Agency notification and retrospective self-reports of childhood maltreatment in a 30-Year cohort: Estimating population prevalence from different data sources.

Abstract Background There is little known about how two sources of child maltreatment reports correspond, specifically for emotional, physical, sexual abuse and emotional or physical neglect. Objective To compare agency and self-reports of child maltreatment and determine whether they are predicted by similar early life course adverse experiences. Participants Data are taken from a long running birth cohort study (Mater-University of Queensland Study of Pregnancy - MUSP). Mothers (N=7223) were recruited in 1981-83 and their children were followed-up 30 years later (2010-14). In 2000 data from the relevant child protection agency were accessed and linked to the survey data. Setting Consecutive women giving birth to a live singleton baby at a major obstetrical service in Brisbane, Australia were recruited and both mother and child were repeatedly follow-up over a 30 year period. Methods Birth cohort study with data linkage of child protection agency records and self-report survey data (using the Childhood Trauma Questionnaire - CTQ) of childhood maltreatment experiences. We compare reports of emotional, physical and sexual abuse and neglect using agency and self-reports (Cramer's V and kappa). Results Most children who are notified cases of maltreatment subsequently self-report they experienced little or no maltreatment in childhood. Most children who report experiencing severe maltreatment have not previously been notified to the protection agency. Teenage mothers have children who are notified 2.89 (1.52, 5.52) times, self-report 2.01 (1.31, 3.09) times and both notified and self-report 3.61 (2.26, 6.10) times more than their older comparison mothers. Conclusions Different methods of assessing maltreatment identify different subsets of those children who have experienced maltreatment.