Predicting Premature Death with Multimorbidity Save

Machine learning (ML) models were used to predict premature death in inflammatory bowel disease (IBD) patients with multimorbidity, the co-occurrence of 2 or more chronic conditions. 

Canadian administrative health data on IBD patients (n = 1856) who died between 2010 and 2020 were utilized in this population-based, retrospective cohort study.  The primary outcome was premature death (before age < 75 yr) among people with IBD and to identify the role of multimorbidity, specifically the presence of 17 chronic conditions. 

While all ML models showed strong performance (AUC 0.81–0.95), the best performing model had an AUC of 0.95 (95% confidence interval 0.94–0.96) and incorporated age at diagnosis for each chronic condition developed at or before age 60 years Salient features for predicting premature death were younger ages at the time of diagnosis for mood disorder, osteo-and other arthritis types, other mental health disorders, and hypertension, as well as male sex.

The development of multimorbidity in IBD patients, especially if developed early in life (age ≤ 60 yr), significantly affects health trajectory and the risk of premature death.