RESUMO
Intake reports from child protective service (CPS) agencies are the foundation for subsequent decisions and services within the child welfare system. They provide valuable information for understanding children's needs, yet the unstructured way that information is captured makes the information ascertained in these reports difficult to summarize. Utilizing CPS intake reports from a mid-sized urban county in the southeastern United States (N = 2,724), our study had three aims: (1) develop a coding system to abstract information from narrative CPS intake reports, (2) examine the prevalence of maltreatment subtypes, and (3) compare prevalence of maltreatment subtypes by screen-in/screen-out decisions. Improper discipline/physical abuse was the most common maltreatment subtype (34.6 %); over 40 % of reports involved a physical act toward the child not resulting in injury. Salient risk factors included caregiver drug use (20.6 %) and domestic violence (19 %). While substantial discrepancies were not found between screened-in and screened-out cases with respect to maltreatment types, maltreatment type-specific codes, or contributory factors, they were found for reporter type and child age. Our coding system to extract and assess child maltreatment intake narrative data can provide local agencies with descriptive information about why children come to their agency's attention and provide nuanced details regarding the children's and families' needs. This coding framework could be used to develop validated intake tools to better document and categorize child maltreatment which could inform the assessment/investigation process and create targeted prevention and intervention services for families that come to the attention of CPS.