Illustration(s) pertain to the topic addressed in this publication, not the specific research or data presented in the publication

Objective birth data and the prediction of child abuse


Data from the USA suggest that many cases of abuse are not reported to child abuse agencies.  It is highly likely that the population characteristics associated with risk of abuse may actually represent risk of abuse coming to the attention of the authorities. Since the lower social classes are likely to be the beneficiaries of this increased risk of being reported, the factors known to be associated with low social class-such as marital instability, poor prenatal attendance, and low birthweight, etc.-will be artefactually found to be associated with risk of abuse. Furthermore, random selection of controls and objective control data will not correct this bias. Until we can be confident that all cases of child abuse are diagnosed and reported or that all families are under equal surveillance, identification of risk factors may merely represent risk factors for child abuse reporting and not for child abuse itself.

Marks JS, Jason J

Arch. Dis. Child. 1982 Jan;57(1):80-1 (Letter to the Editors)

PMID: 7065702