3303.0 - Causes of Death, Australia, 2004  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 14/03/2006   
   Page tools: Print Print Page Print all pages in this productPrint All

APPENDIX 4 ENHANCED DEATH DATA


MULTIPLE CAUSES OF DEATH STATISTICS

Since 1997, the ABS has coded multiple causes of death. Prior to 1997, the ABS produced causes of death statistics where each death was assigned a single underlying cause. WHO has encouraged the use of multiple cause coding and analysis in relation to cause of death.


The underlying cause is defined as the 'disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury'. The medical certificate of cause of death recommended by WHO for international use, was designed to facilitate the selection of the underlying cause. When more than one condition is entered on the death certificate the underlying cause is selected using the coding rules of the relevant version of the ICD. The underlying cause is selected so that preventive strategies can be instituted to address that particular cause. Since its adoption in 1948, statistics based on the underlying cause concept have served the purpose of summarising international cause-specific mortality statistics into a single index, which has been used to assess trends in causes of death.


The current main causes of death are now very different from those that prevailed when the single underlying cause concept was adopted. The main causes of death in most countries have shifted from infectious and parasitic diseases to chronic and degenerative diseases, and from infant and child deaths to deaths among the elderly. As the population continues to age, chronic diseases become increasingly prevalent. Deaths from chronic conditions occur with a number of concurrent or coexisting conditions, whose initiating condition is often difficult to isolate. Some deaths cannot occur without the influence of more than one cause. The ability of the single underlying cause statistics alone to accurately summarise the mortality pattern of a population has, therefore, been questioned.


When only a single underlying cause is selected for tabulating cause-specific statistics, other information provided on the death certificate is lost, for instance, immediate cause of death, causes and conditions that intervene between the underlying and immediate causes of death and many other contributory causes that were involved, but did not directly influence the death. The information lost may be important to the understanding of the process of death. Since the introduction of multiple cause coding, data suggest that, on average, 2 causes (and conditions) would be lost per death if only the single underlying cause was recorded. The loss of information is a particular problem for deaths attributed to external causes (injury, poisoning and violence) which are classified to the circumstances of death, rather than according to the nature of injury.


In response to the change in coding structure, the ABS has from 1997 calendar year data, coded all causes and conditions reported on each death certificate using software developed by the National Center for Health Statistics (NCHS) of the United States of America. The NCHS software has three major components: MICAR (Medical Indexing, Classification and Retrieval System), ACME (Automatic Classification of Medical Entities) and TRANSAX (Translation of Axes). These three components have been developed to read the textual entries and convert them into a file containing multiple causes of death in accordance with ICD coding rules.


The automation of cause of death coding has brought a number of benefits, including improved data quality and timeliness. It will facilitate policy relevant epidemiological and demographic research by focusing on competing risks and disease interactions which are not possible from the cause of death statistics solely based on underlying cause.


The introduction of automatic coding of causes of death has introduced a break in time series data in 1997. Appendix 5 provides comparability factors to assist with the linking of manually and automatedly coded data.