OBJECTIVES: Strong evidence based on case record reviews indicates that the incidence of child homicide reported from death certificates is under ascertained. The characteristics of infant injury fatalities with undetermined, but suspicious, intent were compared for the probability that they should be considered homicides. METHODS: Using linked birth and death certificates for all birth cohorts in the US from 1983-91, 2345 injury fatalities reported as intentional, 7594 as unintentional, and 431 as undetermined intent were identified. Maternal and infant variables potentially predictive of injury fatalities were selected based on increased bivariate associations. Relative risks of injury death by intentional, unintentional, and undetermined intent were assessed for maternal and infant characteristics. RESULTS: Relative risks were consistently higher across all intent categories for infants of mothers with the least education, no prenatal care, young maternal age, and single marital status, as well as for infants who are second or later born, preterm, black, or American Indian. Fatalities with undetermined intent have larger relative risks in the highest risk categories than either intentional or unintentional injuries. Deaths with undetermined intent have risk profiles that more closely resemble profiles for intentional deaths than unintentional. CONCLUSIONS: Injury homicide rates would be almost 20% greater than official classifications indicate if deaths with undetermined intent were included. In analyses of infant homicide, excluding deaths of undetermined intent may lead to an underestimation of the magnitude of the public health problem of intentional injuries among infants. Other studies based on record reviews from multiple sources indicate that misclassification and under ascertainment of homicides may be even greater.
OBJECTIVES: Strong evidence based on case record reviews indicates that the incidence of child homicide reported from death certificates is under ascertained. The characteristics of infantinjury fatalities with undetermined, but suspicious, intent were compared for the probability that they should be considered homicides. METHODS: Using linked birth and death certificates for all birth cohorts in the US from 1983-91, 2345 injury fatalities reported as intentional, 7594 as unintentional, and 431 as undetermined intent were identified. Maternal and infant variables potentially predictive of injury fatalities were selected based on increased bivariate associations. Relative risks of injury death by intentional, unintentional, and undetermined intent were assessed for maternal and infant characteristics. RESULTS: Relative risks were consistently higher across all intent categories for infants of mothers with the least education, no prenatal care, young maternal age, and single marital status, as well as for infants who are second or later born, preterm, black, or American Indian. Fatalities with undetermined intent have larger relative risks in the highest risk categories than either intentional or unintentional injuries. Deaths with undetermined intent have risk profiles that more closely resemble profiles for intentional deaths than unintentional. CONCLUSIONS: Injury homicide rates would be almost 20% greater than official classifications indicate if deaths with undetermined intent were included. In analyses of infant homicide, excluding deaths of undetermined intent may lead to an underestimation of the magnitude of the public health problem of intentional injuries among infants. Other studies based on record reviews from multiple sources indicate that misclassification and under ascertainment of homicides may be even greater.