Performance of Novel Inflammatory Bowel Disease-Specific Models for Risk Adjustment

We aimed to develop and evaluate inflammatory bowel disease (IBD)-specific risk-adjustment models using an actuarial approach.

This retrospective, cross-sectional study was conducted using Optum administrative claims data from patients with Crohn's disease and/or ulcerative colitis between January 1, 2020, and December 31, 2022. Clinical-risk groupers with three severity levels were developed for Crohn's disease and ulcerative colitis based on IBD medical costs associated with diagnosis codes. Linear regression models assessed the relationship between clinical-risk groupers and concurrent or prospective costs. Separate models were developed for commercially and Medicare Advantage-insured patients, and for users and nonusers of targeted IBD therapies. Additional predictors that improved explanatory power (R²) were included.

In commercially and Medicare Advantage-insured patients, the clinical-risk grouper explained 20.99% and 20.72% of the variance in concurrent costs incurred by targeted therapy users, and 2.07% and 2.01% of the variance for nonusers, respectively. Including IBD-related hospitalizations (number and duration of stay) increased R² to 41.56% and 22.71% for targeted IBD therapy users and to 29.20% and 16.56% for nonusers who were commercially and Medicare Advantage-insured, respectively. In prospective models, the clinical-risk grouper explained 12.04% and 31.68% of the variance in costs of commercially and Medicare Advantage-insured targeted IBD therapy users and 6.73% and 4.65% of the variance in nonusers, respectively. No evaluated predictors improved prospective model performance.

Our novel concurrent and prospective risk-adjustment models for patients with IBD explain and predict IBD-related costs using administrative claims data, which may offer real-world utility in multiple settings.