New Set of Standards Aims to Ensure Effectiveness, Equity of AI

In collaboration with various health and research organizations, the National Academy of Medicine (NAM) established an Artificial Intelligence Code of Conduct (AICC) to support the effective and responsible use of this technology within the healthcare field.

The growth of AI and its capabilities has led to a need for further transparency surrounding its impact on healthcare, as along with this expansion comes greater responsibility and a larger room for error.

The NAM Leadership Consortium (LC) supervised the development of the AICC to provide a foundation for AI and its use in healthcare. With an expected duration of three years, the AICC provided guidelines to support interoperable governance standards for the development of AI tools.

Building on similar prior efforts, the AICC will also provide more definitive details surrounding stakeholder responsibilities, particularly related to privacy, accountability, and ethics.

“Involving these accomplished national leaders from across the US is essential for creation of a harmonized, broadly adopted AI Code of Conduct, as well as for development of the national architecture that promotes the equitable and responsible use of AI. This collaborative effort will help ensure that the application of health AI is based on the best science, and is consistent with ethical principles and societal values in pursuit of effectiveness, efficiency, and equity for all members of society,” said Michael McGinnis, MD, the Leonard D. Schaeffer Executive Officer of the NAM, in a press release.

Further, the development of the AICC will include input from significant stakeholders and the public. Through structured gatherings, the NAM intends to update the AICC to continuously broaden adoption while improving practices to grow the potential of AI.

Prior research has indicated that greater attention is being dedicated to ensuring AI implementation increases health equity rather than detracts from it.

Research from May noted the abilities of AI to determine the likelihood of appointment no-show rates. The population used for the study consisted of residents of various Cleveland-based communities, including minority groups.

Although technological resources can assist healthcare access, healthcare disparities can limit this access. Thus, researchers from MetroHealth and Case Western Reserve University used AI to uncover the chances of missed appointments.

This study mainly considered adult internal medicine patients with a likely no-show rate of 15 percent or higher. Researchers used data belonging to these patients to construct an AI model.

Following this, they began a trial involving patients receiving phone calls from a scheduler. In the event that patients could not make an appointment, MetroHealth provided options for telehealth or transportation.

After analyzing the data, they found that Black patients who received a phone call had 36 percent lower no-show rates compared to those who did not get a call.

“The fear was implementing a model that gave more opportunities to patients who were not in dire need of better access, which would widen disparity gaps,” said Yasir Tarabichi, MD, the study’s lead author and MetroHealth’s medical director of the virtual care enterprise and director of clinical research informatics, in a press release.

Yasir added that AI allowed for the creation of a fair model that did not widen gaps. 

“We want to provide an outreach mechanism that is fair and provides a level of equity," he said. "People at a higher risk for not showing up are the ones we are helping the most. Minority patients were more likely to pick up the phone when called, and we met them where they were.”