth as a normal life event, rather than one focused on ensuring a healthy mother and baby at the end”.
She says the Ministry of Health should have been monitoring comparative out- comes of doctors and midwives since the lead maternity carer system was introduced in 1996, but it had not.
Dixon, from the College of Midwives, says although more investigation is clearly required, there’s already research suggesting that “if we can keep healthy women out of hospitals and in small, freestanding primary units, that helps to protect them, so the women who need secondary services and obstetrician input get that without having a lot of other women around”.
However, Exton, a former GP-obstetrician, says the popularity of primary birthing units has declined rapidly. The 2014 Report of Maternity said that between 2007 and 2014, the proportion of women giving birth at a tertiary facility had increased from 40.9% to 46.6%, while the number delivering at a primary facility fell from 15.6% to 9.1%.
“Most women are choosing to give birth where there is also a doctor, and no one has done research into why midwife-only staffed units are falling in popularity.” She says of the families who come to Action to Improve Maternity after problems with births, 90% of those births occurred at home or in primary birthing units.
“OUR SYSTEM IS PREDICATED ON CHOICE”
The Ministry of Health’s senior adviser for maternity services, former midwife Bronwen Pelvin, says she’s seen “enormous strides” in the collegiality of doctor-midwife relationships since she started practice in 1976. “Midwives and obstetricians work hand in glove. Neither can do what they do without the other group being present and engaged in the care of women.”
If women who read the research decided to go with an obstetrician as their lead maternity carer – although many will live in areas where they don’t have that option – that’s their call, Pelvin says. “Our whole system is predicated on choice, but for everyone who doesn’t make that choice, we have secondary services in place and referral guidelines that do work. We know referrals are made to obstetricians in district health boards, and that advice is given back to lead maternity carers on how to manage women’s pregnancies and keep them healthy and well. If they’re not healthy and well, the obstetricians will take responsibility for providing that care. That is how the system should work. We have a highly integrated system and it does work well for the majority of people.”
She says Wernham’s results are particularly surprising given a Cochrane review in 2015 involving nearly 18,000 women in four countries that reported no differences in rates of adverse outcomes and numerous benefits with midwife-led care, including fewer preterm births, fewer interventions and a higher chance of vaginal delivery.
“That’s one of the reasons the outcome of this study is so unexpected. That’s why we need to go back and look at why.”
The system would be better if GPs had remained part of it, either as lead carers or just a part of the care. “We missed a trick, frankly.” “My book described the maternity system as scandalous, and it hasn’t improved.”