AI can predict premature deaths in people with inflammatory bowel disease, suggests study
Almost half of people who died with inflammatory bowel disease (IBD) died prematurely, according to a study published in CMAJ (Canadian Medical Association Journal) that used machine learning models to predict death.
Canada has some of the highest rates of IBD worldwide, which includes Crohn disease and ulcerative colitis. People with IBD have shorter life expectancy than people without such diseases, and they can develop other chronic health conditions related to their IBD. The study found that people with IBD are at risk for premature death (defined as death before age 75) when they develop other chronic health conditions earlier in life.
As machine learning models can predict premature death in the general population, researchers applied the technology to determine whether it could predict premature deaths among people in Ontario with IBD and other chronic conditions using health care data held at ICES.
“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,” writes Dr. Eric Benchimol, a pediatric gastroenterologist and senior scientist at The Hospital for Sick Children (SickKids), professor of pediatrics and epidemiology at the Temerty Faculty of Medicine, University of Toronto, 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,” he says.
Of the total 9278 deaths in people with IBD between 2010 and 2020, almost half (47%) were premature, with higher rates in males than in females (50% v. 44%). The most common chronic conditions at death were various types of arthritis (77%), hypertension (73%), mood disorders (69%), kidney failure (50%) and cancer (46%). The researchers found that including chronic conditions diagnosed before age 60 and the age of diagnosis improved the models’ predictions.
“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,” write the authors.
The study was co-led by medical student Gemma Postill of the Temerty Faculty of Medicine, and Dr. Laura Rosella, professor and Canada Research Chair in Population Health Analytics at the Dalla Lana School of Public Health.
The authors hope that their research will help pinpoint areas for more targeted follow up from a range of health care professionals, from dietitians to mental health professionals and specialists when required.
“These findings provide scientific support for providing multidisciplinary and integrated health care across the lifespan (particularly during young and middle adulthood),” the authors conclude.
Reference:
Machine learning prediction of premature death from multimorbidity among people with inflammatory bowel disease: a population-based retrospective cohort study, Canadian Medical Association Journal, DOI:10.1503/cmaj.241117.
