Shortly after he celebrated his first birthday, Dalyire Damion McFadden was fatally stabbed in the neck by his father, who wrapped his son's body in a deflated air mattress and left it in the basement of his mother's home. It was a shocking homicide even in Dalyire's hometown of Baltimore, a city that often lands at or near the top when homicide rates are calculated.

Despite making frequent headlines, child homicide is an extremely rare event. The murder of a child grabs public attention because it is one of the most unthinkable of crimes. When the homicide occurs at the hands of the person most expected to protect a child from harm—and it most often does—the horror and fascination are compounded, and so is the public's inability to comprehend it.

Children under age 5 are more likely to be killed by a parent than by any other person and are at greater risk of homicidal death than at any time in their life. Filicide, the killing of one's own child, can be maternal or fraternal, but mothers kill their children for reasons very different from those of fathers.

Neonaticide, the killing of a child within the first 24 hours of life, almost always occurs at the hands of the mother, often a young, unmarried woman with an unwanted pregnancy who received no prenatal care and who is not subsequently determined to have a mental disorder. Maternal mental illness is more likely to be implicated in the killing of children older than age 1.

In 1969, Phillip Resnick, M.D., a professor of psychiatry and director of the Division of Forensic Psychiatry at Case Western Reserve University School of Medicine, reviewed the psychiatric literature on maternal filicide and first characterized five major motives of filicidal mothers (see Five Major Motives of Maternal Filicide Perpetrators*). He found a high frequency of depression, psychosis, prior use of psychiatric services, and suicidality among the perpetrators of maternal filicide. "The challenge," he said, then "is the prediction of the likelihood of child homicide being perpetrated by a parent according to [his or her] psychiatric history."

In the May Journal of Clinical Psychology, researchers at Aarhus University in Denmark referred to Resnick's classification scheme and compared homicide rates between children whose parents were previously admitted to a psychiatric hospital and those whose parents had no previous psychiatric admission, as they examined the effect of specific parental diagnoses and assessed how frequently the crime was perpetrated by a parent or parents according to their psychiatric history.

Theirs was a prospective, register-based cohort study using the entire Danish population born from January 1, 1973, to January 1, 2007. Follow-up of the cohort members began on their date of birth and ended on January 1, 2007, their 18th birthday, their date of death, or their date of emigration, whichever came first. They also used police records from 2000 to 2005 to determine if one of the parents was the perpetrator.

In 88 percent of the Danish child-homicide cases examined, a parent was the perpetrator, regardless of whether that parent had ever been admitted to a psychiatric hospital.

But children of parents who had at some point been admitted to a psychiatric hospital "had a higher risk of being homicide victims, and risks were especially high in young children whose mothers were hospitalized with affective disorders or schizophrenia," wrote Thomas Laursen, Ph.D., an associate professor at National Center for Register-based Research at Aarhu University, and colleagues.

Are the motives paternal filicide the same? In the March 2009 Journal of Forensic Science, Resnick and colleagues analyzed the literature on paternal filicide finding that roughly half of filicidal acts are committed by fathers. Fathers who killed their children were, on average, in their mid-30s. The mean age of their victims was 5. Sons and daughters were killed in equal numbers. Reasons for the paternal filicides included death related to abuse of the child, mental illness of the father (including psychosis and depression), and revenge against a spouse. Suicide following the act occurred frequently.

In 1994, Martin Daly, Ph.D., a professor of psychology at McMaster University in Hamilton, Ontario, and his wife, Margo Wilson, Ph.D. (who died after the study was completed), considered that paternity is of interest in cases of paternal filicide because suspicions of nonpaternity may be germane to some cases, and the motivations in such cases may resemble more closely those of steppaternal homicides. Stepfathers, compared with genetic fathers, are more likely to perpetrate lethal abuse of their stepchild and never appear to commit suicide.

Researchers agreed that most aspects of filicide deserve further study, but Laursen and colleagues also emphasized the rare nature of such events: "Because the actual risks of homicide are extremely low, mental health professionals, policymakers, and the popular media must continue to emphasize that the overwhelming majority of mentally ill parents pose no threat to the survival of their children.... While more common health problems affect these children more frequently than in the general population, the threat of homicide is minimal."

The research of Laursen and colleagues was supported by the Stanley Medical Research Institute and the Wellcome Trust.

An abstract of "Filicide in Offspring of Parents With Severe Psychiatric Disorders: A Population-Based Cohort Study of Child Homicide" is posted at <www.ncbi.nlm.nih.gov/pubmed/21034682>.