Overview

This module is intended to serve as an introduction to algortihmic bias for clinicians. It presents basic terminology, essential concepts, and some initial recommendations for clinicians to begin to handle using AI systems in providing clinical care. It is not intended to turn clinicians into experts on algorithmic bias. The primary goal is educate clinicians enough to avoid the major harms of algorithmic bias for patients under their care.

Instructions

1. Click Pre-Module Survey to provide anonymous information that will help determine how well the module is performing. There will also be a follow-up survey at the end.
2. Click Module in the menu to view the Interactive Algorithmic Bias in Health Care module for clinicians.
3. You can view of a summary of the module recommendations any time by selecting "recommendations" from the menu.
4. Consider reading the article listed below under the "suggested reading" header.
5. There are 4 cases (1A, 1B, 2, and 3) for you to work through after viewing the module and completing the reading. The cases can be found in the "cases" link in the menu. View each of the 4 cases. Under each case is a link to questions associated with the case. Please answer at least 1 question for every case. To receive full credit for participation, provide thoughtful original answers or reply to another comment while referencing learning points and recommendations from the module and the reading.
6. Please provide anonymous feedback by taking the survey at the post-module survey link in the menu.

Learning Objectives

This module will help you to...

1. Recognize different forms of algorithmic bias in the use of artificial intelligence for health care and the origins of those biases.
2. Deconstruct algorithmic bias in the use of artificial intelligence for health care by examining example cases.
3. Empathize with the importance of recognizing and addressing algorithmic bias in health care by exploring the problems resulting from algorithmic bias with a disproportionate impact on already marginalized populations and proposing ways to deal with those problems.

Suggested Reading

• Meskó, B., Görög, M. A short guide for medical professionals in the era of artificial intelligence. npj Digit. Med. 3, 126 (2020).

Additional Related Reading

• Parikh RB, Teeple S, Navathe AS. Addressing Bias in Artificial Intelligence in Health Care. JAMA 2019;322(24):2377-78.
• Vokinger KN, Feuerriegel S, Kesselheim A. Mitigating bias in machine learning for medicine. Communications Medicine 2021;1(25):1-3.