The Death Predictor: A Helpful New Tool or an Ethical Morass?
Whenever Eric Karl Oermann has to tell a patient about a terrible prognosis, their first question is always: "how long do I have?" Oermann would like to offer a precise answer, to provide some certainty and help guide treatment. But although he's one of the country's foremost experts in medical artificial intelligence, Oermann is still dependent on a computer algorithm that's often wrong.
Doctors are notoriously terrible at guessing how long their patients will live.
Artificial intelligence, now often called deep learning or neural networks, has radically transformed language and image processing. It's allowed computers to play chess better than the world's grand masters and outwit the best Jeopardy players. But it still can't precisely tell a doctor how long a patient has left – or how to help that person live longer.
Someday, researchers predict, computers will be able to watch a video of a patient to determine their health status. Doctors will no longer have to spend hours inputting data into medical records. And computers will do a better job than specialists at identifying tiny tumors, impending crises, and, yes, figuring out how long the patient has to live. Oermann, a neurosurgeon at Mount Sinai, says all that technology will allow doctors to spend more time doing what they do best: talking with their patients. "I want to see more deep learning and computers in a clinical setting," he says, "so there can be more human interaction." But those days are still at least three to five years off, Oermann and other researchers say.
Doctors are notoriously terrible at guessing how long their patients will live, says Nigam Shah, an associate professor at Stanford University and assistant director of the school's Center for Biomedical Informatics Research. Doctors don't want to believe that their patient – whom they've come to like – will die. "Doctors over-estimate survival many-fold," Shah says. "How do you go into work, in say, oncology, and not be delusionally optimistic? You have to be."
But patients near the end of life will get better treatment – and even live longer – if they are overseen by hospice or palliative care, research shows. So, instead of relying on human bias to select those whose lives are nearing their end, Shah and his colleagues showed that they could use a deep learning algorithm based on medical records to flag incoming patients with a life expectancy of three months to a year. They use that data to indicate who might need palliative care. Then, the palliative care team can reach out to treating physicians proactively, instead of relying on their referrals or taking the time to read extensive medical charts.
But, although the system works well, Shah isn't yet sure if such indicators actually get the appropriate patients into palliative care. He's recently partnered with a palliative care doctor to run a gold-standard clinical trial to test whether patients who are flagged by this algorithm are indeed a better match for palliative care.
"What is effective from a health system perspective might not be effective from a treating physician's perspective and might not be effective from the patient's perspective," Shah notes. "I don't have a good way to guess everybody's reaction without actually studying it." Whether palliative care is appropriate, for instance, depends on more than just the patient's health status. "If the patient's not ready, the family's not ready and the doctor's not ready, then you're just banging your head against the wall," Shah says. "Given limited capacity, it's a waste of resources" to put that person in palliative care.
The algorithm isn't perfect, but "on balance, it leads to better decisions more often."
Alexander Smith and Sei Lee, both palliative care doctors, work together at the University of California, San Francisco, to develop predictions for patients who come to the hospital with a complicated prognosis or a history of decline. Their algorithm, they say, helps decide if this patient's problems – which might include diabetes, heart disease, a slow-growing cancer, and memory issues – make them eligible for hospice. The algorithm isn't perfect, they both agree, but "on balance, it leads to better decisions more often," Smith says.
Bethany Percha, an assistant professor at Mount Sinai, says that an algorithm may tell doctors that their patient is trending downward, but it doesn't do anything to change that trajectory. "Even if you can predict something, what can you do about it?" Algorithms may be able to offer treatment suggestions – but not what specific actions will alter a patient's future, says Percha, also the chief technology officer of Precise Health Enterprise, a product development group within Mount Sinai. And the algorithms remain challenging to develop. Electronic medical records may be great at her hospital, but if the patient dies at a different one, her system won't know. If she wants to be certain a patient has died, she has to merge social security records of death with her system's medical records – a time-consuming and cumbersome process.
An algorithm that learns from biased data will be biased, Shah says. Patients who are poor or African American historically have had worse health outcomes. If researchers train an algorithm on data that includes those biases, they get baked into the algorithms, which can then lead to a self-fulfilling prophesy. Smith and Lee say they've taken race out of their algorithms to avoid this bias.
Age is even trickier. There's no question that someone's risk of illness and death goes up with age. But an 85-year-old who breaks a hip running a marathon should probably be treated very differently than an 85-year-old who breaks a hip trying to get out of a chair in a dementia care unit. That's why the doctor can never be taken out of the equation, Shah says. Human judgment will always be required in medical care and an algorithm should never be followed blindly, he says.
Experts say that the flaws in artificial intelligence algorithms shouldn't prevent people from using them – carefully.
Researchers are also concerned that their algorithms will be used to ration care, or that insurance companies will use their data to justify a rate increase. If an algorithm predicts a patient is going to end up back in the hospital soon, "who's benefitting from knowing a patient is going to be readmitted? Probably the insurance company," Percha says.
