Australian Bureau of Statistics
Menzies School of Health Research
This Occasional Paper is intended to make the results of current research available to other interested parties. The aim is to encourage discussion and comment.
Views expressed in this paper are those of the authors and do not necessarily represent those of the Australian Bureau of Statistics or of the Menzies School of Health Research. Where quoted or used, they should be attributed clearly to the authors.
This report presents information on the distributions of height, weight and body mass index (BMI) in a large, nationally representative sample of Indigenous Australians aged 5 years and over. Where possible, the results have been compared with reference data from other populations, including data from surveys representative of all Australians. The report also provides the results of analysis of a variety of factors which may be associated with BMI. (For more information about BMI, see page 5.)
Data were collected as part of the first National Aboriginal and Torres Strait Islander Survey, conducted by the Australian Bureau of Statistics in 1994. The analysis was undertaken jointly by researchers at the Australian Bureau of Statistics' National Centre for Aboriginal and Torres Strait Islander Statistics and the Menzies School of Health Research in Darwin.
The results highlight the considerable heterogeneity of the Indigenous population. Indigenous children aged 5-9 years were, on average, short for their age and heavy for their height, based on international reference values. However, there were differences by place of residence, with rural children being the shortest for their age on average, and children in capital cities being the heaviest on average for their height. There was greater than expected variability for all geographic locations, especially in weight for height.
Among children aged 7-15 years, both underweight and overweight were more common than expected, based on Australian reference data. The proportion of children in these weight categories varied according to whether they lived in capital cities, other urban areas or rural areas. Underweight was more common in rural areas and overweight was more common in urban areas. However, even in rural areas the proportion overweight was greater than expected. Similarly, the proportion of children in capital cities who were underweight was more than expected.
Among adults, Indigenous males were on average about 3-4 cm shorter and Indigenous females were about 2-3 cm shorter than their all-Australian counterparts up to age 65 years. There was little difference in mean height among those aged 65 or more. Mean weight was generally lower among Indigenous males than all-Australian males, but the reverse was true for Indigenous females. Based on BMI, about 25% of Indigenous males and 28% of Indigenous females aged 18 or more could be classified as obese. This compares unfavourably with 19% of all-Australian males and females aged 18 or more.
Torres Strait Islander adults were, on average, taller and considerably heavier than Aboriginal adults and had a higher mean BMI. Among Torres Strait Islanders, about 43% of adult males and 50% of adult females could be classified as obese, compared with 24% of Aboriginal adult males and 27% of Aboriginal adult females. The appropriateness for these two Indigenous groups of the BMI cutpoints used for classification remains a point of debate, however.
A number of factors were associated with mean BMI among children and adults. Among children, mean BMI increased with age and was generally higher among children who spoke English as their main language and lower among those who lived in a household where it was reported that someone had gone without food in the past month and/or where there were more than two people per bedroom. Among adults, mean BMI increased with age up to a certain point and then decreased. Mean BMI was higher among adults who reported they had diabetes, identified as Torres Strait Islander, said they had not drink alcohol recently or said they did not drink at all, and was lower among smokers and those living in households with more than four people per bedroom.
Not everyone included in the survey was measured. Measurements of height and weight were available for 62% of children aged 5-17 years and 73% of adults aged 18 years or more. In contrast to most studies, some information was available about the characteristics of those not measured, such as age, sex and questionnaire responses. This information was used to assess whether a bias in the reported results due to non-measurement was likely to have occurred. Although a number of factors were significantly associated with whether someone was measured, it appears they may have worked in opposite directions. Taking all these factors into account, the analysis suggested that any such bias is likely to have been quite small.
The heterogeneity in the distributions of height, weight and BMI among Indigenous people means that it would be inappropriate to assume that the national results apply to every location and every age group. Wherever possible, local information should be used to inform decision-making.