Pediatric admissions declined over the past two decades, but a greater proportion of these hospitalized kids have landed in the intensive care unit (ICU), and with increasingly complex cases, a retrospective study spanning 21 states found.
In the population-based cohort study of over 2 million pediatric admissions from 2001 to 2019, hospitalizations decreased from 30.1 to 19.7 per 1,000 children (P<0.001). And while a slight dip in ICU care overall was seen (3.2 to 3.0 per 1,000 children), the percentage of admissions that included a non-neonatal ICU stay increased from 10.6% to 15.5%, reported researchers led by Elizabeth Killien, MD, MPH, of Seattle Children's Hospital.
Furthermore, the medical complexity of ICU cases grew during this time, with an increasing proportion of kids having an underlying comorbidity (46% in 2001 vs 57% in 2019) and needing feeding tubes, respiratory support, or other preadmission technology (16% vs 24%; P<0.001 for both), they detailed in .
"As hospitals increasingly encounter constraints in nursing and respiratory therapy staffing, it is important to note that the average child requiring an ICU bed in 2023 is very different than the average child who required an ICU bed 10 or 20 years ago," Killien told in an email.
"Pediatric ICU patients are increasingly medically complex, more frequently develop organ failure and require mechanical ventilation, and have longer hospital stays," she added. "Estimates of the staffing required to care for a child in the ICU must be based on current and future needs rather than past needs."
Cases involving multiple organ dysfunction syndrome increased from 7% to 21% over the study period, and the percentage of children receiving mechanical ventilation rose from 15% to 17% (P<0.001 for both). As for length of stay in cases involving ICU care, the proportion of children with a stay of more than 28 days increased from 5% to 6% (P<0.001), the researchers found.
The cost of a pediatric hospitalization involving ICU care rose as well, from $10,009 to $15,123 between the two periods in 2019 dollars, which after adjusting for inflation represents a near doubling, according to the study. Nationally, an estimated 239,000 children were admitted to a non-neonatal ICU in 2019, "corresponding to $11.6 billion in hospital costs," noted Killien and co-authors.
Researchers used the Healthcare Cost and Utilization Project (HCUP) state inpatient databases from a total of 21 states across 5 years: seven states in 2001, eight in 2004, 17 in 2010, 20 in 2016, and 16 in 2019.
Hospitalized children, ages 0 to 17 years, were included in the population-based retrospective study, while newborns (during birth hospitalization) and patients admitted to rehabilitation facilities or psychiatric hospitals were excluded. Of 2,157,991 pediatric admissions, 275,656 (12.8%) included ICU care.
Killien's group reported that the proportion of ICU admissions in children's hospitals increased from 51% to 85%, as did the the proportion of ICU admissions in dedicated pediatric ICUs (PICUs), from 62% to 81% (P<0.001 for both).
"ICU care is increasingly regionalized in children's hospitals and dedicated pediatric intensive care units," Killien said.
"While care in dedicated pediatric facilities may be beneficial for many children, regionalization of care may also contribute to lack of timely access to pediatric care for children in rural areas," she continued. "Closures of pediatric units in community hospitals across the country is likely to increasingly strain our pediatric healthcare system, especially as dedicated pediatric units are increasingly admitting children with high medical complexity and illness severity."
In terms of mortality, in-hospital mortality for children admitted to an ICU decreased from 2.5% to 1.8% (P<0.001), largely due to a decrease in mortality in dedicated PICUs and medical ICUs, according to the authors. However, mortality increased over time in surgical ICUs (1.7% to 3.3%) and cardiac ICUs (1.5% to 2.3%).
"By 2016, cardiac failure had become the second most common organ failure among all ICU admissions, despite little change in admissions to dedicated cardiac ICUs between 2001 and 2019," Killien's group stated.
Study limitations included the fact that not all states were represented because some states do not contribute publicly available data to the HCUP and not all include revenue codes to identify ICU care, according to the authors. Other limitations included that all counts refer to admissions rather than unique patients, and that ICU type was identified by revenue codes designated by each institution and could not be independently verified.
The study's findings, "should inform development of a multidisciplinary collaboration across pediatric critical care professionals, epidemiologists, and health services researchers to identify strategies to reduce disparities in pediatric critical illness and better prepare the pediatric critical care community and U.S. health care system for anticipated capacity and resource needs with an increasingly large, diverse, and medically complex patient population," Killien and colleagues wrote.
Disclosures
The study was supported by an award from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and grants to the University of Washington from the National Center for Advancing Translational Sciences (NCATS).
Killien disclosed support from NICHD and NCATS. A co-author disclosed support from the NIH.
Primary Source
JAMA Pediatrics
Killien EY, et al "Epidemiology of intensive care admissions for children in the US from 2001 to 2019" JAMA Pediatr 2023; DOI: 10.1001/jamapediatrics.2023.0184.