Hospitals are one of the last places we might expect to find evidence of possible racial bias, even unconscious bias, in their services. For one thing, all bodies are alike to physicians. For another, health practitioners work under ethics of care and compassion that do not discriminate on any grounds.
So it is surprising and disturbing that a report to Health Minister David Clark by the Perinatal and Maternal Mortality Review Committee has found an apparent racial bias in decisions to resuscitate premature babies.
It reports that resuscitation is attempted on 95 per cent of Pakeha and most Asian babies born between 23 and 26 weeks but only 92 per cent of Māori babies, 89 per cent of Pacific and 86 per cent of Indian babies. This is after accounting for difference in mothers' age, weight, smoking and socio-economic status.
"Institutional bias or implicit biases are likely to play at least some part [in the resuscitation decisions]," the committee concludes.
The 23-26 week phase of gestation is when these decisions are the most difficult. Before 23 weeks, resuscitation is not attempted. After 26 weeks it is almost always attempted. Births that occur in between present an agonising dilemma for the parents and their doctors. Many considerations must come into the decision, including the low rate of survival at that stage and risk of severe disabilities for those that do survive.
But none of the legitimate considerations include the colour of the parent's skin, or any assumed ethnic characteristics.
Yet in one case we report today a woman says a doctor told her when her baby was arriving at 24 weeks he had less chance of survival because he was Maori. The boy survived and is now a law student.
The woman has interviewed mothers in a similar situation and suggests racial bias can be more subtle than she experienced, and institutional, but no less unfair to those poorly served, She cites visiting policies that do not recognise whanau beyond the immediate family.
New Zealand is not alone in trying to combat racism in neonatal services. In the United States African-American premature births are more than twice as likely to die as white babies. Even well-off African-American women in professional occupations were more likely to lose their babies than white women.
In New Zealand, a survey of final-year students at Auckland and Otago medical schools has found both explicit and unconscious bias in them. "Patients" with Maori names were presumed less likely to understand medical advice or take a prescribed treatment.
Ethnicity is much studied in health and social science these days. Maori and Pacific groups register disproportionately on most measures of wealth, education, housing and other characteristics that can influence their health. But risk factors should not influence treatment decisions, especially at birth.
The chairman of the mortality review committee, Dr John Tait, does not believe racial bias determines resuscitation decisions directly. He believes fewer resuscitation attempts are made for Maori, Pacific and Indian babies because their condition, such as weight, makes resuscitation less viable. Nevertheless his committee has recommended compulsory cultural competency training for the entire maternity and neonatal workforce to "address awareness of, and strategies to reduce and minimise the impact of, implicit bias and racism".
Clearly medical leaders have some work to do.