O2061: Association of Medical Complexity and Language with Clinical Outcomes in the Pediatric Intensive Care Unit

Background: Using a language other than English (LOE) is common in the United States, yet children in families who use LOE face barriers to equitable care and are more likely to be technology dependent than patients from English proficient families. We hypothesize that in the pediatric intensive care unit (PICU), patients in families who use LOE are more likely to have children with medical complexity (CMC) than their English proficient counterparts and that the coexistence of medical complexity and LOE will have an additive negative effect on clinical outcomes.

Methods: This is a single-site observational study of PICU encounters from 9/1/2017-8/31/2022. CMC were defined as having at least 3 chronic conditions or technology dependence using the Chronic Condition Classification System. We performed univariate and multivariate analyses between demographic factors and the presence of medical complexity. We described differences in outcomes for CMC between language groups using a mixed effects regression model to compare effects of language on outcomes including PICU and total hospital length of stay, ventilator and organ dysfunction free days, total hospital charges, and mortality after adjusting for covariates including severity of illness.

Results: There were 6,802 unique patients with 10,011 total encounters. A total of 5,199 encounters (51.9%) involved CMC and 1433 (14.3%) encounters involved patients from families who use LOE. In univariate analysis, Spanish (OR 1.44; 95% CI 1.32-1.57) and non-Spanish LOE (OR 1.58; 95% CI 1.32-1.84) were both associated with increased odds of medical complexity compared to families who use English. In multivariate analysis, Spanish LOE (OR 1.30; 95% CI 1.15-1.45) and non-Spanish LOE (OR 1.37; 95% CI 1.08-1.66) were independently associated with medical complexity. CMC who used a non-Spanish LOE had 6% fewer organ dysfunction free days (p = 0.003) than CMC with familial English use. Among CMC who used a non-Spanish LOE, PICU length of stay was 1.49 times longer (p = 0.02), total hospital stay was 1.44 times longer (p = 0.01), and total cost was 2.46 times higher (p < 0.001) compared to CMC with familial English use. Among patients without medical complexity, hospital stay was 1.82 times longer (p < 0.001) for the non-Spanish LOE group than patients with familial English use. There were not significant differences in outcomes for the Spanish use LOE group compared to English for organ dysfunction, ventilator free days, LOS, and cost. No mortality differences were identified.

Conclusion: Familial use of LOE was independently associated with increased medical complexity among patients in the PICU. Among the CMC population, non-Spanish LOE was associated with longer PICU LOS, longer hospital stay, and higher cost compared to the CMC with familial English use cohort.