Australian Bureau of Statistics

Rate the ABS website
ABS Home > Statistics > By Release Date
4363.0.55.001 - National Health Survey: Users' Guide - Electronic Publication, 2007-08  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 14/09/2009   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

This document was added 09/17/2009.



CONTENTS

Introduction
Tobacco smoking
Alcohol consumption
Exercise
Body mass
Dietary behaviours


INTRODUCTION

A range of genetic, social, economic and environmental factors are recognised as affecting the risk of ill-health, i.e. the chance an individual has of developing a particular illness or injury. Specific lifestyle and related factors which have been identified as positively and/or negatively impacting health include diet and nutrition, use of medicines, being overweight or obese, physical activity, high blood cholesterol, high blood pressure, smoking and alcohol use.

It is clearly not possible, and in some cases inappropriate, in a survey such as the NHS to attempt to address the whole range of factors likely to affect health. The approach taken in this survey was to focus on selected lifestyle-related health risk factors identified through consultations with health professionals, administrators and policy makers as major issues of concern, and considered appropriate for inclusion in an interview survey of this type.

Health risk factor topics included in the 2007-08 NHS were:

  • tobacco smoking;
  • alcohol consumption;
  • exercise;
  • body mass; and
  • dietary behaviours.

Other health risks may be indicated through information obtained in the survey about other health and related characteristics, such as the presence of particular long term conditions. The collection of information about health risk factors and behaviours in conjunction with other health and population characteristics enables all elements to be analysed together. However, while data from the survey may suggest apparent associations between particular risk factors and certain medical conditions, the data should not be interpreted as indicating causal relationships.

Some caution should be used in drawing together data for the different risk factors covered, as the reference periods used differ, e.g. smoking at time of interview, alcohol consumption in the last week, exercise in the last week and last two weeks. However, when used with care, data from the NHS can describe populations which may be at special risk due to the presence of combinations of risk factor behaviours and characteristics.

Most of the specific risk factors covered have been addressed in previous ABS surveys, either at national or State/ACT levels. Major changes in the coverage of risk factors between the 2007-08 NHS and the 2004-05 NHS are summarised in the table below.

Coverage in 2007-08 and 2004-05

Topic
Coverage

Smoking
Similar, with additional data items, now from 15 years and over
Alcohol consumption
Similar, now from 15 years and over
Exercise
Similar, with additional data items
Height, weight, BMI
Similar, with additional data items including measured
height and weight, and waist and hip circumference
Dietary indicators
Similar, with additional data items
Breastfeeding
Not collected in 2007-08
Adult immunisation
Not collected in 2007-08
Child immunisation
Not collected in 2007-08
Supplementary women's health topics
Not collected in 2007-08



Where appropriate to the survey vehicle and consistent with the data requirements of users, similar methodologies were employed in the 2007-08 NHS to those used in previous surveys to enhance comparability and enable use of the data for analysing changes over time. Comments regarding comparability between the 2007-08 and 2004-05 NHS are contained in the individual topic descriptions which follow. A more general discussion of time series issues relating to the 1989-90, 1995 and 2001 surveys is contained in Occasional Paper: Health Risk Factors - a Guide to Time Series Comparability from the National Health Survey, Australia (ABS cat. no. 4826.0.55.001).


SMOKING

Definition

This topic refers to the smoking of tobacco, including manufactured (packet) cigarettes, roll-your-own cigarettes, cigars and pipes, but excluding chewing tobacco and smoking of non-tobacco products. The topic focused on ‘regular smoking’, where 'regular' was defined as one or more cigarettes (or pipes or cigars) per day as reported by the respondent.

The topic primarily describes smoking status at the time of interview; i.e. current smokers (daily, weekly and other), ex-smokers, and those who had never smoked 100 cigarettes, nor pipes, cigars or other tobacco products at least 20 times, in their lifetime.


Methodology

Respondents were asked whether they currently smoke. Respondents who answered yes were asked whether they smoked daily. Those who did not smoke daily were asked whether they smoked at least once a week. Along with respondents who reported that they did not currently smoke, they were then asked whether they had:
  • ever smoked regularly (that is, at least once a day),
  • smoked at least 100 cigarettes in their life, and
  • smoked pipes, cigars or other tobacco products at least 20 times in their life.

If a respondent did not currently smoke, or had never smoked at least 100 cigarettes, nor smoked pipes, cigars or other tobacco products at least 20 times in their life, they were classified as persons who had never smoked, and sequenced to the questions about other people in the household.

Current daily and ex-daily smokers were asked the age they had started smoking. Ex-daily smokers were asked whether they had stopped smoking regularly in the last 12 months, and the age they were when they stopped smoking regularly. Current smokers were asked whether their smoking had increased, decreased or stayed the same in the last 12 months, and whether they usually smoked inside the house.

