To provide some data helpful in making determinations on aggressive care or palliative care. The author is a Dean at the Stanford University School of Medicine. A good short article.
The first patient I saw during my clinical rotations as a medical student was dying from chronic myeloid leukemia (CML). Today, with treatment, the 5-year survival rate for CML is about 90%. But when I was a student, CML was incurable. It was an early and poignant reminder of the limitations of medicine that I have never forgotten.
We didn’t understand cancer like we do now, nor did we have the tools to treat it, which meant there was agonizingly little we could do. With the national hospice and palliative care movements only beginning, our discussions with this man and his wife focused mostly on which chemotherapies might be able to prolong his life.
The attending physician and I answered questions about the severity of the side effects of these treatments, but we could not answer the one question he most desperately wanted to know: How long have I got left?
For all the advancements we’ve made in biomedicine since I was in medical school, answering this kind of question with anything approaching certainty still vexes doctors. It’s a question that informs many others: How long should a dying patient “fight” a terminal illness, and when should that person focus instead on minimizing suffering? Is a day at home more valuable than a week in the hospital? Even a small improvement in our ability to gauge the life expectancy of a seriously ill patient could provide enormous value for them and their families.
A new algorithm developed at Stanford Medicine could help. Analyzing data from hundreds of thousands of anonymized medical records, the model predicts which patients are likely to die in the next 3 to 12 months. In early tests, the algorithm analyzed medical data of patients who had already passed away and correctly predicted their remaining life expectancy in 9 out of 10 cases.
The idea of using algorithms in end-of-life care understandably makes people uneasy, so it’s important to be clear: AI isn’t going to make decisions for patients or for doctors, and it’s not going to deny nor discourage care. If a patient wants to proceed with an aggressive treatment regimen, that choice will of course be honored. What AI can do is give patients information they have never had before that can help them realize their preferences as they near the end of their lives, whether that’s remaining mentally aware, being able to spend time with family, avoiding severe pain, or exhausting every possible avenue to defeat their disease.
Entering palliative care too late can mean more time in the hospital pursuing aggressive treatments that offer little chance of improvement, and precious time lost for families. It can also create a financial burden, as patients continue to undergo expensive, curative treatment. Between 2000 and 2014, the average Medicare spending on a beneficiary who died at some point during the year more than doubled. On the other hand, those who enter palliative care too early risk missing out on treatments that could improve their condition and extend their life.
It’s a delicate balance, and according to data from the National Palliative Care Registry, one that we can improve. Less than 50% of hospital patients who need palliative care actually receive it. Just like that first CML patient I cared for, they may suffer because no one knows when to steer them toward a gentler ending.
I don’t pretend AI is a panacea. An algorithm isn’t going to make decisions for doctors or patients, but it can help inform their choices by providing them with insight they’ve never had before.
I’ve seen incredible success stories over the course of my career, many of them driven by advances in biomedicine and technology. Today, clinicians who treat patients with CML are able to have very different discussions than when I was in medical school. As we work to transform more terminal illnesses into treatable ones, AI can help inform the difficult conversations about diseases that remain incurable. It may seem more like silver lining than technological breakthrough, but in the face of illnesses that take away absolutely everything, it’s incredibly powerful to be able to help those who are terminally ill understand how much time they have left. It empowers them to spend that time doing what brings them joy, and to be with those whom they love.
Thank you for this one Tom. I agree wholeheartedly.
Posted by: ellen mullin | Monday, May 07, 2018 at 11:10 AM