Research article
Increased incidence of inflicted traumatic brain injury in children after a natural disaster

https://doi.org/10.1016/j.amepre.2003.10.023Get rights and content

Abstract

Background

The incidence of child abuse following natural disasters has not been studied thoroughly. However, parental stress and decreased social support have been linked to increased reports of child maltreatment. We hypothesized that a large-scale natural disaster (North Carolina's Hurricane Floyd) would increase the incidence of inflicted traumatic brain injury (TBI) in young children.

Methods

An ecologic study design was used to compare regions affected to those regions unaffected by the disaster. Cases of inflicted TBI resulting in admission to an intensive care unit or death from September 1998 through December 2001 in North Carolina were ascertained. Poisson regression modeling was employed to calculate rate ratios of injury for each geographic area by time period.

Results

Inflicted TBI in the most affected counties increased in the 6 months post-disaster in comparison to the same region pre-disaster (rate ratio 5.1, 95% confidence interval [CI]=1.3–20.4), as did non-inflicted TBI (rate ratio 10.7, 95% CI=2.0–59.4). No corresponding increased incidence was observed in counties less affected or unaffected by the disaster. The rate of inflicted injuries returned to baseline in the severely affected counties 6 months post-hurricane; however, the rate of non-inflicted injuries appeared to remain elevated for the entire post-hurricane study period.

Conclusions

Families are vulnerable to an elevated risk of inflicted and non-inflicted child TBI following a disaster. This information may be useful in future disaster planning.

Introduction

In September 1999 Hurricane Floyd dropped 20 inches of rain in eastern North Carolina, which caused extensive flooding in some counties. An estimated 2.1 million people were affected, with 52 deaths directly attributable to the hurricane, most from drowning.1 Damage was estimated at $6 billion. We hypothesized that the flooding and subsequent loss of, and disruption to, lives, property, and community ties may have contributed to an increase in parental stress and depression, and thus contributed to an increase in child maltreatment.2

A report examining the changes in both reporting and incidence of child abuse after three separate natural disasters—Hurricane Hugo in South Carolina, the Loma Prieta earthquake in California, and Hurricane Andrew in Louisiana—had mixed results.3 It found that following Hurricane Hugo, child abuse reports and confirmations were higher at 3 and 6 months compared to the previous year after adjusting for population change. The same pattern was seen for child abuse reports after the Loma Prieta earthquake, although confirmation of abuse was not possible because of inadequate data. However, this pattern was not seen after Hurricane Andrew. Curtis et al.3 theorized that the population may have experienced a lower stress level, given the relatively high frequency of severe weather (hurricanes and tropical storms) in Louisiana.

The potential for disasters to increase the rates of child abuse is proposed to stem from the additional stress conferred by the event, and the subsequent effects on mental health. The mental health effects of natural disasters including hurricanes, flooding, earthquakes, and volcano eruptions have been reported. Reactions to natural disasters are proposed to stem both from the stress caused directly by the disaster such as loss of life and property and disruption of the social fabric of the community.4 Mental health effects that have been documented include post-traumatic stress disorder, depression, and anxiety.5, 6 Three disaster experiences (injury, fear of death, and property loss) were associated with increased scores on the Trauma Symptom Inventory designed to measure psychological sequelae of traumatic events.5

Physical child maltreatment is the leading cause of death from injury among infants.7 Inflicted traumatic brain injury (TBI) (commonly referred to as shaken baby syndrome) is a common form of child abuse in the first year of life, estimated at about 30 per 100,000 infants,8, 9 and approximately 17 per 100,000 child-years in the first 2 years of life.9 Families at highest risk appear to be those with a first child, low maternal education, and minority status.9 Severe TBI requires urgent medical treatment, which allows ascertainment even under disaster conditions. Therefore, this ecologic study examined whether there was a temporal and geographic increase in inflicted TBI in young children in relation to Hurricane Floyd in North Carolina.

Section snippets

Case ascertainment

The study population consists of children aged ≤24 months who sustained a fatal TBI and/or a TBI serious enough to be admitted into any of the nine pediatric intensive care units (PICUs) or step-down units in North Carolina between the dates of September 1, 1998 through December 31, 2001. Data from the period of January 1, 2000 to December 31, 2001 were prospectively collected as part of an ongoing project about incidence of inflicted TBI in North Carolina. Data prior to January 1, 2000 were

Results

A total of 245 children were identified as having received a TBI during the entire 40-month study period. Thirty-seven (15%) injuries occurred in the 16 severely affected counties and 208 (85%) occurred in the less-affected/unaffected counties. Intentionality was similar in the two regions across all time periods with 20 (54%) of inflicted injuries in the severe counties and 108 (51.9%) of inflicted injuries in the less-affected/not-affected counties (p =0.81). In the severe counties, the

Discussion

These data demonstrate a county-level temporal and geographic increase in the rate of inflicted TBI following a natural disaster. While the rate of inflicted injuries rose and then returned to baseline, the rate of non-inflicted injuries rose even more dramatically and remained moderately above baseline at >6 months post-flooding, possibly reflecting increased injury risk due to prolonged stress.

These study results are consistent with the results found after Hurricane Hugo and the Loma Prieta

Acknowledgements

This project was funded by a faculty small grant from the University of North Carolina Injury Prevention Research Center, National Center for Injury Prevention and Control (grant R49/CCR402444).

References (14)

There are more references available in the full text version of this article.

Cited by (114)

View all citing articles on Scopus

The full text of this article is available via AJPM Online at www.ajpm-online.net.

View full text