AI offers answers to patients about multiple myeloma
But can a chatbot provide information a patient can trust?
Written by |
Have you seen “The Pitt?” The medical drama on HBO Max is not for the faint of heart as it races from case to case. I only started watching this year, but I’m hooked.
One plot line in the recently finished second season dealt with artificial intelligence (AI), casting it as a potential tool to help doctors with their charting, thus freeing them to spend more time with patients.
As it played out on the show, it didn’t quite work that way. The fictional AI tool made some transcription mistakes, upsetting some of the doctors. But art doesn’t quite imitate life on this point.
AI as the doctor’s helper
A 2025 American Medical Association survey found 66% of doctors use a form of AI for some tasks. The survey focused on “health care augmented intelligence,” defined as “a conceptualization of artificial intelligence that focuses on AI’s assistive role, emphasizing that its design enhances human intelligence rather than replaces it.”
That sounds to me like someone to do the grunt work. More than half of the doctors surveyed thought AI was helpful in cutting down on their administrative workload.
Cancer researchers have used AI for years. A search for “artificial intelligence” on Rare Cancer News turned up articles dating back to 2017, the oldest of these about an AI platform used to detect a biomarker to tell doctors which multiple myeloma patients might benefit from stem cell transplant (SCT). More recently, Rare Cancer News reported on steps toward AI-based technologies for diagnosing myeloma.
AI as the patient’s helper
Earlier this month, Gallup published a survey about how patients use AI. They partnered with the nonprofit West Health to survey more than 5,500 adults in the U.S., and found that 25% have used AI tools or chatbots to get health information or advice. Most of the time they were preparing for a doctor’s appointment, and a large number did research after a visit.
The report also said 14% of those AI users skipped a provider visit after consulting AI. But only 4% of respondents said they strongly trusted the answers they got.
Even before I saw that survey, I was curious about chatbots and how useful they might be for myeloma patients. So I asked three different chatbots a question I often see in online support groups: “What’s the best diet for multiple myeloma?”
Two of the chatbots were myeloma-specific. Myelo was created by the International Myeloma Foundation, and Leif is the HealthTree Foundation’s chatbot. The third was ChatGPT, one of the most popular tools cited in the Gallup survey. It told me, “There isn’t a single ‘best’ diet that can cure or directly treat multiple myeloma, but nutrition plays a real role in supporting your body during treatment, maintaining strength, and managing side effects.”
In its 323-word response, ChatGPT also made recommendations for managing side effects through diet, warned against some supplements and “anti-cancer” diets advertised online, and offered to tailor a plan specific to my needs.
Myelo gave me a 380-word response that recommended plant foods as my dietary base, along with dairy and meat proteins, plus healthy fats. Myelo also recommended hydration and warned me that during treatment was not the best time to go vegan or vegetarian.
Leif was a bot of few words. “There isn’t one ‘best’ diet for multiple myeloma, as nutritional needs can vary,” it counseled. “However, a balanced diet rich in fruits, vegetables, lean proteins, and whole grains is generally recommended to support overall health during treatment.”
All three bots advised me to consult with my care team about my dietary choices.
Chatbot advice vs. human doctor guidance
The chatbots’ advice did not contradict what I heard from the two doctors who guided me through induction and SCT in 2019. My local hematologist recommended a plant-based diet, and to go easy on the proteins since I had some kidney damage. My transplant specialist said to eat in a way that wouldn’t add to my medical burden. I got instructions for a low-microbial diet immediately after my SCT, but in the years since then, moderation has been the key.
I asked the chatbots some other questions, too. Each time, the answers aligned with each other and the chatbots advised me to talk with my care team. I didn’t get any wrong answers, but the questions were general rather than complex, and not specific to my case.
Leif, Myelo, and ChatGPT were helpful and much more targeted than a search engine. I think they can be helpful. But I don’t think they can replace a human doctor with a human touch.
Note: Rare Cancer News is strictly a news and information website about the disease. It does not provide medical advice, diagnosis, or treatment. This content is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition. Never disregard professional medical advice or delay in seeking it because of something you have read on this website. The opinions expressed in this column are not those of Rare Cancer News or its parent company, Bionews, and are intended to spark discussion about issues pertaining to rare cancer.
Leave a comment
Fill in the required fields to post. Your email address will not be published.