Counting injury deaths: a comparison of two definitions and two countries. (17th December 2013)

• Record Type: Journal Article
• Authors: Lu, Tsung-Hsueh; Hsiao, Annjhih; Chang, Pei-Ching; Chao, Yao-Cheng; Hsu, Chuan-Chiang; Peng, Hua-Chun; Chen, Lea-Hua; Kawachi, Ichiro
• Is Part Of: Injury prevention. Volume 21(2015)e1
• Page Start: e127
• Page End: e132
• Publication Date: 2013-12-17
• Subjects: Children's accidents -- Prevention -- Periodicals; Accidents -- Prevention -- Periodicals; 617.1
• DOI: 10.1136/injuryprev-2013-040974
• Languages: English
• ISSNs: 1353-8047

Abstract

Abstract : Objectives: This study sought to examine the discrepancy in counting injury deaths according to two definitions, that is, (a) external cause of injury as the underlying cause of death (UCOD), and (b) with mention of external cause of injury, and differences in certification practices between Taiwan and the USA. Methods: We calculated the percentage (U/M%) in which external cause of injury was selected as the UCOD (U) among deaths with mention (M) of external cause of injury anywhere on the death certificate to assess the discrepancy between the two definitions in counting injury deaths. Results: The discrepancy was small in Taiwan and in the USA for transport injuries (with U/M% 99% and 98%) and intentional self-harm (100% and 100%) and large for accidental suffocation (64% and 29%) as well as complications of medical and surgical care (61% and 10%). The magnitude of discrepancy was not consistent between countries in accidental falls (96% vs 81%) and accidental drowning (80% vs 97%). Certifiers in Taiwan were more likely to report an accidental fall in Part I of the death certificate and less likely to report medical conditions (such as stroke, Alzheimer's disease or Parkinson's disease, etc) with accidental suffocation than their counterpart US certifiers. Conclusions: The counting of injury deaths according to two definitions varied by external cause of injury as well as between countries. For some external causes of injury with high discrepancies, we suggest presenting the number of injury deaths according to two definitions.