Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 142 - Fairness and Equity in Clinical Risk Prediction: Healthcare Data for the Public Good
Type: Invited
Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 11:50 AM
Sponsor: Biometrics Section
Abstract #309171
Title: Can Individualized Risk Calculators Reduce Racial/Ethnic Disparities in Cancer Screening Guidelines?
Author(s): Hormuzd Katki* and Corey Young and Li Cheung and Rebecca Landy
Companies: US National Cancer Institute and Morehouse School of Medicine and US National Cancer Institute and US National Cancer Institute
Keywords: screening; prediction; risk assessment; disparities; medical decision-making

Risk calculators could mitigate, or exacerbate, racial/ethnic health disparities in screening guidelines. In lung cancer screening, current guidelines do not account for race/ethnicity, even though african-americans have higher risk, and hispanic/asian-americans have lower risk, after controlling for all lung cancer risk factors. We show that some prominent risk calculators drastically overestimate risk, and other prominent calculators poorly estimate risk for racial/ethnic groups. We show that use of these risk calculators could mitigate some, but also exacerbate other, health disparities. We propose criteria for rigorous evaluation of risk calculators to avoid these problems. In the second half of the talk, we propose an alternate approach that instead estimates the individualized gain in life-expectancy due to screening, which optimizes the life-years gained in a population. We show the superiority of selecting people for screening based on life-gained vs. risk. However, if life-expectancy is lower for racial/ethnic groups because of deprivation or historic discrimination, then benefit calculators are inherently unfair. We suggest some remedies for this issue.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program