Patterns and Predictors of Steroid Use in a Real-World Inflammatory Bowel Disease Cohort

Patterns of steroid use in inflammatory bowel disease remain poorly characterized in real-world settings. Steroid exposure is associated with adverse effects and often indicates suboptimally controlled disease. Therefore, patterns and predictors of steroid use in a large inflammatory bowel disease cohort were examined.

Steroid exposure over a 3-year window was explored. Use was classified by duration-short (1-28 days), moderate (29-56 days), prolonged (> 56 days), and recency (within last year, prior years or no exposure). Associations with demographic and disease-related factors were assessed using multivariable logistic regression.

Among 5436 people (median age 42 years, IQR 32-56), 18.3% (n = 994) were steroid exposed. 57.6% had Crohn's disease and 50.2% were female. Crohn's disease was associated with lower odds of both prolonged and recent exposure compared to ulcerative colitis (AOR 0.72, p = 0.001 and AOR 0.78, p = 0.037, respectively). Females had a greater likelihood of both prolonged and recent exposure (AOR 1.22, p = 0.048 and AOR 1.23, p = 0.041, respectively). Young adults (20-29 years) had higher odds of prolonged and recent use than those > 70 years (AOR 6.59 and 9.12, respectively, p < 0.001). Combination immunomodulator and advanced therapy use was associated with a higher likelihood of both prolonged and recent use compared to 5-aminosalicylic acid therapy alone (AOR 4.01, p = 0.002 and AOR 4.54, p < 0.001). Age at diagnosis had a modest effect size (AOR 1.03, p < 0.001).

Steroid use was modest, with over 80% unexposed over 3 years. Proactive optimization of therapy, particularly in younger individuals and those with ulcerative colitis, may further reduce steroid exposure.