The Potential For Bias In Machine Learning And Opportunities For Health Insurers To Address It

By Kaelyn Varner

In the health insurance industry, computers can be trained to study health data and use it to predict events like the chances of a patient developing a disease or being hospitalized. This is a useful tool that helps the healthcare industry better distribute resources, design outreach programs, and reduce health care costs. These predictions, however, can make unfair and incorrect assumptions about some patients because the information used to make the predictions isn’t diverse enough or collected consistently. In order to make these predictions more accurate for all groups of people, especially marginalized people of color, health insurance companies must collect more data about other social and environmental factors that affect health outcomes like race, ethnicity, gender, education level, income, and housing quality, to name a few. The insurance industry must also create guidelines for the best ways to collect this health information and use it responsibly to ensure equitable, accessible health care for all patients.

  1. Gervasi, S. S., Chen, I. Y., Smith-McLallen, A., Sontag, D., Obermeyer, Z., Vennera, M., & Chawla, R. (2022). The potential for bias in machine learning and opportunities for health insurers to address it. Health Affairs, 41(2), 212–218. https://doi.org/10.1377/hlthaff.2021.01287