Critical and fatal child maltreatment in Oregon: Escalating violence or distinct behavior

This study examines 924 cases involving children under the age of 6 from Oregon's public child welfare system to determine if 24 family/parental factors and 22 child characteristics used by front-line child protection workers are valuable in identifying the most severe maltreatment cases.;Results suggest that: (1) Specific factors are associated with mild, moderate, and severe cases. However, these factors have little value in predicting maltreatment severity. (2) Although the data do not demonstrate an association between children's disabilities and increased severity, 40 of the 42 fatal cases in the sample involved children with a documented physical, mental, or behavioral problem. (3) Abuse and neglect cases are two distinct populations. (4) Cases do not form an unbroken severity continuum. Cases in the mild group are the most distinguishable because they share the fewest factors with the other severity groups. (5) Fatal cases of child maltreatment have fewer of the factors commonly associated with maltreatment severity than any of the other groups in this study.;The findings of this study have broad implications for child welfare practice. The factors commonly used by child protection workers may not be useful in differentiating the most severe cases in the child welfare system. Given the increasing work load, the lack of reliable and valid measures of risk, and the incomplete information available to front-line child protection workers, it is unlikely that the child protection agency in Oregon can successfully predict which families will critically or fatally maltreat their children. A new approach to risk assessment, a broader review of fatal cases, more efficient sharing of case-related information and additional research efforts are suggested.;Using data from two previous studies, sample cases are assigned to four severity groups and then compared to one another on the basis of family factors, parental factors, and child characteristics. Findings include demographic data, frequency distributions, Z-scores, and tests for association.