Racially disparate care has been shown to contribute to suboptimal healthcare outcomes for minorities in some settings. Using data from the ImproveCareNow (ICN) network, we sought to investigate differences in the medical management and outcomes of pediatric patients with Crohn's disease (CD) at time of diagnosis and 1-year post-diagnosis. ICN is a learning health network for children with inflammatory bowel disease. It contains data collected prospectively and longitudinally during outpatient encounters. Inclusion criteria: all CD patients ≤21 years of age with a visit between September 2006 and October 2014, with the first recorded encounter ≤90 days from the date of CD diagnosis and an encounter 1-year post diagnosis (within 60 days). Analyses were conducted at 2 time periods: baseline (diagnosis) and the 1-year encounter. Analyses included summary statistics, nonparametric Kruskal-Wallis test, Χ2, and Fisher's exact tests as appropriate. Baseline: 976 patients (Black = 118 [12%], White = 858 [88%], mean age = 13 years, 63% male) from 39 sites were included. Black children had a higher percentage of Medicaid insurance (44% versus 11%, P < 0.001). There were no race differences in gender, age at diagnosis, BMI, disease behavior or disease location. Black children had more active disease according to Physician Global Assessment (PGA) at diagnosis (remission 29% versus 35%, mild 37% versus43%, moderate/severe 34% versus 22%; P = 0.027), though there was no difference according to the short Pediatric Crohn's Disease Activity Index (sPCDAI) (P = 0.67). Black children had a lower hematocrit (34.8 versus 36.7, P < 0.001) and albumin level (3.6 versus 3.9, P = 0.001). There were no differences in initial treatment (Black versus White: anti-TNFα 12% versus 11%, P = 0.94; thiopurines 32% versus 34%, P = 0.75); corticosteroids 58% versus 61%, P = 0.71). Black children had more psychosocial concerns noted at diagnosis (30% versus 12%, P = 0.021). At 1 year following diagnosis, Black children had more active disease according to PGA (remission 63% versus 75%, mild 26% versus 29%, moderate/severe 11% versus 6%; P = 0.016); however, there was no difference in disease activity based on the sPCDAI (P = 0.060). Black children had a lower hematocrit (36.6 versus 37.9, P = 0.004), lower albumin level (3.9 versus 4.2, P <0.001), and higher erythrocyte sedimentation rate (20.0 versus 15.2, P = 0.03). There were no differences in treatment (Black versus White: anti-TNFα 35% versus 33%, P = 0.69; thiopurines 47% versus 44%, P = 0.52); corticosteroids 7% versus 10%, P = 0.32). In this cohort, minor differences were identified between Black and White children at baseline and 1-year, with Black children having slightly worse disease activity by some measures. Specifically, statistical differences were identified with respect to hematocrit, albumin, and erythrocyte sedimentation rate (at 1-year), although these differences may not have been clinically meaningful. Remission rates were lower in Black children according to PGA, although not according to C-reactive protein and sPCDAI, suggesting that either Black children were perceived to be more ill despite similar disease activity, or that Black children actually were more ill, but this was not adequately reflected by the sPCDAI. Racial differences in treatment were not identified.
Read full abstract