P2.D069: Hospital-Level Differences in Delivery of Low-Value Care by Race and Ethnicity

Background: Low-value care is a common problem in pediatric emergency medicine. Racial and ethnic differences in low-value care delivery are understudied. Better understanding such differences is essential for anticipating the impact of quality improvement efforts and may inform interventions to improve health equity as a core dimension of quality.

Methods: To assess racial and ethnic differences in the provision of LVC for common pediatric emergency conditions within and across US children’s hospitals, we performed a retrospective, cross-sectional study of children’s hospitals in the Pediatric Health Information System (PHIS). We included emergency department (ED) encounters for asthma, bronchiolitis, headache, and minor head injury (MHI), as defined by the PHIS Low Value Care Report, from January 2021 to June 2023 for children 0-18 years. We used a composite LVC outcome for each condition, classifying an encounter as LVC if any LVC intervention for the condition occurred (e.g., bronchodilator, steroids, CXR, antibiotic, or blood culture for bronchiolitis). We estimated multivariable logistic regression models for the outcome of LVC by race and ethnicity (non-Hispanic White (NHW) vs. non-Hispanic Black (NHB), Hispanic vs. NHB, NHW vs. Hispanic), stratified by hospital and condition. We adjusted for patient age, sex, weekend presentation, hour of presentation, insurance type, and distance from the hospital, reporting adjusted odds ratios (aOR) with 95% confidence intervals (CI).

Results: There were 314,138 included encounters during the study period (Table 1) with an overall LVC prevalence of 24.1%. LVC prevalence varied by condition: asthma 18.2% (range 6.0-52.7 across hospitals); bronchiolitis 31.8% (10.6-74.7); headache 24.4% (10.8-38.9); and MHI 25.5% (9.3-42.8). NHW compared to NHB patients had increased odds of LVC, reported as pooled aOR across hospitals, for all conditions except bronchiolitis: asthma aOR 1.51 (1.46, 1.58); bronchiolitis aOR 0.94 (0.90, 0.99); headache aOR 1.57 (1.48, 1.67); MHI aOR 1.10 (1.04, 1.56) (Figure 1a). There were also differences between NHW vs. NHB within hospitals (21/37 hospitals had significant aOR >1.0 for asthma; 2/37 for bronchiolitis; 21/37 for headache; 9/37 for MHI) (Figure 1b). Similarly, NHW compared to Hispanic patients had increased odds of LVC for all conditions except bronchiolitis. Hispanic compared to NHB patients had increased odds of receiving LVC for asthma (pooled aOR 1.33 (1.28, 1.38); 13/37 hospitals with significant aOR >1), but no difference for bronchiolitis (pooled aOR 1.03, 95% 0.99-1.08; 0/37 hospitals) and headache (pooled aOR 1.27, 95% CI 1.20, 1.34; 7/37 hospitals), and decreased odds for MHI (pooled aOR 0.83, 95% CI 0.79, 0.88; 3/37 hospitals).

Conclusion: There are considerable differences by race and ethnicity LVC use by race and ethnicity, both within and across PHIS hospitals. Future examination of hospital factors associated with these differences as well as the differential impact of quality improvement efforts by race and ethnicity may help advance equitable care.