About this publication
Data derived from the annual registered deaths this report presents data and a brief analysis of the trends and changing patterns of causes of infant and child deaths in Australia by socio-demographic characteristics for the years 1982-96. The levels and trends in causes of infant deaths are examined according to its two major components: neonatal and postneonatal deaths. The report also compares the cause-specific infant mortality rates between the Indigenous and non-Indigenous infants and presents an overview of trends and differentials in infant mortality across states and Territories.
The current infant and child death rates in Australia are low by international standards. Although infant and child deaths form only a small proportion (less than 2%) of all deaths, they nevertheless have important public health policy significance.
Infant mortality, defined as deaths of children under one year, has been traditionally viewed as an indicator of the general level of mortality, health and wellbeing of a population and as such has received special attention in public health policy. It is also an indicator of the social development of a population.
Infant mortality has a strong association with both fertility (births) and life expectancy (at birth). The increased survival of infants and young children is generally accompanied by a decline in fertility. The survival of infants and young children is highly affected by preventive health measures and public health programs which aim to improve life expectancy (United Nations, Mortality of Children Under 5: World Estimates and Projections, 1950-2025, Population Studies Series, no. 105, 1988).
Infant deaths are, in general, divided into neonatal and postneonatal periods. Deaths in the neonatal period are those occurring in the first 28 days after birth (0-27 days), while postneonatal deaths are those occurring in the remainder of the first year (28 days to 364 days).
Age and sex pattern
Numbers of deaths decrease significantly with the increasing age of infants. About 35% of infant deaths occurred on the day of birth . A further 16% had occurred by the end of the first week. In total, 62% of infant deaths occurred in the neonatal period. The remaining 38% of deaths occurred in the postneonatal period, with the proportion of deaths declining steadily over this period.
In general, infant mortality rates are higher among males than females. Throughout the 15-year reference period, the infant mortality rate for males was about 27% higher than that for females. Higher male infant death rates were found for almost all leading causes, although the extent of variation differed. SIDS showed a greater differential than did perinatal conditions or congenital anomalies. Average male infant death rates from SIDS exceeded the female rates By 55% in 1982-86, dropping to 35% in 1992-96.
Between 1982-86 and 1992-96 the overall neonatal mortality rate declined by 33%, from 596 deaths per 100,000 live births to 398. The rate of decline was similar for males and females.
Two major groups of causes accounted for 95% of all neonatal deaths during the 15-year period.
These were: perinatal conditions (62% of neonatal deaths) and congenital anomalies (33%). The main specific causes of neonatal deaths in these groups included: hypoxia, birth asphyxia and respiratory conditions (International Classification of Diseases, revision 9 (ICD-9) codes 768-770) (26%); disorders relating to length of gestation and fetal growth (ICD-9 codes 764-766) (18%); congenital anomalies of the heart and circulatory system (9%); congenital anomalies of the nervous system (7%); and haemorrhage (6%).
The perinatal conditions group includes a number of causes that relate to pregnancy, fetal growth, labour, delivery and life of the newborn in the first month. Diseases and conditions originating during pregnancy and in the neonatal period, even though the death may occur later, are included in this group (see Glossary). Disorders specific to the perinatal period such as respiratory conditions and infections also fall into this cause group. From 1982-86 to 1992-96, the overall neonatal death rate from this group of causes declined by 28%, from 353 deaths per 100,000 live births to 254. This rate was lower than the decline in the overall neonatal death rate observed for the same period. As a result, the proportion of neonatal deaths attributed to perinatal conditions increased slightly, from 59% in 1982-86 to 64% in 1992-96.
Respiratory distress syndrome (a disorder characterised by difficult and laboured breathing and cyanosis)(ICD-9 code 769), and other respiratory conditions (ICD-9 code 770), were the leading causes of perinatal deaths in the neonatal period. T he average death rate from these causes combined declined from 164 deaths per 100,00 live births in 1982-86 to 86 in 1992-96. The proportion of neonatal deaths due to these causes also declined, from 28% in 1982-86 to 22% in 1992-96.
The neonatal death rate from respiratory distress syndrome declined by 66% from 61 deaths per 100,000 live births in 1982-86 to 21 in 1992-96. Death rates from other respiratory conditions and intrauterine hypoxia and birth asphyxia also declined during the same period, but at a slower rate, 40% and 30% respectively.
