Identification of Infant Victims of Abusive Head Trauma Hospitalised in France from 2015 to 2017

To our knowledge, there are no recent national epidemiological studies in France on abusive head trauma (AHT). The objective of this study was to quantify cases of AHT as a result of child abuse admitted to French hospitals, based on the French hospital discharge database (PMSI) from 2015 to 2017. Among infants aged between one and 11 months who were hospitalised for subdural haematoma or traumatic subarachnoid haemorrhage, we distinguished probable cases (stays with International Statistical Classification of Diseases and Related Health Problems, 10(th)Revision codes suggestive of child abuse or retinal haemorrhage) from possible cases (all other children), in order to make comparisons and estimate hospital incidence rates. We identified 512 probable cases and 703 possible cases. The incidence was 22.1 cases (20.2-24.0) per 100 000 live births for probable cases, and 52.4 cases (49.5-55.4) per 100 000 live births for combined probable and possible cases. The overall median age was four months with a male predominance. Among probable cases, lesions suggestive of abuse were more frequent, and there were fewer skull fractures and accidental falls than in possible cases. This exploratory analysis shows that it is complicated to determine precisely the number of cases of AHT from PMSI data. The implementation of a specific code for AHT in this database would identify these cases more accurately. Key Practitioner Messages Among probable cases, lesions suggestive of abuse were more frequent, and there were fewer skull fractures and accidental falls than among the possible cases. Implementation of a specific code for AHT in hospital discharge databases would identify these cases more accurately.