“Predictive algorithms that power these tools should be continually reviewed and refined, and supplemented by information such as socio-economic data, to help clinicians make the best-informed care decisions for each patient.”
Racial bias found in algorithm used to predict healthcare needs of millions of Americans
BY: Harriet Alexander
An algorithm used to predict the healthcare needs of patients in the United States has been found to discriminate significantly against black people.
The computer programme is sold by a leading health services company, Optum, is used to guide care decision-making for millions of people.
Researchers examined how the system worked, and found that it underestimated by half the number of black patients needing extra care.
Their study, published on Friday, found that the results were distorted because it looked at the health costs in order to judge health needs.
Black patients incurred about $1,800 (£1,400) less in medical costs per year than white patients with the same number of chronic conditions, but the algorithm failed to take into account that less money is spent on black patients who are equally ill as white patients.
The discrepancy is because of myriad complicating factors, including explicit racism, access problems, lack of insurance, mistrust of the medical system, cultural misunderstandings or unconscious biases from doctors.
Previous studies have found that black patients are less likely to receive pain treatment, potentially lifesaving lung cancer surgery or cholesterol-lowering drugs, compared with white patients.
The four researchers, based in Chicago, Boston and Berkeley, found that there was an easy tweak to the system to make it more accurate.
Instead of just predicting which patients would incur the highest costs and use the most health care in the future, they tweaked the algorithm to make predictions about their future health conditions.
Optum said they welcomed the research, which would also be useful to creators of other healthcare algorithms – many of which use similar systems. The team were only able to carry out their research thanks to Optum sharing their processes.
“Predictive algorithms that power these tools should be continually reviewed and refined, and supplemented by information such as socio-economic data, to help clinicians make the best-informed care decisions for each patient,” said Tyler Mason, a spokesman for Optum.
“As we advise our customers, these tools should never be viewed as a substitute for a doctor’s expertise and knowledge of their patients’ individual needs.”
Read the article: https://www.telegraph.co.uk/news/2019/10/25/racial-bias-found-algorithm-used-predict-healthcare-needs-millions/