Babies born close to death are less likely to get life-saving treatment if they're Māori, Pacific or Indian. Nicholas Jones investigates how racial bias reaches back before the cradle.
For 21 days Alysha McVeigh lay still, aware movement would bring on labour and death for her unborn child.
She'd been rushed to Wellington Hospital after her amniotic sac bulged into her birth canal.
Aly was just 20 weeks pregnant. Hospital policy was to attempt resuscitation only on babies born after 23 weeks.
Even on total bedrest, the odds were against her unborn child lasting a few days, let alone 21.
She and husband Adrian named their daughter Tia-Jane. In a sterile hospital room they spoke to her about a life she might not see.
"I would talk with her about the days we had to go. We set up an email account where I would send emails of things I wanted her to know," Aly said.
"I told her about family members, hopes and dreams for her, what her name meant, how I was feeling."
Tia-Jane was an IVF baby, and her parents' last chance for a child together.
Ultra-cautious, they hardly left home in the early weeks of the pregnancy. Then, at the 20-week scan the sonographer stopped pushing the probe over Aly's belly and said, "Can you just wait here a moment?"
The problem: a shortened cervix, which greatly increases the risk of miscarriage or very premature birth.
Aly was soon in hospital and being told birth would most probably come too early for resuscitation to be attempted.
Even if it could be, there was a high chance Tia-Jane would die or be left with major disability. The couple declined termination.
Lying on an angle and trying not to move, Aly tried to put on a brave face for Adrian. But the minute-by-minute pressure of the situation was extraordinary.
• READ MORE: Māori babies less likely to be resuscitated
"Adrian washed my hair. He gave me showers. It's really embarrassing, but he'd empty my bag, because I had a catheter in for the entire time.
"I felt like I was losing my mind. You are lying there, you have no fresh air ... on that last day I said to Adrian, 'You've got to take me outside, I've got to get out.'"
Steroids sped Tia-Jane's production of a protein that prepares lungs for breathing. Then baby's foot went through mum's cervix and "hold her in" turned to "get her out".
Tia-Jane McVeigh was born at 2.36am on June 15 2015, just a couple of hours past 23 weeks and weighing 516 grams.
Extremely prem babies provide a window to the womb — Tia-Jane's eyes were fused shut, her eyebrows still on the side of her head, her skin translucent without pigmentation that comes later in gestation.
Adrian saw his just-born daughter whisked to a table. Doctors tipped back her head. Once. Twice. On the third try they got the breathing tube down. Oxygen inflated tiny, fluid-filled lungs.
"It was one of the scariest things I've ever seen," he said. "But amazing all at the same time."
Tia-Jane was born at the edge of viability, becoming New Zealand's smallest and most premature baby to survive.
Doctors almost always resuscitate from 25 weeks and never earlier than 23. Each baby in the in-between "grey zone" is a question — try for life, or give palliative care?
Parents make one of the hardest decisions of their lives, in shock and often with little time amid a blur of medical advice.
The chance of survival can be slim, and brings with it the possibility of conditions requiring lifelong care, including cerebral palsy, blindness and deafness.
Infants can endure weeks of invasive treatments only to die. The act of resuscitation is a trauma in itself, and may be a child's only, fleeting experience of the world.
Birth weight and steroids before delivery can make a crucial difference to the chance of survival.
A Weekend Herald investigation can reveal another, more disturbing circumstance: skin colour.
Babies of certain ethnicities are significantly less likely to be resuscitated — and experts think racial bias is at least partly to blame.
That bombshell discovery emerges from a close look at a decade of data for Kiwi babies born from 23-26 weeks.
Resuscitation was tried on 92 per cent of Māori babies, 89 per cent of Pacific, and 86 per cent of Indian. That compared to 95 per cent for "other", mostly Pākehā and non-Indian Asians.
The divides were outlined in a report to Health Minister David Clark, by the Perinatal and Maternal Mortality Review Committee, a taxpayer-funded watchdog tasked with reviewing deaths of babies and mothers.
"Māori, Pacific and Indian live born babies were statistically significantly less likely to have an attempt at resuscitation," the committee stated.
"Institutional bias or implicit biases are likely to play at least some part."
Extremely premature Māori and Pacific babies are also less likely to survive their first four weeks, even after accounting for factors like mother's age, weight, smoking and socioeconomic status.
That "suggests there were other factors increasing risk for these women," the committee stated.
There's a similar search for answers in the US, where African-American infants are more than twice as likely to die as white babies.
Poverty can't fully explain the gap. African-American women with advanced education, like doctors and lawyers, were more likely to lose their babies than white high school dropouts.
The 1990s saw a hunt for genes linked to premature birth, but there's currently interest in a "weathering" hypothesis — that a lifetime of discrimination damages women's health and makes early birth more likely.
In New Zealand, a ground-breaking study has now found evidence of ethnic bias among final-year medical students at University of Auckland and University of Otago.
