CHAPEL HILL - Painstaking analysis of a decade of homicides among children ages 10 and under in North Carolina shows vital records poorly reflect the number of child abuse deaths, according to a new University of North Carolina at Chapel Hill study. The true number is more than three times higher than those records indicate.
Within homicide cases, most child abuse deaths in North Carolina could not be identified through computerized files, the study's principal investigator says. Records in other states likely are just as inaccurate since all states use the same international system of coding causes of death.
Researchers estimated that from 1985 through 1996, 9,467 homicides among U.S. children under age 11 resulted from abuse rather than the 2,973 reported, said Dr. Marcia E. Herman-Giddens, adjunct associate professor of maternal and child health at the UNC-CH School of Public Health.
"Using medical examiner data and extensive interviews with police and others, we found that significant under-reporting of child abuse homicides in vital records systems persists in spite of greater attention to abuse fatalities," Herman-Giddens said. "Improved recording of the incidence of these deaths should be a priority so that we can do a better job in monitoring and preventing them, especially given the rising abuse death rate. Deficiencies in recognizing abuse fatalities are due to the system and certain other causes, not lapses in vital records."
A report on the findings appears in the Aug. 4 issue of the Journal of the American Medical Association. Other authors are Dr. Gail Brown, former director the N.C. child fatality prevention team; Sarah Verbiest and Pamela J. Carlson, project managers for the study; Elizabeth G. Hooten, statistician in the Office of the Chief Medical Examiner; Eleanor Howell, statistician at the N.C. Center for Health Statistics; and Dr. John D. Butts, N.C. chief medical examiner and clinical professor of pathology and laboratory medicine at the UNC-CH School of Medicine.
Researchers found 259 N.C. children under age 11 were victims of homicides from 1985 through 1994. Of those, 220 -- or 85 percent -- were due to child abuse, 22 were not abuse-related and 17 could not be determined, according to a case-by-case review. Abuse deaths include children killed by caregivers and do not include deaths due to neglect.
Other findings were that:
- The N.C. rate of abuse homicides increased from 1.5 per 100,000 in 1985 to 2.8 in 1994, a rise of about 12 percent a year.
- N.C. black children were killed at three times the rate of white children.
- Males comprised two-thirds of probable killers.
- Almost two-thirds of assailants were one or both of the victim's biological parents.
"Further study to determine the reasons for the increasing fatal violence against children would help in determining where to direct prevention efforts," Herman-Giddens said. "Factors that appear to put African-American children at three times the risk of dying at the hands of caregivers than white children should be distinguished from the effects of social class and socioeconomic status."
The need is urgent to collect accurate data on child abuse homicides and to devise strategies for preventing them, she said.
"Knowing that 85 percent of the time child homicides result from abuse by caregivers should also help law enforcement officers know where to look first," the researcher said.
Herman-Giddens said her group's best estimate is that each year about 800 children are being killed in the United States by those most responsible for their care and safety. This does not count children dying from neglect.
"I think this is pretty incredible when you consider the to-do that arises every year around Christmas when a few children get hurt by some toy," she said. "Where is the outcry for these 800 children killed by parents or other caregivers?"
Support for the study came from the U.S. Department of Justice, the N.C. Department of Crime Control and Public Safety, the Governor's Crime Commission, the U.S. Centers for Disease Control and Prevention, the N.C. State Child Fatality Prevention Team and the Injury Prevention Research Center and School of Public Health's department of maternal and child health, both at UNC-CH.
Note: Herman-Giddens can be reached at 919-542-2529.
Journal
JAMA