AI is coming to health care. Here are 9 challenges you need to prepare for.

Artificial intelligence (AI) offers major new capabilities for health care—but also raises complex questions that require "well-informed and thoughtful leadership," two legal experts write for Hospitals & Health Networks (H&HN).

Your cheat sheet for understanding Artificial Intelligence (AI)

What is AI?

Writing in H&HN, Jennifer Geetter and Dale Van Demark, partners in the health law practice of McDermott Will & Emery, note, "Modern AI solutions are the 'natural' descendants of other technologies that use the latest computer science advances to solve problems, gain insights, and automate in ways that would be out of reach otherwise."

While AI systems are largely self-sufficient, and are "freed from human-dictated logic," Geetter and Van Demark note that they are not error-proof, noting, "[p]oor data quality or training can result in biased outcomes."

What AI can do for health care

Geetter and Van Demark list several existing areas where AI is used in health care, such as cybersecurity, diagnostic assistance, precision medicine, revenue cycle management, and patient engagement.

In addition, they note that AI tools are being used to observe indicators of health that providers haven't traditionally used—such as media posts and consumer shopping habits—and are allowing consumers to more directly engage with "the 'internet of things,' creating the first stepping stone to a diffuse but integrated health 'system' with significantly more patient/consumer control and engagement than currently exists."

With AI come challenges

Providers looking to integrate AI systems into their practices must look beyond the usual technology implementation questions, such as proper vetting, assessment, and financial analysis, Geetter and Van Demark write.

They outline nine "unique challenges requiring additional considerations and strategies" for integrating AI into health care and ways to address those issues, saying that providers should:

  • Identify how AI will disrupt or supplement the current workflow, which could require "extensive education and training, and institutional culture changes";

  • Educate and engage patients who may "push back" against AI systems being used in their care;

  • Address AI systems' "black box" issue—which makes it hard to determine why the system reached a particular conclusion—in "contractual relationships with vendors, professional liability insurance terms, and workflows";

  • Be aware of AI systems' potential for bias, which can be affected by the quality of the data the AI system operates on and how the system trains itself over time;  

  • Be aware of and monitor the use of consumer-generated data and "diagnosis" among patients, which Geetter and Van Demark write "will likely result in an evolution of the patient-practitioner relationship and produce different types of consumer demands of practitioners and systems";

  • Monitor regulatory changes as professional organizations, legal experts, and federal regulators consider the implications of advanced AI systems in health care "from a policy and regulatory perspective";  

  • Manage expectations by "clearly describing what the system can and cannot do" and what the provider hopes to achieve from AI use;

  • Put in place a comprehensive privacy and security system to address the fact that AI systems "may deduce information that would otherwise remain private" and "recognize that bad actors are also adopting technology—including AI technology—to achieve their … goals"; and

  • Realign their strategic vision to make better use of an AI system's full functionality.

Geetter and Van Demark write, "With the advent of digital health tools that present very real solutions to health issues and offer improvements to processes related to clinical research, health system operations, and patient engagement—as well as with the adoption of value-based payment systems—the integration of information technology solutions into the DNA of health systems is a necessary and critical step."

And when it comes to AI, they conclude, "The common elements with all of these challenges are preparation and leadership" (Geetter/Van Demark, Hospitals & Health Networks, 9/18).

Your cheat sheets for understanding Artificial Intelligence, EMR optimization, and more

Download our cheat sheets to keep track of the fast-changing technologies transforming health care:

  • Artificial Intelligence (AI)
  • Interoperability
  • EMR optimization
  • 3D Printing
  • Digital health systems
  • Get all the Cheat Sheets


    Next in the Daily Briefing

    Weekend reads: These sheep can recognize Baaa-rack O-baaa-ma

    Read now


    Join the discussion

    Please log in to comment.
    Close

    Forgot your password?


    Not an Advisory Board Member? Click here to register

    Close

    Members please Log In

    LOG IN

    Forgot your password?


    Not an Advisory Board Member? Click here to register