Defining suspected child maltreatment using International Classification of Diseases codes: A scoping literature review

Abstract

BackgroundAdministrative medical records and claims are increasingly used to study prevalence of and outcomes for children exposed to child maltreatment. However, suspicion of child maltreatment is often under-documented in medical records using International Classification of Diseases (ICD) codes. Although researchers have developed strategies to more broadly capture the injuries, illnesses, and circumstances that are suggestive of maltreatment, there is no consensus on which codes to use for this purpose. ObjectiveTo systematically examine the types of research being conducted with ICD codes related to suspected maltreatment, summarize ICD codes for suspected maltreatment, synthesize the strategies researchers employed to identify the codes, and propose future direction. MethodsWe searched five electronic databases for studies that reported ICD codes suggestive of child maltreatment in any medical setting, included participants aged 0–18 years, and published in a peer-reviewed journal in English. Two reviewers independently screened the titles, abstracts, and the full texts. Data were synthesized in a qualitative manner. This review was guided by the PRISMA extension for scoping reviews. ResultsThirty-seven studies met inclusion criteria. Most studies focused on determining the incidence or trends of maltreatment-related injuries or illnesses. Studies varied greatly in the codes used to identify suspected maltreatment. Only four articles reported on the validation of selected codes. ICD codes for transport accidents were the most frequently excluded co-occurring codes. Studies frequently relied on two seminal studies and two national-level guidelines. ConclusionsSubstantial heterogeneity existed in the ICD codes and strategies used to identify suspected maltreatment. The age range for codes, excluded co-occurring codes, and validation of codes are areas future research should address. This review may reduce costs for future researchers so that they could choose optimal measures of suspected maltreatment from our summarized list of codes without reinventing the wheel. Our review provides a basis for the development of recommended guidelines in establishing uniform codes for suspected maltreatment that could promote public health surveillance and allow for more efficient and uniform policy or program evaluation.