REAL-WORLD RISK STRATIFICATION IN CROHN’S DISEASE: PATTERNS IN DIAGNOSIS, TREATMENT, AND UTILIZATION

Abstract Crohn’s disease (CD) is a chronic and recurrent condition requiring long-term management. The American Gastroenterological Association (AGA) provides a risk stratification tool to guide treatment selection.1 Using this tool in claims databases is challenging due to missing clinical indicators (ie, anatomic involvement). Without medical record abstraction, assessing moderate to high (M/H) risk CD prevalence is difficult. This study quantified the proportion of CD patients meeting AGA-defined M/H risk criteria and described their demographic, clinical, and healthcare resource utilization (HCRU) patterns by risk status. This retrospective cohort study used the Healthcare Integrated Research Database (HIRD®), a large, nationwide claims-based real-world data source of commercially insured and Medicare Advantage populations in the US. Adults diagnosed with CD between 11/17/2019 and 8/16/2024 were eligible if they had an outpatient CD encounter with a gastroenterologist (most recent being the index date), were ≥18 years old on index, had 1 additional CD diagnosis during the year prior to the index date, no prior evidence of ulcerative colitis (UC), and ≥685 days of continuous enrollment prior to and ≥45 days following the index date. To address missing clinical indicators in claims, a subset of 700 patients underwent medical record abstraction. This enabled application of AGA criteria for M/H risk classification, as detailed in the figure legends. Patient characteristics and HCRU were assessed using claims data during the 1-year pre-index period and stratified by AGA risk status. Among 7,835 eligible patients, mean (SD) age was 49.2 (16.9) years, and 54.2% were female. Of the 700 abstracted patients, 490 (70.0%) were M/H risk. Compared to low-risk patients, M/H risk patients were younger (mean [SD] age 49.5 [17.1] vs 56.2 [13.8] years old), and less likely to be female (54.1% vs 61.4%). CD-related diagnostic testing and imaging (e.g., CT, MRI) were common (52.9% low risk and 63.7% M/H risk) and surgeries (colectomy, resection) were more frequent among M/H risk patients. Of the M/H risk, 257 (52.4%) received any advanced therapy during the same year (vs. 94 [45.7%] for those at low risk). HCRU was higher among M/H risk patients, including all-cause inpatient and emergency department (ED) encounters, compared to patients at low risk (Inpatient: 17.3% vs 10.0%, ED: 27.3% vs 19.0%). In this retrospective cohort study, 70% of CD patients met the AGA definition of M/H risk and 52.4% received advanced therapy. Patients with M/H risk had higher rates of inpatient and ED use, imaging, surgeries, and diagnostic testing than low risk patients. 1Sandborn WJ. Crohn’s disease evaluation and treatment: clinical decision tool. Gastroenterology. 2014 Sep;147(3):702-5. https://www.gastrojournal.org/article/S0016-5085(14)00918-4/fulltext. Figure 1:Baseline demographics Abbreviations: N/n, number; %, percent; SD, standard deviation; IQR, interquartile range; NH, non-Hispanic) Risk classification: Based on AGA criteria. M/H risk: Age at initial diagnosis < 30 years, extensive anatomic involvement, perianal and/or severe rectal disease, deep ulcers, prior surgical resection, stricturing and/or penetrating behavior. Patients not meeting M/H risk criteria were classified as low risk. * Cells with values of 1 to 4, or cells where values 1 to 4 can be derived, have been masked per Carelon Research privacy policies. The Northeast cell (>1,440*) was masked to prevent back-calculation of the of the value in the ‘unknown’ region.Figure 2:Clinical measures and healthcare resource utilization during the 1-year pre-index period Abbreviations: N/n, number; %, percent; SD, standard deviation; HCRU, healthcare resource utilization; CD, Crohn’s Disease. Risk classification: Based on AGA criteria. M/H risk: Age at initial diagnosis < 30 years, extensive anatomic involvement, perianal and/or severe rectal disease, deep ulcers, prior surgical resection, stricturing and/or penetrating behavior. Patients not meeting M/H risk criteria were classified as low risk. * Cells with values of 1 to 4, or cells where values 1 to 4 can be derived, have been masked per Carelon Research privacy policies. 1 CD-related imaging includes CT scans, endoscopy, MRI, and ultrasonography. 2 CD-related surgeries includes cecostomy, colectomy, endoscopic balloon dilation of the colon, fistulectomy/fistulotomy, incision/drainage of the colon, rectum, or anus, intestinal resection, placement or removal of seton, J-pouch formation, proctectomy, proctocolectomy, repair of anal fistula, or J-pouch, and stricturoplasty. 3 CD-related diagnostic testing includes abdominal MRI, any CT scans, colonoscopy, endoscopy, HRCT, magnetic resonance enterography, MRI, and ultrasound (abdominal, pelvic). 4 Advanced therapy for CD includes infliximab and biosimilars, adalimumab and biosimilars, ustekinumab and biosimilars, certolizumab pegol, vedolizumab, natalizumab, risankizumab, mirikizumab, guselkumab, and upadacitinib.