The research, reported here for the first time, examined both explicit (intentional) and implicit or unconscious bias (automatic and outside awareness).
Online testing included reaction to hypothetical mental health and heart disease patients. Surnames — Wiremu/Williams or Tipene/Stephens — indicated ethnicity, with all other vignette details identical.
Students were also probed on their expectations of the patient — how likely they were to understand medical advice or refuse treatment.
On average, there was implicit bias toward Pākehā, and students rated Māori less likely to take prescribed antidepressants.
Real-life pressure would likely draw out more bias.
"There is evidence to support the notion that stereotypes and biases are invoked more in situations of high cognitive load," the authors, co-led by University of Otago academics Dr Donna Cormack and Dr Ricci Harris, wrote.
Could doctors' ingrained, unconscious biases make them less likely to push for resuscitation if a baby is brown?
No, said Dr John Tait, chairman of the mortality review committee and chief medical officer at Wellington's Capital & Coast DHB.
"The bias doesn't come into the actual act of resuscitation. No baby wouldn't be resuscitated because of its ethnicity."
Rather, bias could strike earlier in the chain of events and explain why more Māori, Pacific and Indian women have babies whose condition, such as birth weight, makes resuscitation less viable.
"The fact they're in those circumstances is where the unconscious bias is," Tait told the Weekend Herald.
"Is enough effort going into Māori and Pasifika health to prevent them going into labour early? If people don't attend antenatal clinics, it may be they can't afford the bus ... should we be going to them?
"There's no deliberate bias. But there's that underlying bias for years that's led to poor Māori health outcomes, and poor Pasifika outcomes as well."
As well as the resuscitation skew, the committee found Māori, Pacific and Indian women had worse access to antenatal care and large hospitals, where specialists work.
They were also less likely to be given antenatal steroids, of the type which helped mature Tia-Jane's lungs.
Ultimately, parents of grey zone babies decide on resuscitation after talks with the neonatal pediatrician and obstetrician. Could doctors' biases affect those discussions?
"That is always a possibility," Tait said. "We are still looking through the lens of Pākehā whereas the lens of Māori may be quite different. And that may have an effect on the ability to communicate."
The committee has recommended compulsory cultural competency training for the entire maternity and neonatal workforce. This should explicitly "address awareness of, and strategies to reduce and minimise the impact of, implicit bias and racism".
Medical and midwifery councils already require such training, and police recruits are now put through "awareness training" to combat bias.
Police Commissioner Mike Bush has said unconscious bias could contribute to Māori over-representation in offender statistics.
In the US, Starbucks put 175,000 employees through "anti-bias training", after workers in a store wrongly called police on two ' customers.
Videos shown on iPads included historic civil rights marches, and workers were asked questions including when they first noticed their own racial identity.
Personal experience means Keri Thompson isn't surprised by the resuscitation divide.
"The horrible part is it should shock me," she told the Weekend Herald.
Thompson's son Anaru Thompson Adams, now 22 and studying law, was born at 24 weeks. Before delivery, doctors ran over what could happen, including resuscitation. There was talk of blindness and other disability.
"I kind of remember what things I was told, but I didn't know what I was agreeing to. Technically you just remember, 'Please, keep my baby alive.'"
Anaru spent his first six months in Waikato Hospital's neonatal intensive care unit (Nicu), battling a lung condition. His heart and breathing stopped on three different occasions, but doctors brought him back.
Thompson remains grateful for the medical care received, but said at times she and her whanau felt "like aliens" and "culturally inept".
In one example, she said a doctor bluntly told her Anaru was less likely to live because he was Māori and a boy.
"They had a lot of medical reasons ... that it was about our physiology — our breathing airways.
"We got the best medical care that was on offer at the time. I have no complaints. But the judgements that come with that care ... "
(Boys do worse than girls, but this shouldn't enter doctors' decision-making. The alleged claim about Māori physiology is incorrect.)
Thompson realised her experience wasn't isolated, and subsequent interviews with Māori about their time in Waikato's Nicu formed a 2009 thesis.
Problems included strict visiting policy not recognising whanau extends beyond the notion of a nuclear family. One interviewee recalled how staff wouldn't use her baby's Māori name.
Relatively small actions could either demoralise or lift up, Thompson said.
"You're away from your own family. You may not have a lot of money and come from out of town. And you add the cultural differences — missing out on karakia, all that sort of stuff that would naturally happen."
One way to bridge such divides is to have a workforce as diverse as those being treated.
Almost one in five Kiwis will be Māori by 2025, but to match that share of the workforce would require another 3000 nurses, 2510 doctors, 380 dentists and 320 midwives.
Auckland DHB has responded with a radical change.
All eligible Māori and Pacific job candidates are now automatically fast-tracked straight to interview.
If they're not hired, managers must give specific feedback to HR, so the unsuccessful candidate can be coached to improve their chances in future interviews.
