Methods. This analysis was based on unique dataset constructed by linking child welfare records to birth records from California. Included are all children born in 2002 and reported for maltreatment before the age of five. Unadjusted and adjusted multinomial logistic regression models were used to estimate the association between reporter type and report disposition (evaluated out, unfounded, substantiated). Several models were specified with reporters grouped along different dimensions (e.g., mandated vs. non-mandated reporters; family reporter, medical reporter, legal reporter, community reporter, other professional, and anonymous reporter). Multivariate models included child covariates of race, age, gender, and a prior allegation of maltreatment. Adjustments were also made for family-level sociodemographic characteristics captured as maternal education, birth place, and age, as well as birth payment method and the establishment of paternity on the birth record. Models were stratified by allegation type, with predicted probabilities of a substantiated disposition computed.
Results. Of the over 530,000 children born in 2002, 14% were reported for maltreatment before the age of five. Approximately 21% of these children were substantiated as victims of maltreatment, while 64% received an investigation that was unfounded, and another 15% were evaluated out over the phone. Nearly 20% of children had a first report with no identified reporter, followed by a report originating from law enforcement (18%), other professionals (15%), a medical provider (13%) and schools or child care providers (11%). Reports arising from law enforcement consistently had the highest substantiation rate. Reports from family members were more frequently evaluated out than reports from other sources. In both crude and adjusted models, the odds of report substantiation for first allegations of maltreatment did not vary by mandated reporter status. When follow-up allegations were modeled, however, reports made by mandated reporters were significantly more likely to be substantiated. Allegations of maltreatment for African-American and Latino children were less likely to be substantiated than white children, after adjusting for reporter type, maltreatment type, and child/family-level sociodemographic characteristics.
Conclusions and Implications. This analysis examines the association between reporter identity and the subsequent disposition assigned to an allegation of maltreatment. Findings suggest that mandatory reporter status is predictive of substantiation among later, but not first allegations of maltreatment. Aligning with prior research, findings support variations in allegations made by different types of reporters.
