NIH Press Release
NATIONAL INSTITUTES OF HEALTH
National Institute of Child Health
and Human Development

EMBARGOED FOR RELEASE
Wednesday, October 21, 1998
5 PM Eastern Time
Robert Bock
(301) 496-5133

Researchers Identify Risk Factors For Infants Most Likely to be Homicide Victims

An infant's chances of becoming a homicide victim during the first year of life are greatest if he or she is the second or later born child of a teenage mother, according to an analysis of birth and death certificates by researchers at the National Institute of Child Health and Human Development (NICHD). Homicide is the leading cause of infant death due to injury.

In the study appearing in the October 22 issue of The New England Journal of Medicine, the authors also found that the likelihood of being killed was greatest for infants whose mothers were less than 15 years old, had less than 12 years of school, or did not have prenatal care. One half of the infants killed were dead by the fourth month of life.

The authors noted, however, that other studies of nonfatal child abuse suggest that a program to have home nurses visit expectant teenage mothers regularly could reduce the infant homicide rate.

"To a large extent, very young teens aren't ready to be either pregnant or parents," said the leader of the NICHD research team, Mary Overpeck, DrPH, a researcher with NICHD's Division of Epidemiology, Statistics, and Prevention Research. "Since these needless, tragic deaths occur early in a child's life, the key to preventing them is to reach the mothers early in pregnancy, before the child is born."

Other NICHD members of the research team were Ruth Brenner, MD, MPH, Ann C. Trumble, PhD, Lara B. Trifiletti, MA, and Heinz W. Berendes, MD, MPH.

Dr. Overpeck added that visiting home nurse programs that have reduced the incidence of child abuse have focused on helping teenage mothers gain behavioral skills and support.

"Most of these girls don't feel as if they're in control of their environment," Dr. Overpeck said. "The visiting health care professionals can help the girls develop their options--to finish school and take care of themselves and their babies."

In the article, Dr. Overpeck and her coauthors noted that a study of the most successful intervention to prevent child abuse was conducted in Elmira, New York, by researcher David Olds.

Dr. Overpeck explained that the intervention strategies used in the Olds study probably would reduce the infant homicide rate as well. Earlier studies, she said, have shown that more than 80 percent of infant homicides are due to abuse. The Olds study, of low-income, primarily white, unmarried, pregnant teenage girls, found that visits by trained nurses during the girls' pregnancies and the first two years of the children's lives reduced the incidence of child abuse and neglect among first born children. The girls who took part in the program also had fewer subsequent child births and were more likely to complete their education than were teenage mothers who did not take part in the study. Dr. Overpeck said this finding was extremely important, as the risk of infant homicide is higher for second or later born children of teenage mothers. A follow-up study of these girls, conducted 15 years later, showed that they were also less likely to abuse alcohol and drugs. She stressed, too, that since most infant homicides are committed by fathers or other males who are left to care for the infants, the resultant educational and behavior modifications learned by the girls may allow the infants to be cared for in a safer environment.

She added that 8 states account for about 50 percent of the total births to girls below the age of 17: California, Texas, Florida, Georgia, Illinois, New York, Ohio, and Michigan. Visiting nurse programs targeted to these areas would probably reduce the infant homicide toll significantly.

In The New England Journal of Medicine article, the researchers compared birth certificate data from all 34,895,000 births that occurred in the U.S. from 1983 to 1991 to data from death certificates completed during the same period. After analyzing the data, they determined that 2,776 infants died from homicide. An additional 52 deaths, caused by neglect, abandonment, or exposure to severe weather (but not considered intentional) were not included in the analysis.

Over all, second or later children born to mothers younger than 19 were 10 times more likely to be killed than the first child of the oldest mothers. The risk of death to children born to mothers with less than 12 years of education was 8.4 times that of children born to mothers who had completed 16 years of education. The researchers added, however, that it was difficult to separate the risk of death to infants born to mothers younger than 17 from the risk to infants born to mothers with less than 12 years of education, because many mothers under 17 years of age have not had the time to complete 12 years of education.

Still, when the researchers excluded girls younger than 17 from their analysis, the risk of homicide among children born to mothers who were old enough to complete 12 years of education but had not done so was still greater than for mothers who had completed additional years of school. In fact, children whose mothers were older than 17 but had not completed 12 years of school had 8 times the risk of death as did infants born to mothers who had 16 years of education.

The researchers also found that 1 out of 4 homicides occurred by the second month of life, one half occurred by the fourth month, and two-thirds by the sixth month. The investigators noted that they were able to obtain little information from death certificates about the circumstances surrounding the death. For example, only 10 percent of the available certificates listed the relationship of the infant to the person identified with killing him or her. One third of the certificates listed the cause of death as "battering or other maltreatment." About 28 percent listed the cause of death as "assault from unspecified means."

The authors noted, however, that other studies have reported that most infant homicides are carried out by either parents or stepparents, and a slight majority are carried out by males. Other studies have also found that most homicides of children older than two years of age are carried out by someone who is unrelated to the child.

Still other studies, the authors added, have found that when the child is killed during the first week of life, the homicide was usually conducted by the mother. The authors also reported that 5 percent of infant homicides occurred during the first day of life; of these, 95 percent were not born in a hospital. One explanation for these cases, the authors wrote, is that the mother committed the homicide to hide the pregnancy and birth.

The researchers also noted, however, that infant homicides are probably under reported. For example, some homicides committed on the day of the child's birth may be so well hidden that they simply go undetected. In addition, other homicides may have been attributed to death due to accidental injury or some other cause. In fact, other studies have shown that from 7 to 27 percent of deaths attributed to unintentional injuries actually may have been due to child abuse or neglect.

To remedy this situation, the researchers recommended that reviews of child deaths be conducted not just from death certificates and records of medical examiners, but also abuse registries, crime reports, hospitals, and ambulatory care records. The researchers added that one such review of the records from several agencies indicated that the number of deaths from abuse and neglect for children up to four years of age were double the rate reported by medical examiners' records.

The researchers also uncovered a disturbing trend: the number of homicides increased from 7.2 for every 100,000 births from 1983 to 1987, to 8.8 homicides for every 100,000 births from 1988 to 1991. The authors added that this increase is probably not due to better detection and reporting of infant homicide cases and most probably represents an increase in infant homicides. In fact, from 1988 through 1991, an average of 351 infant homicides were committed each year--almost one each day. Based on studies of the under reporting of infant homicides, the authors surmise that about twice the amount of infant homicides actually may have been committed during the same time period.

Dr. Overpeck said that no studies have been conducted to determine why the infant homicide rate has risen. She theorized, however, that the increase coincides with an increasing need for mothers to enter the work force, combined with a shortage of affordable child care for infants.

The NICHD is one of the Institutes of the National Institutes of Health, the world's premier biomedical research organization, located in Bethesda, Maryland. NICHD supports and conducts basic, clinical, and epidemiological research on the reproductive, developmental, and behavioral processes that determine and maintain the health of children, adults, families, and populations.