Respondents in households other than single person households were asked whether anyone else in the household smoked regularly, and if so, the number of people and whether they usually smoked inside the house.


Population

Information was collected for persons aged 15 years and over. Please note that this differs from previous NHSs where these questions were asked of persons 18 years and over.


Data items

Output categories for the data items used in the 2007-08 survey are available from the list of output data items available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).

Please note that:
  • Although the items 'Numbers of daily smokers in household' and 'Whether any daily smokers smoked at home indoors' are household level characteristics, the items are on each person's record.
  • Respondents were asked whether they smoked or had ever smoked 'regularly, that is, at least once a day'. The term 'regular' is replaced by the term 'daily' in the data items.


Interpretation

Points to be considered in interpreting data from this survey include the following:
  • Some under-reporting of persons identifying as current smokers is expected to have occurred due to social pressures, particularly in cases where other household members were present at the interview. In the 2007-08 survey, interviewers were given the opportunity to indicate whether a parent was present at the time of the interview for persons aged 15 to 17 years, in order to assist with analysis of some aspects of under-reporting. However, the extent to which under-reporting has occurred and its effect on the accuracy of survey estimates are unknown.
  • Concepts such as ‘regular’ were open to different interpretation by respondents and may not have been consistently applied in reporting information in this survey, despite a prompt to respondents that regular meant ‘at least once a day’.
  • The selected adult respondent may not have known the smoker status of all other members of the household if, for example, another member only smoked when at work, or children kept their smoking hidden from parents. As a result, some undercounting may have occurred. Estimates of the prevalence of smoking in the population should therefore be based on person level data rather than responses to the 'smokers in household' questions.
  • The categories of smoker status, and the concepts on which they are based, align with those in the National Health Data Dictionary (NHDD).
  • Duration of smoking is derived from reported age commenced daily smoking to current age at the time of the survey (for current smokers), and from age commenced daily smoking to age last ceased daily smoking (for ex-regular smokers). The items are therefore subject to errors around the ages reported by respondents, and the derivation of 'duration' takes no account of periods (potentially long periods) when the respondent may have ceased smoking only to start again.
  • Whether levels of smoking had increased, decreased or stayed the same since 12 months ago is based on self-perception.


Comparability with 2004-05

Data for most smoking items are directly comparable between the 2004-05 and 2007-08 surveys. New items collected in 2007-08 are:
  • smoking level compared to 12 months ago;
  • whether stopped smoking daily in the last 12 months; and
  • presence of parent during smoking questions.

In both surveys, other smokers in the household could include children.

As noted above, smoking information for the 2007-08 survey was collected for persons aged 15 years and over, whereas in the 2004-05 NHS, it was only collected for persons aged 18 years and over.


ALCOHOL CONSUMPTION

Definition

This topic refers to consumption of alcoholic drinks, and focuses on two aspects of consumption:
  • intake of alcohol, derived from information about the types and quantities of alcoholic drinks (including homemade wines and beers) consumed on the three most recent days in the week prior to interview on which alcohol was consumed; and
  • the frequency of consuming 'at risk' amounts of alcohol in the previous 12 months. Amounts are defined in terms of 'standard drinks', where an Australian Standard Drink contains 10 grams (equivalent to 12.5 mls) of alcohol.

Intake of alcohol refers to the quantity of alcohol contained in any drinks consumed, not the quantity of the drinks themselves.


Methodology

Respondents aged 15 years and over were asked how long ago they last had an alcoholic drink. Those who reported they had a drink within the previous week were asked the days in that week on which they had consumed alcohol (excluding the day on which the interview was conducted), and for each of the most recent three days in the last week on which they drank, the types and quantities (number and size) of drinks they had consumed. They were also asked whether their consumption in that week was more, about the same, or less than their usual consumption.

Information was collected separately in respect of the following categories of alcoholic drinks:
  • Beer
      • light beer
      • mid-strength beer
      • full-strength beer
      • type not known
  • Wine
      • red wine
      • white wine
      • low alcohol wine
  • Champagne/sparkling wine
  • Ready to drink spirits/liqueurs
  • Liqueurs
  • Spirits
  • Fortified wine
  • Cider
  • Other alcoholic drinks

Respondents who reported having beer or wine were asked supplementary questions to identify the type (e.g. light beer, white wine), as shown above. If interviewers were unsure in which category a reported drink belonged, details were recorded in ‘other alcoholic drinks’ for checking/reclassifying as appropriate during office processing.

Respondents were asked to report the number of drinks of each type they had consumed, the size of the drinks, and where possible the brand name(s) of the drink(s) consumed on each of the most recent three days in the last week on which they had consumed alcohol.