A new assessment tool prompts interviewers to think about "reflecting our communities and prioritised health outcomes", along with traditional skills and experience.
The policy began in June, but has escaped wider attention.
Fiona Michel, chief HR officer at Auckland DHB, said more Māori and Pacific candidates were being interviewed and hired.
"No one is employed just because of their cultural background. Candidates need to meet the core criteria.
"We are always looking to recruit the best person overall for the job."
The change has similarities to the NFL's "Rooney Rule", which since 2003 has required teams in the US competition to interview at least one minority candidate for head coaching openings.
It's been mirrored in the US private sector including by online retail giant Amazon, despite derision as ineffective box-ticking by some.
Jo Baxter, Associate Dean (Māori) at University of Otago, said hiring strategies made sense, given other DHBs were eyeing the same limited pool of candidates.
"It may not be an affirmative action thing. This may actually be, 'Can we get in early and get these ones before someone else snaffles them?'"
Baxter, director of a unit aiming to grow the number of Māori health science graduates, said not having Māori and others "in the room" during decision-making hurt those groups.
For example, aged care services can ignore the fact Māori die much younger, meaning the needy miss out despite being near the end of life.
More research was needed to understand the resuscitation divide, Baxter said. However, upping cultural awareness was "low-hanging fruit" in health.
"Some people in my generation make comments which make you think, 'Gosh, I don't know if I would want to be their patient.'
"Will they make a judgment that this person deserves or doesn't deserve a particular treatment? You know it happens."
Survival for our tiniest babies can hinge on postcode as well as ethnicity.
Palliative care is normally given to babies born at 23 weeks at some hospitals, including Middlemore. Parents hold their child as long as they are alive, and beyond. Baptisms or other rituals can be arranged.
Wellington and Dunedin stand out because doctors try to revive the vast majority of babies born that early.
That helped save 36 babies over the past decade at those two hospitals. Twenty-nine died. By comparison, the 23-week survival rate is 14 per cent at Middlemore.
Wellington and Dunedin's results aren't tempered by higher rates of serious disability, and a sector working group including specialists and obstetricians is now considering aligning practice across the country.
In the meantime, word of the regional differences has seen at least one family move.
A friend of Aly McVeigh in Hamilton knew she could have a premature birth and moved to Wellington just in case. She gave birth at 23 weeks and, like Tia-Jane, her daughter was resuscitated and survived.
That's impossible for women who don't have warning. It also takes money, and 40 per cent of very early babies are born into the most deprived families.
The Weekend Herald asked DHBs for Nicu resuscitation and survival rates by ethnicity, information not detailed by the mortality review committee.
Responses show how Māori, Pacific and Indian families take the burden of extremely premature birth.
They accounted for 85 per cent of births at 23 weeks at Counties Manukau, for example, with fewer than one in five surviving.
The DHB doesn't resuscitate from 23 weeks to 23 weeks and six days, unless after counselling the family is adamant.
In Wellington, with its aggressive resuscitation policy, more than 40 per cent of Māori, Pacific and Indian babies born in that age range survived.
DHBs warned regional comparison was fraught because of small numbers not adjusted for factors like deprivation.
Dr John Tait agreed, but supports a national approach.
His committee has asked Nicus to "investigate and address" regional differences in survival rates.
"You would hope, in a country of this size, there wouldn't be much variation and they would all manage and look after people in a similar way," he said.
Tia-Jane started kohanga reo the week before the Weekend Herald visited her home in Wellington's Broadmeadows.
Peppa Pig kept her occupied during the interview. She wears glasses and has had minor hearing issues, but otherwise you wouldn't know about her extraordinary beginning to life.
Her best friend is another 23 weeker. Their parents call them twinnies. Survivors. Both Māori. British, Swedish and Kiwi doctors cared for them.
Staff "treated us so well", Aly said. Communication during such a stressful time was critical, given how easy it was to misconstrue what was said or done.
Age comes into that as much as ethnicity or culture, she said. Aly had her son at 19, and Tia-Jane 20 years later.
"My maturity level was a lot different ... I wasn't interested in a career, going out, I wasn't missing out on study. So my entire focus and time was on Tia-Jane.
"It's about support networks. I saw a few young ones in there who had a lot of family around, almost every day. And you can see the outcome for those families was different."
Whether their own daughter would live wasn't always clear, especially early in her 137 days in intensive care.
"It's about every minute. You want to be there, you want to have sung her songs," Aly said. "We would go down in the middle of the night and read her a story."
The big moment came when consultant neonatologist Dr Max Berry stopped Aly and Adrian in a hallway.
"She said, 'Well, I hope you have her room ready'. At the time you are in this mode of every day counts. When she said that a light went on.
"I had taken nothing out of the cupboard. I didn't even have a teddy bear on the bed. I refused to set up even her cot, because if she wasn't coming home I didn't want to have anything to break down. I must have gone away and cried for a good 20 minutes after that."
His ratings have probably improved since the CCO reforms, says the mayor.