Machine learning identifies clusters of multimorbidity among decedents with inflammatory bowel disease
Bibliographic record
Abstract
Multimorbidity is the co-occurrence of two or more chronic conditions in one person. Providing quality, patient-centered care requires understanding multimorbidity. Our objective was to identify patterns of multimorbidity that occur prior to death among people with inflammatory bowel disease (IBD). Using a retrospective population-based matched cohort derived from linked health administrative data of individuals with and without IBD who died between 2010 and 2020 in Ontario, Canada, we compared multimorbidity accumulation and leveraged unsupervised machine learning to identify multimorbidity clusters. Here we show decedents with IBD have a greater prevalence of complex multimorbidity (42% vs 34% with 8+ conditions, standardized difference: 22%). Among those with IBD at death, IBD is commonly developed as their first condition. At death, people with IBD have high prevalences of osteo- and other arthritis (77%), hypertension (73%), mood disorders (69%), renal failure (50%) and cancer (46%). Among those with IBD, we identify 3 clusters: ( $$\alpha $$ ) mood disorder and/or osteo- and other arthritis; ( $$\beta $$ ) cancer and low multimorbidity; ( $$\gamma $$ ) cardiovascular comorbidities. The clusters that we identify are stable across numerous validation techniques, including re-derivation and sex-specific clustering. These findings can inform future research and potential multimorbidity care in populations with IBD. Consideration of these clusters also lends to the need for further research, guidelines, and care programs to manage distinct subgroups of comorbidities among those with IBD, highlighting avenues for greater personalized care. People with inflammatory bowel disease (IBD) have inflammation in their digestive system that can cause stomach pain and diarrhoea. They often also live with other long-term health conditions. This study aimed to understand how different health conditions occur together before death in people with IBD. We used health records to compare people with and without IBD who died between 2010 and 2020. We assessed the conditions that occur with IBD and looked for common patterns of conditions co-occurring using computational methods. We found that people with IBD commonly had other health conditions, specifically arthritis, high blood pressure and mood disorders. Three main patterns of health condition co-occurrence in IBD emerged. These findings highlight the importance of developing tailored care programs to better support people with IBD facing multiple health challenges. Postill et al. apply unsupervised machine learning methods to cluster longitudinal administrative health data of deceased people with IBD. Their model identifies three clusters of multimorbidity present at death among those with IBD: mood disorder and/or osteo- and other arthritis; cancer and low multimorbidity; and cardiovascular comorbidities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".