Michael P. Cary Jr. and coauthors conduct a scoping review of literature related to mitigating racial and ethnic bias in clinical algorithms.

They find that studies "tended to be either highly specific technical guidance or high-level, nontechnical surveys of strategies," with limited evidence of the multidisciplinary approaches needed to overcome algorithmic bias.

In a Commentary regarding clinical algorithms, Tina Hernandez-Boussard and coauthors call for "striking the right balance between the race-neutral approach, which avoids using race as a risk factor in clinical decision making, and the race-aware approach, which incorporates data on disparities in an effort to advance health equity."

Noting the need to address inequities across multiple dimensions (such as race and rurality), Denis Agniel and coauthors describe how assigning quantitative goals for each dimension of equity, or "equity weighting," can yield more optimal results than traditional quality improvement incentives.

Zachary Dyer and coauthors identify forty-two variables available at the census tract level that they combine into a Structural Racism Effect Index.

Deciles of the index are monotonically related to life expectancy, diabetes prevalence, and the percentage of residents identifying as people of color.