Longitudinal trends in child protection statistics in South Australia: A study of unit record data

Abstract

In Australia, it is commonly reported that rates of child protection notifications have increased over time. More and more children in any given year are subject to a child protection notification. On the whole, these conclusions have been based on cross-sectional notification counts or rates recorded in a given year (e.g. AIHW 2009). Although useful, such analyses are limited in that they do not account for the fact that child protection incidents are unevenly distributed across individual cases. Crosssectional analyses also do not indicate the incidence of notifications within a given cohort of children.In this paper, we summarise the longitudinal and comparative analysis of data relating to children born in 1991, 1998 and 2002. The results highlight the increasingly early involvement of child protection systems in children's lives, higher annual incidence rates, as well as increasingly steep cumulative involvement curves for cohorts tracked from their year of birth. The implications of these findings for mandatory reporting policies are discussed.