A standard screening test given to newborns minutes after birth is a less accurate predictor of infant mortality for Black babies than other children, a new study shows, but the authors said the Apgar test should still be used.
The study, published Tuesday in PLOS Medicine, found that more Black infants were assigned low Apgar scores than white infants and that the scores were better at predicting death in the first year of life in white infants compared with Black babies.
An Apgar score, ranging from zero to 10, is assigned to infants one, five, and sometimes 10 minutes after birth in nearly every delivery room, based on heart rate, muscle tone, reflexes, breathing, and appearance. Low scores, generally below a four, are a sign of potential trouble and that a baby might need extra medical care. They correlate with neonatal and post-neonatal adverse outcomes.
“Apgar scores should definitely still continue to be used,” Emma Gillette, first author on the study, said. “Our findings demonstrated that they were still highly predictive of mortality across all participants all across all race groups.”
This work, based on a large dataset of all births in the United States over two years, supports findings from a previous study that found there is a significant difference between Black and white infants’ Apgar scores and another study that found lower Apgar scores were more predictive of neonatal mortality in Mexican American and white infants than Black infants. The new paper included a broader number of racial categories and was more representative of the U.S. population.
Gillette, a clinical research coordinator at the Arnhold Institute for Global Health at the Icahn School of Medicine at Mount Sinai, said the project emerged from her dissertation research for graduate school and a desire to investigate racial inequities in infant mortality.
The authors found that non-Hispanic Black infants had 1.7 times the odds of being assigned a low Apgar score compared with white infants, and non-Hispanic Asian and Hispanic infants had 23% lower odds of being assigned low scores. However, the association between low Apgar score and mortality was weakest among Black infants.
“These are significant but are minor differences,” said Mario Rüdiger, a neonatologist working at University Hospital in Dresden, Germany, who was not affiliated with the study. Since past research has demonstrated that caregivers are very likely to give different Apgar scores to the same baby, the score is already not very likely to provide a strong prediction of outcome, he added.
“I’m not sure this study has great clinical impact,” agreed Jochen Profit, a professor of pediatrics at Stanford who was not involved in the study. He explained that physicians don’t typically use Apgar scores to make mortality predictions for full-term infants and instead mainly use it as an indicator of the effectiveness of resuscitative efforts. When scores are extremely low, skin color is not the determining factor because none of the babies will be pink.
“I would agree with the authors that one potential but not sufficient reason [for the findings] might be systematically lower assigned Apgars based on skin color,” he wrote in an email. “It could also be that Black infants, on average, are born sicker as a result of all the structural inequities that weigh on Black women’s health.”
These structural inequities or lower quality of care — such as babies being delivered at places with suboptimal neonatal care — might lower the Apgar scores they are assigned, without necessarily resulting in increased mortality. This might explain the weaker overall link between Apgar score and mortality, Profit speculated.
A “race-adjusted” Apgar score, which would have to be defined carefully to not “adjust away lived inequities experienced during pregnancy, birth, or postpartum,” might be worth tracking, Profit added. “It might be a good indicator of progress on health equity.”
But Rüdiger said he doesn’t think the Apgar score is affected by race or ethnicity. “The Apgar score is used everywhere,” he said, explaining that the tool has been successfully used across continents, races, and ethnicities.
The authors used maternal race as a proxy for the infant’s race and note that future research should incorporate paternal race as well as include data about the socioeconomic status of families and location of the birth. Including other maternal characteristics and prenatal care would also illuminate how newborn mortality can be affected by ethnicity.