Based on national studies, white patients who came to the emergency room with symptoms resembling Covid-19 were tested far more often than Black patients with identical symptoms.
While the story of American medicine is filled with scientific advancements, the issue of racism is woven into the country’s medical institutions; something made even clearer by the Covid-19 pandemic. From testing to treatment, patients of colour continue to receive a lower quality of care compared to white Americans.
Artificial intelligence can go a long way to help address these health disparities and break down the barriers to healthcare inequalities.
1. Unequal testing and treatment
Research has shown that two out of three clinicians harbor an “implicit bias” against African Americans and Latinos. Though the doctors may not even be aware that they carry these biases, they are still detrimental to minority patients. Certain data reveals that Black individuals die from Covid-19 at a rate that is two to three times higher than white patients. Based on national studies, white patients who came to the emergency room with symptoms resembling Covid-19 were tested far more often than Black patients with identical symptoms. Studies also show Black women are less likely to be offered breast reconstruction after mastectomy than white women and that Black patients are 40 percent less likely to receive pain medication after surgery than white patients.
Technology can help combat these discrepancies. Some experiments using AI have shown some success in supplementing the physician’s judgment when diagnosing a patient’s medical needs. By using data from each doctor’s electronic health record, AI algorithms can compare treatments administered to patients of different racial or ethnic backgrounds within the individual physician’s practice. That data can be used to notify doctors when they are providing unequal treatment to a patient of a different ethnicity or race.
2. Bias in medical research and data
Sometimes, healthcare algorithms themselves discriminate against Black patients. While the fault appears to be the algorithms, in reality, it is the doctors who fail to provide sufficient medical care to Black patients in the first place who are to be blamed.
This is because artificial intelligence algorithms can only be as accurate as the data they’re fed. If the human inputs are biased, the data will be biased as well. AI cannot detect if it’s being fed biased information. That’s why it is essential that humans acknowledge their faulty assumptions while adjusting for bias in research and data aggregation.
With universities and funding agencies increasing their focus on racial issues in healthcare, racial bias will be factored into every AI project in the future. Once this happens, researchers will have no choice but to confront bias in healthcare, resulting in more equitable conclusions and outcomes.
3. Institutional racism
Often, the problem is institutional (or systemic) racism, which is a much deeper problem. This invisible, yet omnipresent racism is woven into the fabric of American healthcare and as a result, it can’t be resolved by modifying the behavior of individual doctors. However, with the help of AI, the actions that contribute to institutional racism can be identified and rectified.
For instance, Black women in the United States remain three times more likely to die from childbirth than white women. Most of these deaths could be prevented by analyzing more data and AI is best suited for this task.
Over a year into the Covid-19 pandemic, Black people continue to die at a rate three times higher than white people. Science and technology offer us the opportunity to combat racism in medicine. Combining AI with a national commitment to root out biases in healthcare is a great step to putting the medical system on a path toward equitable and fair healthcare practices.