Cognitive Outcomes and Delirium After Cardiac Neurodevelopmental Program Implementation for Children With Congenital Heart Disease

Abstract

Importance: A recent advisory from the American Heart Association delineated the potential benefits of developmental care for hospitalized children with congenital heart disease (CHD) and a critical gap in research evaluating the association of such inpatient programs with neurodevelopmental outcomes. Objective: To investigate associations between the Cardiac Inpatient Neurodevelopmental Care Optimization (CINCO) program interventions, delirium, and neurodevelopment in young children (newborn through age 2 years) hospitalized with CHD. Design, Setting, and Participants: This cohort study used quality improvement data from inpatient cardiac units at a tertiary care children's hospital in the US. Participants were children aged 0 to 2 years who were admitted for at least 7 days from September 1, 2018, to September 1, 2023. The CINCO program was implemented on September 1, 2020, in 6-month plan-do-study-act phases. Exposures: The 5 CINCO interventions were medical and/or nursing order panels, developmental kits, bedside developmental plans, caregiver mental health support handouts, and developmental care rounds. Main Outcomes and Measures: Number of days with delirium per patient, which was measured using the Cornell Assessment of Pediatric Delirium (a score higher than 9 indicated delirium). Neurodevelopment was measured using the Bayley Scales of Infant and Toddler Development, Fourth Edition (BSID-4). Results: The full sample included 1331 qualifying admissions for 1019 unique pediatric patients (median [range] age at admission, 3.65 [0-34.62] months; 771 males [57.9%]), with a subcohort of 121 unique patients (median [range] age at admission, 0.00 [0-9.85] months; 77 males [63.6%]) whose initial hospitalization occurred before age 10 months and who underwent BSID-4 evaluation at age 12 months or older. The mean (SD) number of days with delirium per patient was stable for 2 years prior to CINCO implementation, decreased by 54.0% between phases 1 and 2 of the CINCO program (from 3.05 [0.60] to 1.38 [0.21]), and then remained stable over time. Each of the 5 CINCO interventions was associated with lower delirium after false discovery rate (FDR) correction (eg, medical and/or nursing order panel: B = -1.376 [95% CI, -1.767 to -0.986]; F2,1273 = 47.767; partial η2 = 0.036; P