A new AI model can predict the likelihood of premature death from inflammatory bowel disease (IBD) with around 95% accuracy. This is good news, though – because, by looking beyond the gut for early treatment of other chronic illnesses, the early mortality rate can be significantly reduced.
A new study by Canadian researchers from the University of Toronto, the Institute of Clinical Evaluative Sciences (ICES), and the SickKids Research Institute, along with McGill University and Trillium Health Partners, has found that AI intervention can provide crucial intervention that addresses multimorbidities – two or more other health conditions – in patients with IBD, allowing for early treatment of those other issues that may not get addressed until they've worsened.
It's still unclear whether IBD triggers other conditions such as arthritis and mental health issues, but scientific evidence has recently shown there is a link. Those conditions include systemic inflammation, which can lead to arthritis, as well as cardiovascular and kidney disease. IBD is also linked to immune system dysregulation, mental health issues such as mood disorders like depression, and hypertension and diabetes triggered by IBD medication.
“The clinical implication is that chronic conditions developed early in life may be more important in determining a patient’s health trajectory, although further causal research is needed to elucidate this relationship,” wrote Dr. Eric Benchimol, a pediatric gastroenterologist and senior scientist at SickKids, and a senior core scientist at ICES. “Although our insights are not causal insights, they identify patients potentially at higher risk of premature death, and therefore who might benefit from more coordinated care of their IBD and other chronic conditions."
The team tested several AI models on a dataset of 9,278 deceased individuals with IBD, in which 47.2% had died prematurely – before the age of 75. Men represented 50% of these premature deaths. A majority of the cohort had been diagnosed with IBD before turning 60, as most individuals generally seek medical attention in adolescence or early adulthood.
The findings suggest that if those multimorbidities are treated early on, as well as IBD, there's a critical period of time to also address those other serious issues, rather than only in later life or when they've worsened.
IBD, different to irritable bowel syndrome, is an umbrella term that includes Crohn's disease and ulcerative colitis, which is marked by chronic inflammation of the gastrointestinal tract. It affects around one in 77 Americans, and an estimated 4.9 million people worldwide. Notably, Western countries are overrepresented in the statistics.
As machine learning advances, models can be tweaked to focus on specific premature death scenarios – in this case, for the population living with IBD, who are at much higher risk of one or more of 17 other serious conditions. While the scientific evidence varies, people with IBD – across both sexes – are most likely to die between two and eight years sooner than the overall non-IBD population.
In the study's cohort, the common multimorbidities at the time of death were arthritis (77%), hypertension (73%), mood disorders (69%), kidney failure (50%) and cancer (46%). But if chronic illnesses were identified in middle age or earlier, the AI model predicted a lower incidence of premature death.
So what does this all mean? Essentially, while IBD and its associated chronic conditions present complex problems for individuals and healthcare professionals, identifying and intervening in those multimorbidities at the onset of non-IBD illnesses can have a significant impact on life expectancy.
“The use of premature death as the outcome more directly identifies opportunities for health system improvements, as premature deaths are considered avoidable through appropriate prevention or early and effective treatment,” the researchers wrote.
The study provides new avenues of research and treatment that addresses a patient's health beyond IBD.
“These findings provide scientific support for providing multidisciplinary and integrated health care across the lifespan (particularly during young and middle adulthood),” the scientists noted. "Our work points to a need to expand availability to other medical specialties and to enhance care coordination among specialties.
“Machine learning prediction models are not causal tools, but they can highlight high-risk patterns that deserve clinical and research attention," they added.
The study was published in the Canadian Medical Association Journal.
Source: Canadian Medical Association via Scimex