Disorders relating to length of gestation and fetal growth
Deaths attributed to disorders relating to length of gestation and fetal growth (ICD-9 codes 764-766) have fluctuated, but generally showed little change over the period. In 1992-96, the average neonatal death rate due to this cause was 85 deaths per 100,000 live births. About 95% of the neonatal deaths in this group were accounted for by extreme immaturity (see Glossary), a predisposing or contributory condition to other causes. It should be noted, however, that when autopsies are performed deaths are less likely to be classified to this cause.
Congenital anomalies are mental and physical conditions present at birth that are either hereditary or originating from pregnancy. The neonatal death rate from congenital anomalies fell by 41%, from 212 deaths per 100,000 live births in 1982-86 to 124 in 1992-96.
The major external deformities such as anencephalus, spina bifida, or anomalies of the limbs are usually easier to diagnose and therefore likely to be accurately reported, when present.
Other congenital anomalies are difficult to diagnose without post-mortem examinations, and are often underreported, because post-mortems are not always performed.
Congenital anomalies of the heart and circulatory system and of the nervous system were the two largest categories, accounting for 27% and 21% respectively of all neonatal deaths due to congenital anomalies. Between 1982-86 and 1992-96, the average neonatal death rate from congenital anomalies of the heart and circulatory system declined by 35%. Death rate from congenital anomalies of the nervous system declined by 58%, from 50 deaths per 100,000 live births in 1982-86 to 21 in 1992-96. The average annual number of deaths from this cause dropped from 122 to 55 during the same period.
Screening methods used during pregnancy, such as amniocentesis and ultrasound screening, can detect serious defects in the foetus and some pregnancies may then be terminated. A termination of pregnancy is not counted as a live birth. In the 1970s elevated alpha feta-protein levels were used to detect spina bifida; subsequently, this method has increasingly been used to detect anomalies of the central nervous system, resulting in substantial reductions in the number of neonatal deaths from this cause.
Three causes accounted for 76% of all postneonatal deaths during the 15-year reference period. These were: SIDS (49%), congenital anomalies (18%) and perinatal conditions (9%).
Sudden Infant Death Syndrome
SIDS emerged as the leading cause of death in the postneonatal period accounting for nearly half of all postneonatal deaths. Over the 15-year reference period, the average postneonatal death rate from SIDS declined by 57% from 189 deaths per 100,000 live births in 1982-86 to 81 in 1992-96. This decline was, however, not uniform throughout the 15-year period, being mostly concentrated in the period since 1990.
Throughout the period the male death rate from SIDS remained higher and fluctuated more widely than the female rate. During the 1990s the gap between male and female death rates narrowed.
The fall in death rates observed since 1990 could reflect the success of the national health educational campaign which was launched in that year. The campaign highlighted the risk factors thought to contribute to SIDS such as sleeping posture, feeding practices and exposure to passive smoking.
From 1982-86 to 1992-96, the average postneonatal mortality rate from congenital anomalies fell from 67 to 44 deaths per 100,000 live births. The three leading causes of congenital anomalies-congenital anomalies of the heart and circulatory system (43%), congenital anomalies of the nervous system (17%) and chromosomal anomalies (15%)- accounted for 75% of all postneonatal deaths reported in the 15-year period.
Over the reference period, postneonatal death rates from all these categories declined. The largest decline occurred in congenital anomalies of the nervous system, which decreased by about 70% from 14 per 100,000 live births in 1982-86 to 4 in 1992-96.
Vaccine preventable deaths
The health of infants is affected by periodic epidemics of diseases that are preventable by vaccination (see Explanatory Notes, paragraph 20). However, deaths from vaccine preventable causes make up a relatively small proportion of infant deaths. Between 1982 and 1996 there were a total of 56 deaths from diseases covered in the National Immunisation Strategy: 5 deaths were from measles, 14 from whooping cough, 1 from rubella and 36 from haemophilias influenza type B (HiB). These vaccine preventable deaths accounted for less than 1% of all postneonatal deaths reported during the 15-year period.
The death rate for young children (aged 1-4 years) was much lower than for infants and the rest of the population. The child death rate was 52 deaths per 100,000 children during 1982-86 and declined to 36 in 1992-96. The declines in death rates were similar for boys and girls.
The male disadvantage in mortality observed in infancy continued, with young boys showing about 30% higher overall death rate than young girls throughout the period. For instance in 1982-86 the death rate for boys was 59 per 100,000 boys (1-4 years) compared with 45 for girls.
Leading causes of child deaths
The leading cause of death among young children was accidents, poisoning and violence (external causes) which accounted for about 46% of all deaths among children. This was followed by congenital anomalies (14%) and neoplasms (11%). Overall death rates and death rates due to external causes showed a higher mortality among boys, and congenital anomalies and cancers accounted for a higher proportion of deaths among girls than boys.
