Aims Accurate recording of child maltreatment in electronic hospital records is important for clinicians to identify children at risk of maltreatment and to inform service providers about the burden of maltreatment. The authors examined variation in the recording of child maltreatment and related diagnoses in administrative hospital data.

Methods Cross-sectional analyses of electronic hospital data (Hospital Episode Statistics, HES) and a national trauma registry (Trauma Audit Research Network, TARN). The authors analysed acute injury admissions to the NHS in England of children under 5 years of age between 1997 and 2008, and the annual incidence of ICD-10 codes specifying child maltreatment (as well as other maltreatment-related codes consistent with alert features referred to in clinical guidelines reflecting assault, adverse social circumstances, or undetermined cause). The authors also compared recording in severe injury admissions in TARN and equivalent admissions in HES.

Results 2.6% of injury admissions in infants, and 0.4 to 0.6% in older age groups, had maltreatment-specific codes in HES. This prevalence more than doubled when maltreatment-related codes were also included (6.4% in infants, 1.5–2.1% in older age groups). From 1997 to 2008, the annual incidence of admissions with maltreatment-specific codes declined in infants and in 1–3 year-olds, while admissions with maltreatment-related codes increased in all age groups. Maltreatment was recorded approximately 20% less often in equivalent admissions in HES than in TARN. Recording of maltreatment-specific and maltreatment-related codes varied substantially between Trusts, despite adjustment for deprivation. Children under the care of a paediatrician were more likely to have maltreatment codes recorded, but 23.7% of all acute injury admissions in infants, 47.2% for 1 to 3 year olds and 63.7% for 3 to 5 year age olds, had no record of a paediatric consultant episode during the admission.

Conclusion Child maltreatment is under-recorded in hospital administrative data. Despite a shift from maltreatment-specific to maltreatment-related codes, the overall burden has remained stable. In combination, maltreatment-specific and maltreatment-related codes identify children likely to meet thresholds for suspecting or considering maltreatment and taking further action, as recommended in recent the National Institute for Health and Clinical Excellence guidance, and indicate a considerable burden to which hospitals should respond.