Sample Review
Accidents

Annual risk of death resulting from short falls among young children: Less than 1 in 1 million. David L Chadwick, Gina Bertocci, Edward Castillo, Lori Frasier, Elisabeth Guenther, Karen Hansen, Bruce Herman, Henry Krous. Pediatrics 2008;121:1213-1224. (from San Diego CA, Louisville KY, and Salt Lake City UT)

Review from Winter 2009 issue

The objective of this research was to develop an estimate of the risk of death resulting from short falls of less than 1.5 meters in vertical height, affecting infants and young children between birth and the fifth birthday. The study sought to review critically the pediatric fall literature and three relevant public injury databases in an effort to approximate the incidence of child deaths resulting from short falls in the population of young children.

For purposes of this review, a short fall was defined as a fall of less than 1.5 meters (less than 4.7 feet) in vertical height, not involving horizontal velocity beyond what a young child can achieve alone. Fall height was best defined as the difference in the positions of the child’s center of gravity at the start and end of the fall. The authors concede that consistent definitions were not found in the literature review, so the articles they reviewed did not necessarily conform to this definition. The definition of a fatal case they used was one in which the injury was the proximate cause of death.

The study set was derived from several sources:

1. Eight hundred literature citations were initially recovered and, of these, 188 were selected for review by more than one of the authors.  An analysis was written for each case. Excluded articles included those that dealt primarily with injuries among adults or older children involving sports activities or those that were transport-related injuries.
2. The authors used data from the State of California Epidemiology and Prevention for Injury Control Branch (EPIC) database, which uses discharge data submitted by all California hospitals and death certificate data submitted by all California county medical examiners. Death certificates for the 13 possible short fall fatalities were reviewed. Of these, only six remained as possible short fall deaths. This results in an incidence of 0.48 cases per 1 million children per year.
3. Web-based Injury Statistics Query and Reporting System of the Centers for Disease Control and Prevention allows determination of the total fall death rate for children under four years of age but does not provide stratification according to fall type/height. The “all fall” death rate was three cases per one million young children per year.
4. The National Electronic Injury Surveillance System (NEISS) of the Consumer Product Safety Commission uses active surveillance of a sample of hospital emergency departments. Extrapolated data from this source indicated that there would be 0.625 fatalities per one million children per year. This data set has several limitations, however, not the least of which is specialized selectivity of cases since the data are compiled by coders using a data dictionary that provides over 800 product-related elements but few or no elements that allow for the inclusion of cases with false fall histories. The result is that the data set captures almost all product-related injuries but is not adapted for violence-related data acquisition and, therefore, may fail to capture deaths related to short falls that are not involved with products.

5. The medical literature is meticulously categorized by the aims and the parameters of the studies. Although these varied widely, they are carefully identified in the paper. Included were the following:
   1. Studies of multiple and reliably witnessed falls
   2. Studies of child care-related injuries
   3. Studies using biomechanical analyses
   4. Studies of large clinical populations
   5. Studies comparing abusive and unintentional injuries
   6. Studies of long falls
   7. Articles addressing short-fall deaths specifically
   8. Pathologic and cadaver studies
   9. Studies focusing on fractures
   10. Population-based clinical studies of injuries and falls
   11. Studies of injuries occurring on playgrounds
   12. Falls down stairs
   13. Parents’ observations of infants’ falls
   14. Walker-related falls
   15. Cervical spine and cord injuries
   16. Single case and small series (less than 50 cases)
   17. Observational studies of children’s behavior

In their discussion, the authors reiterate the difficulty involved in determining the likelihood of death resulting from a short fall in a young child. That difficulty derives from the impossibility in determining the accuracy of the history of the incident. However, the authors opine that the “best solution to this problem is the ongoing collection of data on hospitalized and fatal injuries from large numbers of injury events affecting infants and young children in large states.”

The authors conclude that the best estimate for risk of death from short falls in an infant or a young child is less than one in a million young children per year. No evidence can be found in the rich literature on children’s fall-related injuries or in other public databases to support a higher incidence.

    Reviewed by Robert M. Reece, M.D.