Why is it a problem that pulse oximeters don’t work as well on patients of color | MIT News

Pulse oximetry is a non-invasive test that measures a patient’s blood oxygen saturation level and has become an important tool for monitoring many patients, including those with Covid-19. But new research links faulty pulse oximeter readings to racial disparities in health outcomes, which could lead to higher rates of death and complications, such as organ dysfunction, in patients with darker skin.

It is well known that non-white intensive care unit (ICU) patients receive less accurate readings of their oxygen levels using pulse oximeters, the common devices that are clipped to patients’ fingers. Now, a paper co-authored by MIT scientists reveals that inaccurate pulse oximeter readings can cause critically ill patients of color to receive less supplemental oxygen during their ICU stay.

Paper,Evaluation of racial and ethnic differences in oxygen supplementation among patients in the Intensive Care Unit,” published in JAMA Internal Medicinefocused on the question of whether there were differences in supplemental oxygen administration between patients of different races and ethnicities that were associated with discrepancies in pulse oximeter performance.

The findings showed that inaccurate readings in Asian, black and Hispanic patients resulted in them receiving less supplemental oxygen than white patients. These results provide insight into how health technologies such as pulse oximetry contribute to racial and ethnic disparities in care, according to the researchers.

The lead author of the study, Leo Antonio Celidirector of clinical research and principal research scientist of the MIT Computational Physiology Laboratoryand a principal research scientist at MIT Institute of Engineering and Medical Sciences (IMES), says the challenge is that health care technology is routinely designed around the majority population.

“Medical devices are typically developed in rich countries with fit, white people as test subjects,” he explains. “Drugs are evaluated through clinical trials that disproportionately enroll white people. The genomic data overwhelmingly comes from people of European descent.”

“So it’s not surprising that we see disparities in outcomes across demographics, with worse outcomes among those who were not included in the care design,” adds Celi.

Although pulse oximeters are widely used due to their ease of use, the most accurate way to measure blood oxygen saturation (SaOtwo) is by taking a sample of arterial blood from the patient. False normal pulse oximetry readings (SpOtwo) can lead to hidden hypoxemia. Elevated bilirubin in the bloodstream and the use of certain ICU medications called vasopressors can also alter pulse oximetry readings.

More than 3,000 participants were included in the study, of whom 2,667 were white, 207 black, 112 Hispanic, and 83 Asian, using data from the Medical Information Center for Intensive Care version 4, or MIMIC-IV data set. This data set is made up of more than 50,000 patients admitted to the Beth Israel Deaconess Medical Center ICU and includes pulse oximeter readings and oxygen saturation levels detected in blood samples. MIMIC-IV also includes supplemental oxygen administration rates.

When the researchers compared SpOtwo levels taken by pulse oximeter to oxygen saturation of blood samples, found that black, Hispanic and Asian patients had higher SpO2two readings that white patients for a given level of blood oxygen saturation measured in blood samples. Turnaround time for arterial blood gas analysis can take anywhere from several minutes to an hour. As a result, clinicians often make decisions based on pulse oximetry readings, unaware of their suboptimal performance in certain patient demographics.

Eric Gottlieb, the study’s lead author, a nephrologist, MIT professor and fellow at Harvard Medical School at Brigham and Women’s Hospital, called for more research to better understand “how disparities in pulse oximeter performance lead to to worse outcomes Potential differences in ventilation management, fluid resuscitation, triage decisions, and other aspects of care need to be explored Then we need to redesign these devices and properly evaluate them to ensure they work equally well for all patients.”

Celi emphasizes that understanding the biases that exist in real-world data is crucial to better developing algorithms and artificial intelligence to help clinicians in decision making. “Before we invest more money in developing artificial intelligence for health care using electronic medical records, we need to identify all the drivers of disparities in outcomes, including those that arise from the use of suboptimally designed technology,” he argues. she. “Otherwise, we risk perpetuating and magnifying health inequalities with AI.”

Celi described the project and the research as a testament to the value of data sharing that is at the core of the MIMIC project. “No team has the experience and perspective to understand all the biases that exist in real-world data to prevent AI from perpetuating health inequalities,” she says. “The database we analyzed for this project has more than 30,000 accredited users consisting of teams that include data scientists, clinicians, and social scientists.”

The many researchers working together on this topic form a community that shares and QAs code and queries, promotes reproducibility of results, and collaborates on data curation, says Celi. “There is harm when health data is not shared,” she says. “Limiting access to data means limiting the perspectives with which data is analyzed and interpreted. We have seen numerous examples of misspecified models and faulty assumptions leading to models that ultimately harm patients.”

Leave a Reply

Your email address will not be published.