P346 Decreasing disparities in and use of systemic corticosteroid and disparities in use among paediatric patients with Crohn disease in a multicentre collaborative

Adler, J.(1);Kandavel, P.(2);Eder, S.(1);Saeed, S.(3);

(1)University of Michigan, Susan B. Meister Child Health Evaluation and Research Center, Ann Arbor, United States;(2)Children's Hospital of Los Angeles, Pediatric Gastroenterology, Los Angeles, United States;(3)Dayton Children's Hospital, Pediatric Gastroenterology, Dayton, United States; ImproveCareNow Network


Systemic corticosteroids (CS) have long been used to treat Crohn disease. However, cumulative CS exposure is associated with adverse effects, particularly among growing children. Steroid sparing therapies are increasingly available. It remains unknown to what extent CS use has decreased among children with Crohn disease.


This was an observational cohort study using data from ImproveCareNow Network, multi-centre international paediatric inflammatory bowel disease collaboration (2007-2018). Paediatric patients diagnosed with Crohn disease were included. We performed serial cross-sectional analyses of period prevalence for entire population. Patients enrolled <30 days from diagnosis were included in logistic multivariate regression to predict odds of CS use 120-365 days after diagnosis. All families provided consent/assent.


The total population included 17,646 patients (82% white, 59% male). Systemic CS were used by 5,862 (33%), more commonly among black than white patients (37% vs 32%; p<0.001). Those with inflammatory phenotype had more CS use than stricturing or penetrating (34%, 28%, 22% respectively; p<0.001).

Overall, CS use decreased from 28% (2007) to 9% (2018). Excess CS use among blacks compared to whites reduced from mean 17% in first 3 years to 3% in last 3 years (Figure 1). Changes in other medications appear in Figure 2. Multivariate results are in Table 1. CS or 5-aminosalicylate use <120 days after diagnosis predicted later increased odds of CS use. Anti-tumour necrosis factor-α (anti-TNFα) use <120 days after diagnosis was associated with reduced odds of later CS use.


Corticosteroid use among paediatric patients with Crohn disease in the ImproveCareNow Network has decreased over time. Racial disparities in CS use were found, but gradually improved. Early 5-aminosalicylate or CS use predicted later CS use. Early anti-TNFα treatment is associated with reduction in later CS use.

Figure 1. Corticosteroid use 120-365 days after Crohn disease diagnosis by race and year

Figure 2. Changes in medication use over time
Figure 2. Changes in medication use over time

Table 1. Multivariate predictors of corticosteroid use 120-365 days after Crohn disease diagnosis (N=2,257).

CharacteristicOdds ratio
(95% Confidence interval)
Age at diagnosis1.00 (0.96-1.04)0.94
Gender0.87 (0.67-1.13)0.29
White race1.18 (0.84-1.67)0.34
Height Z-score1.00 (0.87-1.16)0.95
Weight Z-score1.06 (0.93-1.22)0.36
Phenotype0.96 (0.83-1.11)0.56
Year of diagnosis0.92 (0.86-0.98)0.01
Corticosteroid use
<120 days after diagnosis 
3.42 (2.52-4.63)<0.01
5-aminosalicylate use
<120 days after diagnosis
1.41 (1.08-1.86)0.01
Immunomodulator use
<120 days after diagnosis
0.94 (0.71-1.25)0.68
Anti-tumour necrosis factor-α use
<120 days after diagnosis
0.50 (0.35-0.71)<0.01