P3.J199: Trends of Pediatric Inpatient Admissions and ICU Utilization from 2018-2023

Background: Current trends show a rapid increase in children receiving care in an intensive care unit (ICU) setting before and during the pandemic. A changing health care landscape alongside the closure of community ICU beds may further exacerbate an already strained system. Current, continuously reporting time trends of pediatric inpatient admissions are lacking. This study examines patient-level descriptives and trends of pediatric inpatient admissions from 2018-2023 to highlight changes in ICU and resource utilization, and in-hospital outcomes.

Methods: A retrospective cohort study of n=2,326,107 pediatric inpatient admissions were identified from the Vizient® Clinical Data Base (CDB) between 01/2018-12/2023. Patient/hospital-level characteristics were abstracted from claims, the Vizient Vulnerability Index was leveraged to examine ZIP-code level vulnerability, and cost-to-charge ratios were used to estimate cost to produce care. MS-DRGs were used to classify cases. Two sample t-tests and Chi Square tests were utilized to compare continuous and categorical variables, respectively.

Results: Trends in pediatric inpatient admissions, among those without ICU utilization (n=1,888,005), showed seasonality with dips in admissions in early 2020 followed by peaks in Q3 of 2022/2023, well above Q3 2018/2019.

Among those with ICU (n=438,102; non-NICU), trends were less stable post-2020, with more significant differences in total admissions quarter to quarter. Average percent utilization increased from 2018 (17.5%) to 2023 (18.2%). Average ICU length of stay increased 6% in 2023 over 2018 (p < 0.001); examined by quarter, there was an increase of 7% in Q3 of 2022 and 10% in Q3 of 2023, over Q3 of 2018 (p < 0.001). Regional trends show that ICU utilization in the South trends 13% higher compared to all other regions (p < 0.001).

Children with inpatient admissions and any ICU utilization were significantly younger (ICU:7±6 years; non-ICU:8±6 years), more likely Black (ICU:21%; non-ICU:18%), Medicaid usage (ICU:68%; non-ICU:53%), higher vulnerability (ICU:22%; non-ICU:20%) and at a teaching hospital (ICU:78%; non-ICU:68%), compared to those without ICU (p < 0.001). Examining in-hospital outcomes, those with any ICU utilization had significantly (p < 0.001) longer length of stays (ICU:9±22; non-ICU:5±12), higher cost (ICU:$41,478±120,089; non-ICU:$13,827±42,877) and had a higher rate of in-hospital mortality (ICU:2%; non-ICU:0.2%).

Those with ICU utilization were more likely (p < 0.001) to have an MS-DRG related to the respiratory (ICU:30%; non-ICU:18%), circulatory (ICU:10%; non-ICU:2%), or nervous system (ICU:15%; non-ICU:9%); whereas those without ICU utilization were more likely to have an MS-DRG related to mental diseases or disorders (ICU:15%; non-ICU:0.4%).

Conclusion: Despite bed closures and strained workforce, ICU lengths of stay are increasing, with significant regional variations. The differences in patient-level characteristics are notable when combined with disproportionate odds of in-hospital mortality, higher cost to produce care, longer lengths of stay, and increased clinical complexity. The demand increase for critical care services, particularly among vulnerable populations, must be evaluated across the health care continuum.