The collection of accurate data on quantity of alcohol consumed is difficult, particularly where recall is concerned, given the nature and possible circumstances of consumption. Interviewers were provided with extensive documentation and training to assist with recording of amounts consumed. Where possible, information was collected in terms of standard containers or measures; i.e. 10 oz glass, stubbie, nip, etc. Where the size of the drink did not readily fit into the list provided to interviewers, they were asked to record as much information as necessary to clearly indicate quantity.

Reported quantities of drinks consumed were converted to millilitres of alcohol present in those drinks, and then summed to the drink type, day, and week level as required. The methodology to convert drinks to mls of alcohol consumed is as follows:
      Alcohol content of the drink consumed (%) x number of drinks (of that type) consumed x vessel size (in mls).

This conversion was performed electronically, supported by clerical coding for cases which could not be coded automatically.

Where precise brand x type of drink information was not recorded, default alcohol content values based on drink type were applied. These values are shown below:


Light beer
0.027
Mid-strength beer
0.035
Full-strength beer
0.049
Stout
0.058
Wine coolers
0.035
Low alcohol wines
0.090
Fortified wines
0.178
White wine
0.124
Red wine
0.133
Sparkling wine/champagne
0.133
Spirits
0.400
Liqueurs
0.200
Pre-mixed spirits (e.g. UDL)
0.050
Alcoholic cider
0.047
Extra-strong cider
0.075
Cocktails
0.315
Other alcoholic beverage
0.274



It is recognised that particular types or brands of beverage within each of these categories may contain more or less alcohol than indicated by the conversion factor, e.g. full-strength beers are usually in the range 4% to 6% alcohol by volume. The factors are considered to be sufficiently representative of each category as a whole for the purposes of indicating relative health risk as appropriate to the aims of this survey. However, it should be noted that these categories, defined by the conversion factors used, may not reflect legal definitions.

In addition to information about alcohol consumed in the previous week, respondents who reported they had drunk alcohol in the previous 12 months were asked about the number of times (days) in that period on which they had consumed:
  • 7-10 standard drinks or 11 or more standard drinks in a day if male; or
  • 5-6 standard drinks or 7 or more standard drinks in a day if female.

Respondents who reported that they had drunk alcohol in the last 12 months were also asked about their level of consumption compared to 12 months ago.


Population

Information was collected for persons aged 15 years and over. Please note that this differs from previous NHSs where these questions were only asked of persons 18 years and over.


Data items

Output categories for the data items used in the 2007-08 survey are available from the data item list available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).


Interpretation

Points to be considered in interpreting data on alcohol consumption from this survey include the following:
  • Some under-reporting of consumption, both in terms of persons identifying as having drank alcohol in the reference week, and in the quantities reported, is expected to have occurred. Investigations in relation to previous NHSs showed possible under-reporting to be as high as 50% for some types of drink. In the 2007-08 survey, interviewers were given the opportunity to indicate whether a parent was present at the time of the interview for persons aged 15 to 17 years, to assist with analysis of some aspects of under-reporting.
  • The extent to which under-reporting has occurred and its effect on the accuracy of survey estimates are unknown. Any under-reporting which may have occurred, however, does not invalidate the survey results as indicators of relative consumption levels (current and over time), or the relative health risks of the consumption levels identified.
  • Respondents were asked to record all days in the previous week on which they had consumed alcohol, but details of consumption (type and amount of drink) were only collected for the three most recent days on which they had consumed alcohol. Due to the fact that more people were interviewed early in the week, this methodology may have resulted in the possibility that mid-week drinking occasions could be under-represented in the calculation of level of risk, and weekend drinking occasions could be over-represented.
  • The first column of the table below shows, for each day of the week, the proportion of people who reported drinking on that day in response to the question regarding which days they had consumed alcohol in the last week. The second column shows, for persons who reported drinking on a given day, the proportion who provided consumption details for that day.

Day on which consumed alcohol
% of adults reporting drinking on that day
Of adults who drank on that day, % providing consumption details for that day

Monday
37.7
70.3
Tuesday
37.9
64.1
Wednesday
39.1
55.9
Thursday
39.7
48.4
Friday
55.6
64.4
Saturday
64.0
74.3
Sunday
49.1
78.0