Still, Percha and others say, the flaws in artificial intelligence algorithms shouldn't prevent people from using them – carefully. "These are new and exciting tools that have a lot of potential uses. We need to be conscious about how to use them going forward, but it doesn't mean we shouldn't go down this road," she says. "I think the potential benefits outweigh the risks, especially because we've barely scratched the surface of what big data can do right now."
Story by Big Think
Our gut microbiome plays a substantial role in our health and well-being. Most research, however, focuses on bacteria, rather than the viruses that hide within them. Now, research from the University of Copenhagen, newly published in Nature Microbiology, found that people who live past age 100 have a greater diversity of bacteria-infecting viruses in their intestines than younger people. Furthermore, they found that the viruses are linked to changes in bacterial metabolism that may support mucosal integrity and resistance to pathogens.
The microbiota and aging
In the early 1970s, scientists discovered that the composition of our gut microbiota changes as we age. Recent studies have found that the changes are remarkably predictable and follow a pattern: The microbiota undergoes rapid, dramatic changes as toddlers transition to solid foods; further changes become less dramatic during childhood as the microbiota strikes a balance between the host and the environment; and as that balance is achieved, the microbiota remains mostly stable during our adult years (ages 18-60). However, that stability is lost as we enter our elderly years, and the microbiome undergoes dramatic reorganization. This discovery led scientists to question what causes this change and what effect it has on health.
Centenarians have a distinct gut community enriched in microorganisms that synthesize potent antimicrobial molecules that can kill multidrug-resistant pathogens.
“We are always eager to find out why some people live extremely long lives. Previous research has shown that the intestinal bacteria of old Japanese citizens produce brand-new molecules that make them resistant to pathogenic — that is, disease-promoting — microorganisms. And if their intestines are better protected against infection, well, then that is probably one of the things that cause them to live longer than others,” said Joachim Johansen, a researcher at the University of Copenhagen.
In 2021, a team of Japanese scientists set out to characterize the effect of this change on older people’s health. They specifically wanted to determine if people who lived to be over 100 years old — that is, centenarians — underwent changes that provided them with unique benefits. They discovered centenarians have a distinct gut community enriched in microorganisms that synthesize potent antimicrobial molecules that can kill multidrug-resistant pathogens, including Clostridioides difficile and Enterococcus faecium. In other words, the late-life shift in microbiota reduces an older person’s susceptibility to common gut pathogens.
Viruses can change alter the genes of bacteria
Although the late-in-life microbiota change could be beneficial to health, it remained unclear what facilitated this shift. To solve this mystery, Johansen and his colleagues turned their attention to an often overlooked member of the microbiome: viruses. “Our intestines contain billions of viruses living inside bacteria, and they could not care less about human cells; instead, they infect the bacterial cells. And seeing as there are hundreds of different types of bacteria in our intestines, there are also lots of bacterial viruses,” said Simon Rasmussen, Johansen’s research advisor.
Centenarians had a more diverse virome, including previously undescribed viral genera.
For decades, scientists have explored the possibility of phage therapy — that is, using viruses that infect bacteria (called bacteriophages or simply phages) to kill pathogens. However, bacteriophages can also enhance the bacteria they infect. For example, they can provide genes that help their bacterial host attack other bacteria or provide new metabolic capabilities. Both of these can change which bacteria colonize the gut and, in turn, protect against certain disease states.
Intestinal viruses give bacteria new abilities
Johansen and his colleagues were interested in what types of viruses centenarians had in their gut and whether those viruses carried genes that altered metabolism. They compared fecal samples of healthy centenarians (100+ year-olds) with samples from younger patients (18-100 year-olds). They found that the centenarians had a more diverse virome, including previously undescribed viral genera.
They also revealed an enrichment of genes supporting key steps in the sulfate metabolic pathway. The authors speculate that this translates to increased levels of microbially derived sulfide, which may lead to health-promoting outcomes, such as supporting mucosal integrity and resistance to potential pathogens.
“We have learned that if a virus pays a bacterium a visit, it may actually strengthen the bacterium. The viruses we found in the healthy Japanese centenarians contained extra genes that could boost the bacteria,” said Johansen.
Simon Rasmussen added, “If you discover bacteria and viruses that have a positive effect on the human intestinal flora, the obvious next step is to find out whether only some or all of us have them. If we are able to get these bacteria and their viruses to move in with the people who do not have them, more people could benefit from them.”
This article originally appeared on Big Think, home of the brightest minds and biggest ideas of all time.
Sign up for Big Think’s newsletter
Embrace the mess: how to choose which scientists to trust
It’s no easy task these days for people to pick the scientists they should follow. According to a recent poll by NORC at the University of Chicago, only 39 percent of Americans have a "great deal" of confidence in the scientific community. The finding is similar to Pew research last year showing that 29 percent of Americans have this level of confidence in medical scientists.
Not helping: All the money in science. Just 20 percent of Pew’s survey respondents think scientists are transparent about conflicts of interest with industry. While this issue is common to many fields, the recent gold rush to foot the bill for research on therapies for healthy aging may be contributing to the overall sense of distrust. “There’s a feeling that at some point, the FDA may actually designate aging as a disease,” said Pam Maher, a neuroscientist who studies aging at Salk Institute. “That may be another impetus for a lot of these companies to start up.”
