It has been a while since I wrote on a technology subject. I usually leave it to others who stay much more current and are doing the day-to-day IT work in healthcare organizations. But the rapid acceleration of
generative AI with ChatGPT the past several months has caught my attention.
I put my toe in the water and asked it to draft some marketing messages for me several weeks ago. For first drafts, they weren’t too bad. Using ChatGPT in this way gets you past the blank sheet of paper (or screen) and writer’s block. Or as my minister husband would say at times when writing a sermon, you need to get past picking out the font.
My StarBridge Advisors colleague, David Muntz, recently used ChatGPT to do an internal presentation on AI for our advisor team and gave ChatGPT “co-author” credit to make a point.
Bill Russell has talked about it often on This Week Health podcasts. He and his team are exploring ways they can use it in their business. Here are just a few of the podcasts he’s done on ChatGPT:
Generative AI, the new Medical Generalist
Several health system digital leaders told Becker’s that AI can only do so much in this article dated March 16, 2023, “How far is too far for AI in healthcare?”, and how it needs to be thought as of “augmented intelligence”. Ashish Atreja, MD, CIO and chief digital health officer of UC Davis Health said it well, “Machines lack internal consciousness, and the ability to distinguish right from wrong. Roles that depend on human empathy and critical decision-making in ambiguous situations can be supported but should never be replaced by AI.”
David Muntz’s blog post from October, 2022, “AI and AI – Use Them Responsibly”, talks about artificial intelligence and augmented intelligence. His view is similar to the digital leaders quoted in the Becker’s article. David closes his post by saying, “AI and AI should not be seen as ways to replace humans but as technologies that allow us to instantiate learnings from the human experience. AI applied responsibly and compassionately allows people to share evidence-based practices and focus on the more human-oriented tasks that cannot be automated. Importantly, AI helps us recover time to spend on our community, our providers, our patients, their families, and ourselves.”
The potential for medical AI is so significant that there will be a new AI-themed journal from the New England Journal of Medicine called NEJM AI. The inaugural editor-in-chief is Isaac Kohane, founding chair of the Department of Biomedical Informatics at Harvard Medical School.
Based on a survey of 40 leading health systems from the Health Management Academy, there is growing interest in AI to address staffing and labor issues according to this March 17, 2023, Becker’s article, “Almost half of health systems use AI to tackle workforce issues”.
Concerns about regulating the use of this technology have been raised in recent months. A March 3, 2023, New York Times article by Cecilia Kang and Adam Satariano, “As A.I. Booms, Lawmakers Struggle to Understand the Technology” highlights one of the challenges. A spot-on quote from one representative paints the picture. “The problem is that most lawmakers do not even know what AI is”, according to Representative Jay Obernolte, a California Republican and the only member of Congress with a master’s degree in artificial intelligence.
In a March 9, 2023, Boston Globe article, “Here are nine key issues in the AI debate”, Globe Correspondent Scott Kirsner summarized key questions based on hearing dozens of speakers on the topic at a TEDxBoston event. Here are several of them:
-
- What will inclusive, unbiased AI look like?
- Who does the fact-checking and verification?
- How do we keep AI from being toxic and hurtful?
- Who owns the data – and how intrusive will AI be?
- How do we prepare for the inevitable job loss?
AI appears to have been a big topic at HIMSS23 this week. There is a free resource available from HIMSS for those interested in learning more – HIMSS Resource: AI in Healthcare. As I didn’t attend the HIMSS conference this year, I look forward to reading the recaps from it and hearing about the latest vendor announcements on integrating ChatGPT into their products.
An initial recap from John Lynn, “Biggest themes from HIMSS23” includes these comments: “Of course, while those two topics (workforce burnout and revenue) are top of mind for healthcare organizations at HIMSS 2023, the real talk of the conference is around ChatGPT and Large Language Models. The excitement for these technologies is palpable. However, it also comes with a touch of fear. Many can see the potential and health IT vendors are all evaluating how to incorporate it into their solutions. It’s still too early to really know the answer for how things like ChatGPT are going to impact health IT, but there are a lot of people thinking about it and seeing how they can leverage the potential that comes with it.”
If you’ve read this far and want some basics on ChatGPT and its potential in healthcare, there’s more yet. As I thought about trying to summarize what I’ve read on the potential for healthcare, I decided to ask ChatGPT for help. My query was: “Draft a blog for healthcare IT leaders about the potential of ChatGPT”. Here’s the full response I got:
As healthcare continues to evolve, the industry is being transformed by the power of artificial intelligence and machine learning. One technology that is gaining traction in the industry is ChatGPT, a large language model that can understand and respond to human language in a natural way. ChatGPT is being hailed as a potential game-changer for healthcare IT leaders, and in this blog, we will explore the potential of this technology in the healthcare industry.
What is ChatGPT?
ChatGPT is a natural language processing (NLP) model developed by OpenAI. It uses deep learning techniques to understand and respond to natural language queries, making it possible for users to interact with the system as if they were talking to another human being. The model has been trained on vast amounts of text data, making it capable of understanding a wide range of topics and contexts.
How can ChatGPT be used in healthcare?
ChatGPT has many potential applications in the healthcare industry, including:
Patient communication: ChatGPT can be used to create chatbots that patients can interact with, allowing them to ask questions about their health, schedule appointments, and receive personalized advice.
Medical education: ChatGPT can be used to develop educational resources for healthcare professionals, such as virtual assistants that can answer medical questions or provide guidance on treatments.
Electronic health records (EHRs): ChatGPT can be used to improve the usability of EHR systems by allowing users to interact with them using natural language queries, rather than requiring them to navigate complex menus and interfaces.
Clinical decision support: ChatGPT can be used to provide clinicians with real-time decision support, helping them to make more informed treatment decisions based on the patient’s individual needs and medical history.
Health insurance: ChatGPT can be used by health insurance companies to assist customers in understanding their policies, coverage, and benefits.
Benefits of using ChatGPT in healthcare
Using ChatGPT in healthcare can provide a number of benefits, including:
Improved patient engagement: ChatGPT can help to improve patient engagement by providing patients with a more personalized and interactive experience.
Increased efficiency: ChatGPT can help to increase efficiency by automating tasks that would otherwise require human intervention, such as appointment scheduling or answering common questions.
Enhanced decision-making: ChatGPT can help to enhance decision-making by providing clinicians with real-time, data-driven insights that can inform treatment decisions.
Improved access to healthcare: ChatGPT can help to improve access to healthcare by providing patients with 24/7 access to healthcare resources and support.
Conclusion
ChatGPT has the potential to revolutionize the healthcare industry by providing patients and healthcare professionals with a more intuitive and interactive experience. Healthcare IT leaders should consider how they can leverage this technology to improve patient outcomes, increase efficiency, and enhance decision-making. As with any technology, there are challenges and risks associated with implementing ChatGPT in healthcare, but with careful planning and execution, the potential benefits of this technology are significant.
Don’t worry, I’m not ready to subcontract my weekly blog to ChatGPT but it is certainly interesting what it can produce with a simple query!
I’m in the Bill Russell camp on this – CIOs should be working with their team to figure out where and how to leverage generative AI within their organization.
And if you’re looking for ways to use it yourself, check out this New York Times article by Francesca Paris and Larry Buchanan from yesterday – “35 Ways Real People Are Using AI Right Now”. Fascinating!