Credit Reuters Photograph: Toby Melville
The Guardian
By Andrew Gregory
Tool addresses inequalities in UK, where stillbirth and perinatal death rates are high for black and Asian babies
The NHS is piloting a new artificial intelligence pregnancy screening tool that could dramatically reduce racial disparities in baby death rates.
Stillbirth and perinatal death rates are comparatively high for black and Asian babies in the UK, studies have shown. Now the Tommy’s National Centre for Maternity Improvement, led by the Royal College of Obstetricians and Gynaecologists and the Royal College of Midwives, has successfully designed and trialled a tool that could help end the scandal.
The development comes less than a month after the Guardian revealed how a damning study – the largest of its kind – found “vast” and “widespread” inequity in every aspect of healthcare that it reviewed, and warned that it was harming the health of minority ethnic patients in the NHS.
The story sparked outrage among patients, campaigners and MPs, and prompted black, Asian and minority ethnic doctors to speak out. They said health officials had “paid lip service” to racism in the NHS for years and called for “concrete” action to tackle the racial disparities in healthcare.
The NHS will now begin piloting the pregnancy screening tool, aimed at tackling some of the inequalities exposed by the Guardian, in four areas across England.
It follows trials involving more than 20,000 pregnant women. Researchers found huge racial disparities in pregnancy outcomes when standard screening was used. Perinatal death rates were three times higher among ethnic minority mothers (7.95 deaths per 1,000 births) than among white mothers (2.63 per 1,000 births).
But when the new tool was used in conjunction with targeted care, death rates fell to about the same across all ethnic groups.
Writing in BJOG, an obstetrics and gynaecology journal, the researchers concluded: “First trimester combined screening for placental dysfunction is associated with a significant reduction in perinatal deaths in black, Asian and minority ethnic women – to perinatal mortality rates equivalent to those seen in white women.”
Prof Basky Thilaganathan, clinical director of the Tommy’s National Centre for Maternity Improvement, led the research team at St George’s University hospital NHS trust in London. “It’s incredibly exciting to see that changing from the standard pregnancy risk factor checklists to our new approach can directly address and almost eliminate a large source of the healthcare inequality facing black, Asian and minority ethnic pregnant women,” he said.
“The current maternal risk-factor screening programme is limited and can contribute to ongoing racial inequalities – but our algorithm can account for these deficiencies, enabling us to truly personalise care rather than treating large groups in the same way, and ultimately improve pregnancy outcomes.”
As well as London, the tool is now being piloted in Yorkshire, Lancashire and Surrey. “Given this new evidence, we would urge the government to support and invest in this new system, which will allow us to roll this out more widely across the UK,” added Thilaganathan.
The current method for risk assessment remains largely the same as it was in the 1970s. The new tool uses AI to interpret detailed data routinely gathered during a woman’s antenatal appointments to assess their specific, personalised risk of potential complications developing during pregnancy.
The new tool reduced baby deaths among mothers from ethnic minority backgrounds by 60%. Pregnant women found to be at high risk by the algorithm had extra monitoring during pregnancy, so care teams could act early to prevent problems and save babies’ lives.