But partnering with companies is an important incentive for researchers across biomedical fields. Many scientists – with and without financial ties and incentives – are honest, transparent and doing important, inspiring work. I asked more than a dozen bioethicists and researchers in aging how to spot the scientists who are searching for the truth more than money, ego or fame.
Avoid Scientists Who Sound Overly Confident in messaging to the public. Some multi-talented scientists are adept at publishing in both top journals and media outlets. They’re great at dropping science without the confusing jargon, in ways the public can enjoy and learn from.
But do they talk in simple soundbites, painting scientific debates in pastels or black and white when colleagues use shades of gray? Maybe they crave your attention more than knowledge seeking. “When scientists speak in a very unnuanced way, that can be irresponsible,” said Josephine Johnston, a bioethicist at the Hastings Center.
Scientists should avoid exaggerations like “without a doubt” and even “we know” – unless they absolutely do. “I feel like there’s more and more hyperbole and attention seeking…[In aging research,] the loudest voices in the room are the fringe people,” said the biogenerontologist Matt Kaeberlein.
Separate Hype from Passion. Scientists should be, need to be passionate, Johnston explained. In the realm of aging, for example, Leonard Guarente, an MIT biologist and pioneer in the field of aging, told me about his belief that longer lifespans would make for a better world.
Instead of expecting scientists to be lab-dwelling robots, we should welcome their passion. It fuels scientific dedication and creativity. Fields like aging, AI and gene editing inspire the imaginations of the public and scientists alike. That’s not a bad thing.
But it does lay fertile ground for overstatements, such as claims by some that the first 1,000-year-old has already been born. If it sounds like sci-fi, it’s probably sci-fi.
Watch Out for Cult Behavior, some experts told me. Follow scientists who mix it up and engage in debates, said NYU bioethicist Arthur Caplan, not those who hang out only with researchers in the same ideological camp.
Look for whether they’re open to working with colleagues who don’t share their views. Through collaboration, they can resolve conflicting study results and data, said Danica Chen, a biologist at UC Berkeley. We should trust science as long as it doesn’t trust itself.
Messiness is Good. You want to find and follow scientists who’ve published research over the years that does not tell a clean story. “Our goal is to disprove our models,” Kaeberlein said. Scientific findings and views should zig and zag as their careers – and science – progress.
Follow scientists who write and talk publicly about new evidence that’s convinced them to reevaluate their own positions. Who embrace the inherent messiness of science – that’s the hallmark of an honest researcher.
The flipside is a very linear publishing history. Some scientists have a pet theory they’ve managed to support with more and more evidence over time, like a bricklayer gradually, flawlessly building the prettiest house in the neighborhood. Too pretty.
There’s a dark side to this charming simplicity: scientists sometimes try and succeed at engineering the very findings they’re hoping to get, said Charles Brenner, a biochemist at City of Hope National Medical Center.
These scientists “try to prove their model and ignore data that doesn’t fit their model because everybody likes a clean story,” Kaeberlein said. “People want to become famous,” said Samuel Klein, a biologist at Washington University. “So there’s always that bias to try to get positive results.”
Don’t Overvalue Credentials. Just because a scientist works at a top university doesn’t mean they’re completely trustworthy. “The institution means almost nothing,” Kaeberlein said.
Same goes for publishing in top journals, Kaeberlein added. “There’s an incentive structure that favors poor quality science and irreproducible results in high profile journals.”
Traditional proxies for credibility aren’t quite as reliable these days. Shortcuts don’t cut it anymore; you’ve got to scrutinize the actual research the scientist is producing. “You have to look at the literature and try to interpret it for yourself,” said Rafael de Cabo, a scientist at the National Institute on Aging, run by the U.S. National Institutes of Health. Or find journalists you trust to distill this information for you, Klein suggested.
Consider Company Ties. Companies can help scientists bring their research to the public more directly and efficiently than the slower grind of academia, where “the opportunities and challenges weren’t big enough for me,” said Kaeberlein, who left the University of Washington earlier this year.
"It’s generally not universities that can take technology through what we call the valley of death,” Brenner said. “There are rewards associated with taking risks.”
Many scientists are upfront about their financial conflicts of interest – sometimes out of necessity. “At a place like Duke, our conflicts of interest are very closely managed, said Matthew Hirschey, who researchers metabolism at Duke’s Molecular Physiology Institute. “We have to be incredibly explicit about our partnerships.”
But the willingness to disclose conflicts doesn’t necessarily mean the scientist is any less biased. Those conflicts can still affect their views and outcomes of their research, said Johnston, the Hastings bioethicist.
“The proof is in the pudding, and it’s got to be done by people who are not vested in making money off the results,” Klein said. Worth noting: even if scientists eschew companies, they’re almost always financially motivated to get grants for their research.
Bottom line: lots of scientists work for and with companies, and many are highly trustworthy leaders in their fields. But if a scientist is in thick with companies and checks some of the other boxes on this list, their views and research may be compromised.