While deaths steadily declined over a decade in intensive care units at hospitals with few minority patients, in ICUs with large numbers of minority patients, there was less improvement, according to new research published online in American Journal of Respiratory and Critical Care Medicine.

The disparity was most pronounced among critically ill African American patients, researchers report.


Scientists analyzed nearly 1.1 million patients hospitalized at more than 200 hospitals that participated in the telehealth platform provided by Philips Healthcare from 2006-16.

In additional to studying mortality, the researchers looked at length of stay in the ICU and in the hospital. The data showed a similar pattern of improvement over the decade at non-minority serving hospitals and less improvement at minority serving hospitals.

“We wanted to know whether racial inequalities, previously described across a range of healthcare environments, extend into the highest level of care, namely the ICU,” said lead author John Danziger, MD, MPhil, an assistant professor of medicine at Harvard Medical School and a physician at Beth Israel Deaconess Medical Center in Boston.

For the purposes of this study, the authors defined minority serving hospitals two ways. The first definition defined such hospitals as having twice as many minority patients as expected based on the percentage of African American or Hispanic living in the region according to the U.S. Census. The second defined a minority serving hospital as having more than 25 percent of its ICU patients identify as African American or Hispanic. The different definitions yielded similar results.

The study found that over a decade:

  • A steady decline (about 2%) each year in ICU deaths at non-minority hospitals, but no decline after the first few years at minority hospitals.
  • Longer lengths of ICU and hospital stay at minority hospitals than non-minority hospitals.
  • Nearly a third of critically ill African American patients and half of critically ill Hispanic patients were treated at only 14 of the surveyed hospitals.
  • African Americans treated at non-minority hospitals experienced a 3% decline in mortality each year, compared to no decline in mortality when treated at minority hospitals.

To avoid biasing results, the researchers took into account a range of variables, including age, gender, admission diagnosis, severity of illness and co-existing health problems. They found that minority serving hospitals tended to care for younger, but sicker patients.

The authors said that their study could not determine whether the health disparities they observed “reflect caring for an increasing disadvantaged population” or differences in hospital resources. The researchers did find that patients at minority service hospitals had to wait longer to be admitted to the ICU from the emergency room than patients who were treated in non-minority serving hospitals, suggesting that differences in resources contributed to the findings of their study.

Still, said Dr. Danziger, “The observation that large numbers of critically ill minorities are cared for in poorer performing ICUs gives us an important target for focused research efforts and  additional resources  to help close the healthcare divide amongst different minorities in the United States.”