Accidents, poisoning and violence (external causes)
Although the death rate from external causes declined, it remained the leading cause of child death during the 15-year period from 1982 to 1996. About a third of all child deaths from external causes were due to motor vehicle accidents (on average around 64 deaths per year) and drowning (on average 68 deaths per year).
During the reference period, the child death rates from both these causes substantially declined. The death rate from motor vehicle accidents declined from 9 deaths per 100,000 children per year in 1982-86 to 5 in 1992-96, while the death rate from drowning decreased from 8 in 1982-86 to 5 in 1992-96.
About 6% of all child deaths classified as being due to external causes were homicides. Although there were yearly fluctuations, the average number of deaths attributed to homicides remained similar over the three five-year periods at around 11 deaths per year, with an average homicide rate of 1.1 per 100,000 children aged 1-4 years.
Death rates from external causes as a whole were 39% higher for young boys than for young girls in 1982-86, and 60% higher in the 1992-96 period. For most external causes, the relative disadvantage in mortality for males increased over the reference period.
The average annual death rate for young children from congenital anomalies decreased slightly from 7 deaths per 100,000 population in 1982-86 to 5 in 1992-96. Congenital anomalies of the heart and circulatory system, congenital anomalies of the nervous system and chromosomal anomalies were the three major categories. Within congenital anomalies, the largest decline in the child death rate was recorded for chromosomal anomalies which fell by 49% from 1982-86 to 1992-96.
Although neoplasms caused very few deaths in infancy, they were the third leading causes of death among children 1-4 years of age, accounting for 11% of all deaths between 1982-86 and 1992-96, with an average of 47 deaths per year (or 5 deaths per year per 100,000 children).
About 96% of deaths classified to neoplasms were malignant neoplasms, with slightly more than a third of these due to leukaemia (ICD-9 codes 204-208) and a further quarter of deaths due to brain cancers.
This analysis is restricted to SA, WA and NT, where the identification of Indigenous births and deaths is considered to be relatively complete. Indigenous identification on registration forms has been gradually improving over the years and so analysis has been based on the most recent three-year period available (1994-96). To minimise the effects of annual fluctuations in death rates, the infant and child mortality rates were averaged over the three-year period.
Overall, the Indigenous infant mortality rate (1,857 deaths per 100,000 Indigenous live births) was almost four times higher than the non-Indigenous rate (500 infant deaths per 100,000 non-Indigenous live births).
Deaths from SIDS and diseases of the respiratory system accounted for a greater proportion of Indigenous infant deaths than non-Indigenous infant deaths. The estimated death rate from diseases of the respiratory system for Indigenous infants was more than 23 times higher than that of non-Indigenous infants (233 deaths per 100,000 live births for Indigenous infants compared with 10 for non-Indigenous infants).
The estimated child death rate for Indigenous children (1-4 years of age) for the 1994-96 period was 4.5 times higher than that for non-Indigenous children; 131 deaths per 100,000 children compared with 29, respectively. This difference may also reflect the poor socioeconomic environment in which most Indigenous children live. It is not possible to estimate accurately the death rates for Indigenous children given the small number of deaths that occurred each year, 11 deaths per year on average over the three-year period.
States and Territories
The infant mortality rate varied between States and Territories. Over the period 1982-96 to 1992-96, death rates were considerably higher in NT, which particularly reflects the high proportion of Indigenous children in NT's child population. The infant mortality rate for Australia as a whole declined by 37% over the 15-year period and all States and Territories shared the decline. However, the falls recorded for the ACT (48%), Tas. (46%) and Vic. (44%) were greater than the national rate. While Tas. recorded the second highest rate at the beginning of the period, it was close to the national average at the end. The ACT, SA and Vic. had the lowest infant mortality rates at the end of the period.
The main causes of infant deaths observed for Australia as a whole were, in general, common to all State and Territories. However, deaths attributed to SIDS were higher in Tas. and NT.
All jurisdictions showed a decline in infant death rates from perinatal conditions, with the falls ranging from 9% in NT to 36% in Vic., compared with the national decline of 27%. The decline in the infant death rate from congenital anomalies was greater in Vic., SA and Qld than in other jurisdictions. The national death rate from SIDS was more than halved over the 15-year period. The fall in the SIDS death rate occurred across all States and Territories, with Vic., Tas., the ACT and SA recording the greatest declines.
Owing to the small number of deaths a detailed analysis of the levels and trends in cause-specific child deaths rates by States and Territories has not been made .