283 Improving US national data on child maltreatment fatalities. (1st September 2016)

• Record Type: Journal Article
• Authors: Schnitzer, Patricia; Anderson, Robert; Minino, Arialdi; Covington, Theresa
• Is Part Of: Injury prevention. Volume 22(2016)Supplement 2
• Page Start: A104
• Page End: A104
• Publication Date: 2016-09-01
• Subjects: Child Maltreatment -- data linkage -- child injury; Children's accidents -- Prevention -- Periodicals; Accidents -- Prevention -- Periodicals; 617.1
• DOI: 10.1136/injuryprev-2016-042156.283
• Languages: English
• ISSNs: 1353-8047

Abstract

Abstract : Background: In the United States (US), national data on child maltreatment (CM) deaths are compiled from state child welfare agencies; approximaly 1600 CM deaths are reported for all 50 states annually. Unfortunately, these child welfare data grossly undercount fatal CM in the US. This project used data from death certificates and child death review (CDR) programs to improve identification and better estimate fatal CM in the US. Methods: CDR data from 9 states with a high proportion of deaths reviewed and recorded in the National CDR Case Reporting System were linked to death certificate/mortality data submitted to the US National Centre for Health Statistics (NCHS). Case inclusion criteria were: children between 7 days and 5 years old that died during 2009 or 2010. A CM classification algorithm that included categories of presumptive, probable, possible CM was developed and applied to the linked deaths. Weights were calculated for each cause of death in each CM category; national CM mortality was estimated by applying these weights to 2011 NCHS mortality data for all 50 states. Results: A total of 10927 deaths meeting the inclusion criteria were identified. Of these, 6660 deaths linked to a record in the NCHS mortality file; 169 additional deaths identified by CDR did not link to a NCHS file record. The NCHS mortality file contained an additional 4098 deaths that did not have a CDR review. A total of 3606 deaths met the definition for presumptive (n = 907), probable (n = 387), or possible (n = 2312) CM in these 9 states over the 2 study years. This resulted in a national estimate of 1279 presumptive, 500 probable, and 3083 possible CM deaths for a total of 4862 CM deaths in the US during 2011. Conclusions: It is possible to improve upon the current child welfare-based system of counting fatal child maltreatment in the US by linking CDR and mortality data. In this study, the CDR Program data were essential for confirming child maltreatment in many of the deaths identified.