The Association of Workload and Outcomes in the Pediatric Cardiac ICU*

OBJECTIVES: Healthcare workload has emerged as an important metric associated with poor outcomes. To measure workload, studies have used bed occupancy as a surrogate. However, few studies have examined frontline provider (fellows, nurse practitioners, physician assistants) workload and outcomes. We hypothesize frontline provider workload, measured by bed occupancy and staffing, is associated with poor outcomes and unnecessary testing. DESIGN: A retrospective single-center, time-stamped orders, ordering provider identifiers, and patient data were collected. Regression was performed to study the influence of occupancy on orders, length of stay, and mortality, controlling for age, weight, admission type, Society of Thoracic Surgery-European Association for Cardio-Thoracic Surgery Congenital Heart Surgery Mortality score, diagnosis, number of surgeries, orders, provider staffing, attending experience, and time fixed effects. SETTING: Twenty-seven bed tertiary cardiac ICU in a free-standing children's hospital. PATIENTS: Patients (0-18 yr) admitted to the pediatric cardiac ICU, January 2018 to December 2019. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: There were 16,500 imaging and 73,113 laboratory orders among 1,468 patient admissions. Median age 6 months (12 d to 5 yr), weight 6.2 kg (3.7-16.2 kg); 840 (57.2%) surgical and 628 (42.8%) medical patients. ICU teams consisted of 16 attendings and 31 frontline providers. Mortality 4.4%, median stay 5 days (2-11 d), and median bed occupancy 89% (78-93%). Every 10% increase in bed occupancy had 7.2% increase in imaging orders per patient (p < 0.01), 3% longer laboratory turn-around time (p = 0.015), and 3 additional days (p < 0.01). Higher staffing (> 3 providers) was associated with 6% less imaging (p = 0.03) and 3% less laboratory orders (p = 0.04). The number of "busy days" (bed occupancy > 89%) was associated with longer stays (p < 0.01), and increased mortality (p < 0.01). CONCLUSIONS: Increased bed occupancy and lower staffing were associated with increased mortality, length of stay, imaging orders, and laboratory turn-around time. The data demonstrate performance of the cardiac ICU system is exacerbated during high occupancy and low staffing.