But the authors warn that, while the figures are encouraging, there is no room for complacency because at least one child or young person still dies every week as a result of assault.

The public inquiries following the deaths of Victoria Climbié in 2000, Holly Wells and Jessica Chapman in 2002 and Peter Connelly in 2007 were critical of the ability of child protection services to actually protect children.

The authors used national mortality statistics from the Office of National Statistics showing cause of death and Home Office crime statistics reports to calculate rates of violent death for different age groups.

The figures showed that annual rates of death due to assault fell dramatically in children between 1974 and 2008. There was an almost sevenfold fall in violent deaths in infants from 5.6 to 0.7 per 100 000, and a threefold fall in violent deaths in children aged one to 14 years from 0.6 to 0.2 per 100 000.

The picture in adolescents is less encouraging. During the 1970s rates of death from assault fell among adolescents. They have since remained static in girls in these age groups, but have risen in boys.

When the authors combined the number of deaths from assault with those where it could not be determined whether injury had been caused by violent intent, they estimated that between five and 15 infants in England and Wales died a violent death every year, between 15 and 45 children aged 1-14 years and between 32 and 117 adolescents aged 15-19 years.

The authors suggest that variations in falls in violent death rates according to age, with the largest reductions in infancy, smaller reductions in the middle childhood years, and no change in adolescence, might reflect the different causes of violent deaths in different age groups.

"In infancy and early childhood, violent deaths primarily occur in the context of the family, with parents the usual perpetrators. As children grow, the risks from those outside the immediate family increase, and in adolescence, it is likely that most violent deaths are perpetrated by extra-familial assailants," they say. "This may suggest that policies around protecting children from abuse and neglect within the family are having some effect, while those aimed at protecting older youths from violence have so far been unsuccessful."

The authors conclude: "These reductions are unlikely to be accounted for by changes in categorisation but appear to reflect real improvements in protecting children from severe abuse."

Note: If no author is given, the source is cited instead.

• Children's Health
• Infant's Health
• Teen Health
• Healthy Aging
• Epilepsy Research
• Staying Healthy

• Child
• Sudden infant death syndrome
• Adolescence
• Infant
• Bone age
• Feral child