Statistical Bias

From the more than 700,000 deaths from COVID-19 in the US and the nearly 5 million worldwide, there emerge even more stories than match the statistics when one considers all of the patients' relations. While the numbers are staggering, when we humanize the stories, we are left with even greater devastation, of course. One of the stories among so many that seemed particularly salient and poignant to us was the death of Dr. Susan Moore.

The Lancet Digital Health, Volume 3, June 2021

This Comment describes how systematic biases in data linkage misestimate health needs for ethnic minorities and further entrench existing disadvantages.
This Article supprts SDGs 3 and 10 by assessing the performance of four severity scoring systems used for case-mix determination and benchmarking in intensive care units to identify possible ethnicity-based bias. The study found systemic differences in calibration across ethnicities.