Incidence and Mortality Trends in Critically Ill Children: A Korean Population-Based Study

Jaeyoung Choi, Esther Park, Ah Young Choi, Meong Hi Son, Joongbum ChoJournal of Korean Medical Science 2023 · DOI

Abstract

Mortality among critically ill children improved during the study period, and the improving trend was prominent in children with high treatment requirements. Varying mortality trends, according to ICU organizations, highlight that advances in medical knowledge should be supported structurally.

Introduction

Global data suggests that pediatric critical care outcomes are improving, yet comprehensive, nationwide analyses are often lacking in specific regions. In South Korea, the 2010s marked a period of significant structural shifts in the medical landscape, including a 2014 mandate for dedicated intensivists in tertiary hospitals and the 2015 establishment of specific reimbursement codes for pediatric intensive care units (PICUs). This population-based study sought to determine if these policy and organizational changes translated into better outcomes for the country's most vulnerable pediatric patients. The findings reveal a significant 25% relative decrease in mortality for critically ill children in South Korea between 2012 and 2018, driven largely by structural improvements.

Methods

The researchers conducted a retrospective cohort study using a "big data" approach, analyzing the Korean National Health Insurance (KNHI) database. This database, managed by the Health Insurance Review and Assessment (HIRA) service, covers 97% of the Korean population, providing a high-resolution view of every procedure, prescription, and diagnosis across the national healthcare system.

Study selection flowchart showing the filtering of over 7 million pediatric admissions to 54,843 included ICU cases.

As illustrated in the study selection flowchart, the team screened over 7 million pediatric admissions from January 2012 to December 2018. They narrowed the focus to 54,843 ICU admissions for children aged 0–18. Neonates (under one month old) were excluded because neonatal ICUs in Korea operate under a unique organizational structure that would have skewed the data. The team utilized multivariable logistic regression to calculate the odds ratio of in-hospital mortality, adjusting for age, sex, primary diagnosis, and treatment intensity (e.g., mechanical ventilation or ECMO). Trend significance was evaluated using the Cochrane-Armitage test.

Key Results and Insights

The study found that the average mortality rate for critically ill children was 4.4%. Over the seven-year study period, mortality fell from 5.5% in 2012 to 4.1% in 2018 ($P < 0.001$). After adjusting for patient factors, the risk of in-hospital mortality decreased by 9.2% every year.

The Importance of Specialized Care

A primary insight of the study is that structural factors—specifically the presence of dedicated PICUs and specialized intensivists—are the primary drivers of improved survival. Mortality in dedicated PICUs dropped by a relative 36%, whereas general ICUs only saw a 10% reduction. This suggests that placing a child in an adult-oriented ICU yields significantly worse outcome trends compared to specialized environments.

Furthermore, the presence of dedicated intensivists was a major predictor of survival, validating the 2014 policy mandate. The data shows that having a specialized doctor physically present in the unit is a structural necessity for optimizing outcomes.

Stability Amidst Demographic Change

Despite Korea’s plummeting birth rate—the total pediatric population dropped by nearly 2 million during the study—the rate of ICU admissions remained stable at approximately 8.5 per 10,000. This indicates that the mortality improvement is not due to a shift toward "healthier" admissions; rather, the system has become genuinely more effective at saving the sickest patients. These gains were most prominent in children requiring aggressive interventions like mechanical ventilation or vasopressors, implying that the "ceiling" of medical achievement is rising when supported by specialized staff.

Discussion and Limitations

While the study highlights significant progress, certain subgroups have not seen the same gains. For instance, mortality for children with traumatic injuries did not show a significant decrease, highlighting a potential area for future focus.

The authors acknowledge limitations inherent in using administrative data, such as the lack of physiological data (e.g., blood pressure or lab values) required for standard clinical severity scores like PRISM or PIM3. Additionally, the database records the occurrence of death but not the specific mode (e.g., withdrawal of life support).

Future research should investigate why older children (>13 years) and those with traumatic injuries lag behind in survival gains. There is also a clear need for continued structural support and improved end-of-life care protocols in pediatric settings to ensure that medical advancements reach all patient populations effectively.

Summary updated

Abstract

Mortality among critically ill children improved during the study period, and the improving trend was prominent in children with high treatment requirements. Varying mortality trends, according to ICU organizations, highlight that advances in medical knowledge should be supported structurally.
BibTeX
@article{choi2023incidence,
  author = {Jaeyoung Choi and Esther Park and Ah Young Choi and Meong Hi Son and Joongbum Cho},
  title = {{Incidence and Mortality Trends in Critically Ill Children: A Korean Population-Based Study}},
  journal = {Journal of Korean Medical Science},
  year = {2023},
  doi = {10.3346/jkms.2023.38.e178},
}
Figure