How many child deaths are avoidable?

"Around two-thirds of child deaths caused by abuse could have been avoided last year, campaigners have warned as they accused the Government of failing to protect the rights of children."

The Daily Star, January 21, 2012

Out of 43 children who died in England in 2011-2012 as a result of "deliberately inflicted injury, abuse or neglect," over half could have been saved, a report by the Children's Rights Alliance for England (CRAE) has found.

The report stated that 28 child deaths were deemed to have "modifiable factors." 

But what does "modifiable" mean in this context? Let's first look at the source for this figure. 

The data reported by CRAE is from a July 2012 Department for Education Statistical Release on Child Death Reviews. Between 1 April 2011 and 31 March 2012, the Local Safeguarding Children Boards in England reviewed the deaths of all children excluding stillborn babies, from birth up to 18 years, who resided within their area. This is known as the Child Death Review Process.    The child death overview panels analysed 4,012 child deaths and found that 784 of these had "modifiable factors" (20%). The category of death which had the largest proportion (65%) of cases identified as preventable were those due to "deliberately inflicted injury, abuse or neglect." The table below illustrates this, and other types of deaths which have been deemed "preventable or potentially preventable".     So what does "modifiable death" mean?   The report defines it as:   "a death where there are factors which may have contributed to the death. These factors are defined as those which, by means of nationally or locally achievable interventions, could be modified to reduce the risk of future child deaths."    This is somewhat vague, how do you identify factors on which local or national interventions could impact? One way is to look at the difference of incidence of these sort of deaths on older versus younger children, and on male versus female children.   For example, the report explains that some kinds deaths, like those by suicide and road traffic accidents, occur more frequently in older children.    Aside from suicides and road accidents, other examples of death which were classified as preventable included drowning, falls (for example falls from balconies while abroad), various traumas, and even unexplained deaths like sudden infant death syndrome.    Conclusion It looks like the original claim for this factcheck ("around two-thirds of child deaths caused by abuse could have been avoided last year") is based on sound evidence collected by the government. However, the report is quick to point out that even if a death is identified as "modifiable," it doesn't necessarily follow that it was caused by a failure of the local authority. The responsibility can also be attributed to "any agency, including parents, latent, organisational, systemic or other indirect failure(s) within one or more agency."   Furthermore, it's important to bear in mind that it doesn't necessarily follow that all 784 "modifiable" deaths were preventable. The researchers themselves admitted it was difficult to draw a distinction between those deaths which were preventable and those which were "potentially preventable." In conclusion, we shouldn't be too simplistic in our reading of these headlines.