It’s hard to wrap your brain around. It’s hard to even type these words, but here they are: it’s useful for physicians to be able to predict when a patient will die.
Don’t misunderstand. It’s not about playing God or acting nefarious. It’s about doctors being able to help patients die on their own terms, as comfortably as possible, having received the best, most appropriate care.
In the new issue of Stanford Medicine magazine, writer Kris Newby describes a Stanford pilot program that marries technology and compassion, artificial intelligence and palliative care, with the aim of helping doctors better understand which patients could benefit from end-of-life conversations while there is still time.
In the article, palliative care physician Stephanie Harman, MD, tells Newby:
Ideally with this AI model, we’re identifying patients who are sicker than we realize… And it gives us an excuse to say, ‘It’d be great if we could talk about advanced care planning.’ Or, ‘Have you had a discussion with your regular doctor about what matters most to you if and when you get sicker?’ I think the twist is that we’re using machine learning to add more to a patient’s care without taking anything away.
Developed by Nigam Shah, MBBS, PhD, the model uses an algorithm to calculate the probability that a patient will die within the next 12 months, based on comparisons of the past year of the patient's medical history with records of millions of other patients. Many factors are considered, including the number of hospital admissions, disease classification codes and prescription codes.
The tool provides Harman with a daily report of newly-admitted hospital patients who have a 90 percent or higher probability of dying in three to 12 months. Harman reviews the medical records to decide if the patients have palliative care needs. “She’s found the list to be helpful," the article explains, "and she sees how it can improve hospital care and enable her to spend more time with the most critical patients.”
This is important because, as the article notes, less than half of admitted hospital patients who need palliative care actually receive it. And many more people would want to die at home than actually do.
Hopefully, tools like this will help.
Photo by Edward Caldwell