87 Suicide and Life-Threatening Behavior 40(1) February 2010  2010 The American Association of Suicidology The Associations Between Infant Homicide, Homicide, and Suicide Rates: An Analysis of World Health Organization and Centers for Disease Control Statistics Matthew Large, FRANZCP, Olav Nielssen, FRANZCP, Steven Lackersteen, B Psych (Hons), and Glen Smith, FRANZCP Previous studies have found that rates of homicide of children aged under one (infant homicide) are associated with rates of suicide, but not with rates of homicide. Linear regression was used to examine associations among infant homi- cide, homicide, and suicide in samples of regions in the United States and other countries. Infant homicide rates were significantly and independently associated in all analyses with both total homicide and total suicide rates. The epidemiological evidence supports the clinical observation that infant homicides are due to a range of causes, including maternal mental illness, but are also related to the level of serious violence within a community. Within the cohort of child homicide, rates of killing of a newborn in the 24 hours after birth. It also includes infanticide, a legal term homicide of infants aged under one (here- after, infant homicides) are higher than for for the killing of an infant aged less than one by a mother suffering from a severe mental any other time in childhood. In many coun- tries, they exceed the rates of homicide of illness arising after childbirth. In a recent review of child homicide, adults (Lester, 1986; Marks & Kumar 1993). Infant homicide includes neonaticide, the Friedman and Resnick (2007) noted that a distinguishing feature of infant homicide is that rates of infanticide parallel suicide rates Dr. Matthew Large is with the School of rather than rates of other homicides. Five Psychiatry, University of New South Wales in previous studies have examined the associa- Randwick, Australia; Dr. Olav Nielssen and tion between infant homicide and homicide Steven Lackersteen are with the Clinical Re- or suicide rates. In two studies examining search Unit for Anxiety and Depression, School World Health Organization (WHO) mortal- of Psychiatry, UNSW at St Vincent’s Hospital in Sydney, Australia; Dr. Nielssen is also associated ity statistics for 1966 (Lester, 1974) and 1973 with the Discipline of Psychological Medicine at (Lester, 1991) infant homicide rates were the University of Sydney; and Dr. Glen Smith correlated with suicide rates, but not with to- is with the Northern Sydney Central Coast Area tal homicide rates. Two further studies, also Health Service, Macquarie Hospital, North Ryde, using WHO mortality statistics from 1974 Australia. We would like to thank Dr. Peter Arnold and 1976, respectively, found that rates of in- for his assistance in the preparation of the paper. fant homicide were not correlated with rates Address correspondence to Dr. Matthew of other homicides (Christoffel, Liu, & Large, The Prince of Wales Hospital, Mental Stamler, 1981; Fiala & LaFree, 1981). A fifth Health Services, 299 Barker St., Randwick, Syd- study, examining statistics from the United ney, NSW 2031 Australia; E-mail: mmbl@big pond.com States of America from 1985, found no cor-