Elizabeth Dawes Knight

Concordance Between Adolescent Reports of Childhood Abuse and Child Protective Service Determinations in an At-Risk Sample of Young Adolescents

This study examines the concordance between adolescent reports of abuse and abuse determinations from Child Protective Service (CPS) agencies. It also compares the utility of adolescent reports of abuse, relative to CPS determinations in predicting adolescent psychological adjustment. The sample included 350 early adolescents, ages 12 to 13 years, who were initially identified prior to age 2 years as being at elevated risk of maltreatment. An Audio-Computer Assisted Self Interview (A-CASI) was used to assess lifetime experiences of physical, sexual, and psychological abuse. The A-CASI interview elicited prevalence rates of abuse 4 to 6 times higher than those found in CPS records. However, 20 of 45 adolescents with CPS determinations of abuse failed to report abuse during the study interview. Adolescent psychological adjustment was more strongly associated with self-reports than with CPS determinations. The implications of these findings are discussed for validity of adolescent self-reports of childhood abuse and for the ongoing debate about disclosure patterns among victims of child sexual abuse.

Youth self-report of physical and sexual abuse: A latent class analysis

OBJECTIVE To determine if meaningful groups of at-risk pre-adolescent youth could be identified based on their self-report of physical and sexual abuse histories. METHODS Youth participating in a consortium of ongoing longitudinal studies were interviewed using an audio-computer assisted self-interview (A-CASI) when they were approximately 12 years of age to obtain information about their perceived experiences of physical (18 items) and sexual (12 items) abuse. In addition, Child Protective Service records were reviewed and the taxonomy developed for defining maltreatment characteristics (Barnett, Manly, & Cicchetti, 1993) was applied. A total of 795 youth completed the age 12 interview and had their records reviewed during the period from birth to the time of their age 12 interview. A latent variable modeling approach, specifically latent class analysis (LCA), was used to generate profiles of youth based on their endorsements of the physical and sexual abuse items. These profiles were then compared to CPS reports of physical or sexual abuse to determine their validity. RESULTS The LCA identified 4 interpretable classes or groups of pre-adolescent youth. Based on the pattern of responses to specific items the classes were identified as follows: (1) no physical or sexual abuse; (2) high physical abuse/low sexual abuse; (3) no physical abuse/moderate sexual abuse; and (4) high physical and sexual abuse. Follow-up analyses indicated that the odds of a CPS report for Classes 2, 3, and 4 compared to Class 1 were significantly greater (2.21, 2.55, and 5.10, respectively). CONCLUSION The latent variable modeling approach allowed for the identification of meaningful groups of youth that accounted for both the occurrence of multiple types of abuse as well as differing severities associated with each type. It is suggested that this methodological approach may be most useful in future efforts to identify the antecedents and consequences of maltreatment. PRACTICE IMPLICATIONS The results of the present study not only have implications for future research efforts, but also suggest that in practice, youth at-risk for maltreatment may be reliable and valid reporters of their physical and sexual abuse experiences.

Sexual intercourse among adolescents maltreated before age 12: A prospective investigation

OBJECTIVE: To examine whether child maltreatment (physical, emotional, and sexual abuse, and neglect) predicts adolescent sexual intercourse; whether associations between maltreatment and sexual intercourse are explained by childrens emotional distress, and whether relations among maltreatment, emotional distress, and sexual intercourse differ according to gender. METHODS: The Longitudinal Studies of Child Abuse and Neglect was a multisite, longitudinal investigation. Participants ranged from at-risk to substantiated maltreatment. Maltreatment history was assessed through Child Protective Service records and youth self-report at age 12. Youth reported emotional distress by using the Trauma Symptom Checklist at the age of 12 years and sexual intercourse at ages 14 and 16. Logistic and multiple regressions, adjusting for gender, race, and site, were used to test whether maltreatment predicts sexual intercourse, the explanatory effects of emotional distress, and gender differences. RESULTS: At ages 14 and 16, maltreatment rates were 79% and 81%, respectively, and sexual initiation rates were 21% and 51%. Maltreatment (all types) significantly predicted sexual intercourse. Maltreated youth reported significantly more emotional distress than non-maltreated youth; emotional distress mediated the relationship between maltreatment and intercourse by 14, but not 16. At 14, boys reported higher rates of sexual intercourse than girls and the association between physical abuse and sexual intercourse was not significant for boys. CONCLUSIONS: Maltreatment (regardless of type) predicts sexual intercourse by 14 and 16. Emotional distress explains the relationship by 14. By 16, other factors likely contribute to intercourse. Maltreated children are at risk for early initiation of sexual intercourse and sexually active adolescents should be evaluated for possible maltreatment.

Methodological and Ethical Challenges Associated With Child Self-Report of Maltreatment Solutions Implemented by the LongSCAN Consortium

The conduct of research in the area of child abuse and neglect may be one of the most difficult tasks in social science research. One requirement for valid research is knowledge of the type and amount of exposure. Official reports have been demonstrated to provide a serious undercount of the frequency of maltreatment, and parent report is of limited usefulness. LongSCAN, a consortium of longitudinal studies of abuse and neglect, made the decision to ask children for self-report, but with five independent study sites with unique study-to-sample relationships, ethically implementing this choice demanded customized participant protocols. This article describes the consortiums approach to asking children for direct reports at age 12, the relevant methodological and ethical challenges, and solutions developed with institutional review boards at 4 of the 5 study sites. The wording of consents and the variations in protocol related to reporting to Child Protective Services are discussed.

Reporting participants in research studies to Child Protective Services: limited risk to attrition

This study examines the impact of Child Protective Services (CPS) reports made by research study staff on participant retention and discusses human subjects protocols that may minimize either the need to make such reports or the negative impact of reporting on participants and on participant retention. Among 1, 354 primary caregiver-child pairs in the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN) studies, a total of 15 were reported to CPS by study staff. Within this group, rates of study-generated reports and study participation subsequent to having been reported by researchers were examined. There was an overall retention rate of 93% across a minimum of three interview waves in this sample. Reporting research participants to CPS may have little impact on attrition.