Australian Bureau of Statistics
1307.6 - Tasmanian State and Regional Indicators, Dec 2009
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 29/01/2010
|Page tools: Print Page Print All RSS Search this Product|
Population Health Progress Measures: Body Mass
Use of self-reported BMI estimates for progress measurement
Assessment of Progress
Tasmania Together is a long term social, environmental and economic plan for Tasmania's development for a period of 20 years and an overarching framework for planning, budgeting and policy priorities for the government and non-government sectors. The plan involves twelve goals and it utilises statistical information from a wide variety of sources to monitor progress in achieving the goals. Estimates from the triennial National Health Survey (NHS) are used to monitor progress with the Tasmania Together goal of having 'active, healthy Tasmanians with access to quality and affordable health care services'.
Various NHS topics are used in monitoring progress with this goal, including:
This article focuses on NHS indicators of the the body mass of Tasmanian adults.
Whether a person is overweight or obese is determined in the NHS by their score on the Body Mass Index (BMI). BMI scores are calculated for individuals by dividing their weight in kilograms by the square of their height in metres. Thus an 83 kilogram person of 1.8 metres would have a BMI score of 25.6, i.e. 83/(1.8*1.8). Persons whose BMI score falls within a particular range are categorised as either underweight, normal range, overweight or obese as follows:
Two baseline BMI progress measures from the NHS were adopted by Tasmania Together . These are the proportion of the adult population who were overweight and the proportion who were obese at the time of the 2001 NHS. That survey found that an estimated 34.2% of adult Tasmanians who had reported their height and weight, were in the overweight BMI range while 16.4% were in the obese BMI range.
Since these are survey estimates they contain sampling error. The size of the standard errors (the measure of sample error) for these estimates were 5.2% of the estimate of overweight Tasmanians and 8.5% of the estimate of obese Tasmanians. These standard errors can be used to construct confidence intervals for the progress measures. The 95% confidence interval for the estimate of overweight Tasmanians is between 30.6% and 37.8%. Likewise, the 95% confidence interval for the estimate of obese Tasmanians is between 13.6% and 19.2%.
In the 2007-08 NHS, 13% of the adult population (10% male and 16% female) did not provide their height or weight and this group was predominately populated by young women (18-34 years) and women over 55 years. Although no data is available for these persons' height and weight, users should take into consideration the age and sex profile of the 'not stated' group and the possibility of biasing the results for respondents in these demographics when interpreting the BMI data.
USE OF SELF-REPORTED BMI ESTIMATES FOR PROGRESS MEASUREMENT
In calculating BMI, the National Health Surveys generally rely on survey respondents providing information about their height and weight. It has long been recognised that people tend to underestimate their weight and overestimate their height. So as well as the sample error involved with survey estimates, there is also a problem of incorrect reporting leading to non-sample error in the estimates. An idea of the size of this type of error is available from the 2007-08 NHS survey. For the 2007-08 survey, as well as the reporting of height and weight by respondents, height and weight measurements were taken and recorded. Based on heights and weights reported by respondents, the estimate of the proportion of adult Tasmanians who were in the the underweight or normal range was 41.1%, with 36.4% in the overweight range and 22.5% in the obese range. Based on measured height and weight data the estimate of the proportion of adult Tasmanians who were in the underweight or normal range was 36.0%, with 37.3% in the overweight range and 26.7% in the obese range. Confirmation of this tendency is clear in the following graph.
Although there are known limitations to the efficacy of BMI scores calculated from self-reported height and weight, measured height and weight data are not collected regularly. Assuming that the level of reporting error remains consistent from survey to survey, then the self-reported BMI measure remains the most reliable available indicator for progress measurement. As mentioned above, the baseline measurements for the Tasmania Together BMI progress measures were sourced from the 2001 NHS self-reported data and they are subject to sample error.
ASSESSMENT OF PROGRESS
Since the 2001 NHS, two further measurements - from the 2004-05 and 2007-08 NHS - have been produced and used to monitor progress. These progress measurements have been added to the Tasmania Together on-line reports and a progress assessment published. These Tasmania Together measures are represented in the overweight and obese estimates columns in the following graph.
These columns show an increase in both measures (a negative outcome), with the proportion of Tasmanians in the overweight BMI range rising from 34.2% in 2001 to 36.4% in 2007-08, and the proportion that are obese rising from 16.4% to 22.5%. These results are in line with other states and territory. Obesity rates have risen in all states and territories, while overweight rates have stayed the same in Queensland and dropped marginally in the Australian Capital Territory.
The Tasmania Together plan set targets of 30% (overweight Tasmanians) and 14.5% (obese Tasmanians) for the year 2010. The next triennial NHS is planned for a reference year 2010-11 and a final assessment of targets achievement for 2010 will be undertaken when estimates from that survey are released. However, it seems clear that on current trend it is unlikely that the targets will be achieved since the 2007-08 NHS estimates are 36.4% (overweight Tasmanians) and 22.5% (obese Tasmanians). The following table summarises the estimates and their relation to the targets.
For the 2007-08 NHS the 95% confidence interval for the estimate of overweight Tasmanians is between 33.7% and 39.1%. Since this range does not include the 30% target, if the estimate for 2010-11 does not come down then the target will not be achieved.
For the 2007-08 NHS the 95% confidence interval for the estimate of obese Tasmanians is between 19.6% and 25.4%. Since this range does not include the 14.5% target, if the estimate for 2010-11 does not come down then the target will not be achieved.
These assessments correspond with the assessments in the Tasmania Together Progress Board's reporting system that on both measures progress is not being made to target.
National Health Survey: Summary of Results (ABS cat. no. 4364.0)
National Health Survey: Summary of Results; State Tables (ABS cat. no. 4362.0)
Tasmania Together 2020
These documents will be presented in a new window.
This page last updated 28 April 2010