4398.0 - Causes of Infant and Child Deaths, Australia, 1982 to 1996  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 04/02/1998  Ceased
   Page tools: Print Print Page Print all pages in this productPrint All

Scope and coverage

1 This publication contains statistics on deaths of infants (under 1 year of age) and children aged 1-4 years, compiled from data supplied to the ABS by the State and Territory Registrars of Births, Deaths and Marriages.

Data source

2 As part of the registration process, information as to the cause of death is supplied by the medical practitioner (or by a coroner if the death is subject to a coronial inquest) certifying the death.

3 Other statistical information about the deceased is supplied by a relative or other person acquainted with the deceased, or by an official of the institution where the death occurred. For each death, the information on cause of death and general information on the deceased is provided to the ABS by individual registrars for compilation of statistics. The statistics on infant and child deaths by cause presented in this publication cover those registered during the period 1982-96.

Classification of cause of death

4 From the information provided on the death certificates, an underlying cause of death is coded by the ABS according to the rules and guidelines of the ninth revision of the World Health Organisation's (WHO) ICD-9.

5 The ABS classifies all deaths according to a list of 3-digit (or in some cases 4-digit) ICD codes. Deaths from accidents, poisoning and violence are coded according to the circumstances of injury. Such external causes of death are identified by codes beginning with a letter E.

6 In this publication, data are presented according to the main ICD-9 subdivisions and some causes of death within the subdivisions. When data for causes of death are presented in tables, relevant ICD-9 codes are also provided in parentheses to provide a reference to the ICD.

Data quality

7 In compiling death statistics, the ABS employs a variety of quality control measures to ensure that the data are as reliable as possible. These measures include, where necessary, seeking further information for accurate classification of the underlying cause of death, and a comprehensive system of editing and sample checking.

8 The registration of deaths in Australia is virtually complete. However, there could be a time lag between the occurrence of a death and the registration of the event. Each jurisdiction allows a time period for registering a death without incurring a penalty.

9 For deaths, the time allowed for registration varies from one week following death in Vic. and Tas., to one month in NSW. In all jurisdictions provision for late registrations are available. Thus, some deaths which occurred late in the year MAY be registered in the early part of the following year.

10 For calculation of child death rates, Estimated Resident Population (ERP) for the age group 1-4 years was used as the denominator. Australia conducts population censuses at five-yearly intervals. The ERPs are prepared on the basis of recent population census counts and are revised quarterly according to subsequent trends in fertility, mortality and migration.

11 The ERPs for children aged 1-4 years used in the analysis to compute child death rates for 1992-96 were based on the preliminary estimates which will be revised when the detailed data from the 1996 Census become available. However, such revisions will not cause substantial differences to the number of children aged 1-4 years given in the preliminary estimates.

Indigenous mortality

12 Provision has now been made for the registration forms in all States and Territories to record the Indigenous status of the deceased. While some States have collected such information for a relatively long period of time, Qld has only collected such information since 1996.

13 Even in States and Territories where information on the Indigenous status of deaths has been collected for some time, the extent of identification of Indigenous deaths is far from complete. The current coverage levels of the identification of Indigenous deaths in NT, SA and WA are believed to be sufficiently reliable to warrant publication of results. While in the ACT identification of Indigenous deaths is complete, the events were relatively low for computation of reliable mortality rates.

14 The infant mortality estimates for Indigenous children were derived using infant deaths and live births identified as Indigenous. A live birth is considered Indigenous if either parent is identified on the birth notification form as of Aboriginal or Torres Strait Islander origin. An infant death, like deaths of all other ages, is considered Indigenous if the Indigenous status of the deceased child is reported by the informant on the death notification form as of Aboriginal or Torres Strait Islander origin.

15 The quality of the infant mortality estimates is largely dependent on the accuracy and completeness of birth notification forms, whether or not parents are of Indigenous origin. On forms where the Indigenous status is not reported, the child will be recorded as non-Indigenous.

16 The Indigenous child population aged 1-4 years for SA, WA and NT were drawn from the experimental estimates of the Indigenous population computed by the ABS, adjusted according to the preliminary results of the 1996 Census of Population.

Measurement of mortality

17 All infant death rates used in this publication were derived by conventional method, i.e. by relating infant deaths registered in a specific period to the number of live births registered in that period. However, all infant deaths do not occur in the year the infants were born. The proportion of infant deaths which occurred within the same calendar year that the infant was born was relatively stable and changed only over a narrow range during the reference period, from 86% in 1981 to 90% in 1993. Infant mortality rates estimated by the conventional method are likely to have a minimum impact on the assessment of trends.

18 Estimates of child mortality rates have been carried out taking children aged 1-4 years as a group without disaggregating by single years of age. This approach was adopted mainly to minimise the impact of annual fluctuations arising from the small number of deaths registered.

19 For comparison of trends this publication used infant mortality rates averaged over a five-year period. To compute average infant mortality rates the total number of infant deaths registered in a five-year period was divided by the total number of live births registered in the same period. The average child death rates used in the publication have been computed by dividing average number of child deaths by average ERP estimates for the child population for that period.

Vaccine preventable deaths

20 In Australia programs of mass immunisation are mostly administered by State and Territory Governments. The National Health and Medical Research Council takes an advisory role on immunisation and sets a Standard Childhood Vaccination Schedule. The current schedule includes vaccinations against the following diseases: diphtheria, tetanus, pertussis (whooping cough), poliomyelitis, HiB, measles, mumps and rubella. HiB vaccination has been included on the schedule since 1994, the remaining vaccinations were included on the schedule throughout the period 1982-95. Although not part of the immunisation schedule, hepatitis B vaccine is provided to high risk populations.


21 ABS publications draw extensively on information provided freely by individuals, business, governments and other organisations. This publication is based on the information about registered deaths and live births provided to the ABS by the Registrars of Births, Deaths and Marriages of States and Territories. Their continued cooperation is very much appreciated: without it, these statistics would not be available. Information received by the ABS is treated in strict confidence as required by the Census and Statistics Act 1905.