Keeping children safe: a model for predicting families at risk for recurrent childhood injuries

Area of Science:

• Pediatrics
• Public Health
• Epidemiology

Background:

• Existing research on recurrent unintentional injury (UI) primarily focuses on individual child factors, neglecting family-level risks.
• Identifying families at high risk is crucial for developing targeted public health interventions.

Purpose of the Study:

• To develop a statistical model for identifying families at the highest risk of recurrent unintentional injuries in children.
• To inform the targeting of public health interventions for injury prevention.

Main Methods:

• A retrospective birth cohort study utilized hospital and emergency room (ER) records of children born between 2005 and 2012.
• Negative binomial regression analyzed predictive factors for recurrent child UI using a two-period approach.
• Sensitivity analyses were performed to ensure the model's robustness.

Main Results:

• Eight significant predictive factors for child injury were identified (P < 0.05): male gender, prior UI visits, illness visits, age 36-59 months, low birth weight (<1500 g), maternal ER visits, siblings' UI visits, and number of younger siblings.
• Some factors are established predictors, while others, like maternal ER visits and siblings' UI visits, are novel.
• Five factors remained significant across all sensitivity analyses, indicating model stability.

Conclusions:

• The identified factors can predict a child's risk of repeat UI and a family's cumulative UI risk.
• This novel statistical model offers a valuable approach for targeting public health interventions to high-risk families.
• Early identification and intervention can mitigate the burden of recurrent unintentional injuries.