TL;DR
This study leverages a large national database to identify factors associated with repeated hospital admissions for febrile seizures (FS) in young children. It reveals that younger age (1-3 years), male gender, and very low household income are significant predictors of readmission, offering crucial insights for targeted follow-up care and potential interventions.
What Problem Does This Paper Address?
Febrile seizures (FS) are common in young children, typically considered benign. While the overall incidence and prevalence of FS are well-studied, there's a significant knowledge gap regarding the factors that lead to repeated hospital admissions for these events. Previous research has identified risk factors for developing FS, but the specific drivers for readmission – which can indicate more complex underlying issues or inadequate initial management – remain unclear. This paper aims to fill this gap by analyzing a national dataset to pinpoint which children are most likely to be readmitted after an initial FS hospitalization and why.
Methods
The researchers employed a retrospective observational study design using the National Readmission Database (NRD) from 2010-2014. This database is a robust source of information on hospitalizations and readmissions across the US.
To analyze the data, they used two primary statistical approaches:
- Multivariable Negative Binomial Regression: This was used to investigate factors associated with the rate of repeated admissions. This method is suitable for count data (number of readmissions) and accounts for over-dispersion, which is common in such datasets. They used log-transformed person-months as an offset to calculate incidence rate ratios (IRRs).
- Cox Proportional Hazard Model: This was used to assess the time to readmission, providing a complementary perspective. For this model, patients with more than two readmissions were excluded to ensure robustness.
Both models were adjusted for patient and hospital characteristics, as well as the year of admission, to control for potential confounding factors.
Datasets
The primary dataset used in this study is the National Readmission Database (NRD) for the years 2010-2014. The NRD is a unique database constructed from 21 State Inpatient Databases (SID) and is designed to estimate national readmission rates for all payers and the uninsured. It contains approximately 14 million hospital discharge records annually. The paper highlights that the NRD is a strength because it represents a broad spectrum of the US pediatric population, unlike more specialized databases.
Key Insights and Ideas
- Beyond the Initial Seizure: The study emphasizes that while FS is often benign, repeated admissions signal a need for closer attention. This suggests that some children experiencing FS may have underlying vulnerabilities or complications that aren't immediately apparent.
- Age Matters, But Specifically 1-3 Years: The finding that children aged 1-3 years are at significantly higher risk for readmission, with a peak at age 2, is crucial. This age group is characterized by rapid brain development and a higher susceptibility to infections, potentially explaining the increased risk.
- Socioeconomic Disparities: The strong association between very-low household income and increased readmission risk is a stark reminder of how socioeconomic factors can influence health outcomes, even for seemingly common conditions. This points to potential systemic issues in access to care or environmental factors affecting children in lower-income households.
- Gender Predominance: The finding that boys are more likely to be readmitted aligns with existing literature suggesting higher susceptibility to infections and potentially genetic predispositions in males.
Key Results
- The overall readmission rate for febrile seizures was approximately 3.45 cases per 1000 person-months after hospital discharge.
- Children aged 1-3 years had significantly higher readmission rates, with a peak at age 2 (adjusted IRR, 5.09).
- Male children had 1.75 times higher rates of readmission compared to girls.
- Children from very-low income households were 2.57 times more likely to be readmitted compared to those from very-high income households.
- No significant differences in readmission rates were found based on insurance status, residential area, weekday/weekend admission, emergency department use, or the presence of chronic diseases.
Limitations and Future Work
The authors acknowledge several limitations:
- Potential for Misclassification: The use of ICD-9-CM codes means that miscoding or misclassification of FS and comorbid conditions could lead to underestimation of rates.
- Limited Clinical Detail: The NRD lacks detailed clinical information such as seizure duration, physical examination findings, laboratory results, or imaging data, which could provide deeper insights.
- Inability to Capture Non-Hospitalized Recurrences: The study can only analyze readmissions, meaning it cannot capture the recurrence rate of FS in children who do not require hospitalization.
Future work could focus on:
- Investigating the specific types of complications or underlying conditions that lead to readmission in these high-risk groups.
- Exploring the impact of specific interventions or discharge planning strategies on reducing readmission rates.
- Further research into the mechanisms behind the observed socioeconomic disparities to inform targeted public health initiatives.
Summary updated