  • As shown in the table above, the proportion of persons reporting drinking was highest on weekends, and the proportion of persons providing consumption details was also highest for weekends. ABS analysis has indicated that the 3 day methodology has a small impact on the overall level of health risk at the population level, however, as the effect is considered to be stable over time, analysis of relative risk levels over time should not be affected.
  • To assist users of the data, a Weekend Consumption Flag has been derived to indicate whether consumption during the weekend (i.e. Friday, Saturday, Sunday) is fully, partly or not recorded in the data.
  • Two indicators of alcohol risk level were derived from the average daily amount of alcohol consumed:
      • average over the 1 to 3 days for which consumption details were recorded; and
      • average over the 7 days of the reference week, i.e. average consumption over 3 days x number of days consumed alcohol / 7.
  • Published data are compiled using the 7 day average, which is also the basis for assessing risk level; see point below. Results compiled using the 3 day average are available on request.
  • According to average daily intake over the 7 days of the reference week, respondents were grouped into three categories of relative risk level. Risk levels are based on the 2001 National Health and Medical Research Council (NHMRC) risk levels for harm in the long term, and assume the level of alcohol consumption in the week recorded was typical. The average daily consumption of alcohol associated with the 2001 risk levels is as follows:

Relative risk level Male Female

Low risk Less than or equal to 50ml Less than or equal to 25ml
Risky More than 50-75ml More than 25-50ml
High risk More than 75ml More than 50ml

  • It should be noted that whereas the 2001 NHMRC risk levels assume ongoing consumption at the levels reported, indicators derived in the 2007-08 NHS relate to consumption only during the reference week and take no account of whether or not consumption in that week was more than, less than, or similar to usual consumption levels. In addition, this indicator takes no account of other factors related to health status, or other lifestyle behaviours which may influence the absolute level of personal health risk from drinking alcohol. While data may not reflect the usual drinking behaviour of the respondent at the individual level, at the population level this is expected to average out and be representative of the total population.
  • As noted previously, reported quantities of alcoholic drinks consumed were converted to quantities of alcohol consumed. While brand/drink specific conversions were used where possible, some conversions were based on factors representing the alcohol content of each type of drink category as a whole. To the extent that individuals consumed particular brands/types of drink within each group with an alcohol content higher or lower than that represented by the default factor, the derived intake may over or under-state actual intake. Again, this effect is considered to even out at the population level.
  • Where quantities of alcohol consumed have been converted to standard drinks, a factor of 12.5 mls of pure alcohol per standard drink has been applied (equivalent to 10 grams of alcohol).
  • Whether alcohol consumption had increased, decreased or stayed the same since 12 months ago is based on self-perception.


Comparability with 2004-05

The methodology used in the 2007-08 survey for the collection of data about the quantity of alcohol consumed was essentially the same as that used in the 2004-05 survey. Results for the two surveys are therefore considered directly comparable.

There were, however, some changes to the questionnaire and supporting coding systems used in 2007-08 which involved updating and expanding the index lists supporting the system used to derive alcohol intake. These changes were aimed at improving the accuracy with which alcohol intake was derived from reported consumption. As the main sources of error in this topic are reporting errors, these changes should only have a marginal impact on the overall quality of alcohol consumption data.

In drawing comparisons, consideration should also be given to the social factors and general changes in health awareness which have occurred in the period between surveys and which may have influenced the levels of reporting.

New items collected in 2007-08 are:
  • how often had an alcoholic drink of any kind in the last 12 months;
  • alcohol consumption level compared to 12 months ago; and
  • presence of parent during alcohol questions.

As noted above, alcohol information for the 2007-08 survey was collected for persons aged 15 years and over, while in the 2004-05 NHS it was only collected for persons aged 18 years and over.


EXERCISE

Definition

This topic covers three components of physical activity:
  • exercise undertaken for fitness, recreation or sport during the two weeks and week prior to interview;
  • walking for transport in the week prior to the interview; and
  • time spent sitting during work and when at leisure on a usual day.


Methodology

Current physical activity guidelines for Australian adults include at least 30 minutes of moderate-intensity physical activity on most, preferably all, days. To gauge levels of activity, respondents were asked a series of questions about the exercise they undertook in the last week and last two weeks, expressed in the three categories of walking, moderate exercise and vigorous exercise.

For the purposes of the survey, moderate exercise was defined as exercise undertaken for fitness, recreation or sport that caused a moderate increase in the heart rate or breathing of the respondent. Vigorous exercise was defined as exercise undertaken for fitness, recreation or sport that caused a large increase in the respondent’s heart rate or breathing.

The application of these definitions reflected the respondent’s perception of moderate or vigorous exercise or walking, and the purpose of that activity. Responses may have varied according to the type of activity performed, the intensity with which it was performed, the level of fitness of the participant, and their general health and other characteristics (e.g. age). For example, some respondents may consider a game of golf to be moderate exercise while others may consider it walking. Information was not recorded in the survey about the type of activities undertaken.

Respondents were asked whether they did any:
  • walking for fitness, recreation or sport during the previous two weeks;
  • walking for fitness, recreation or sport during the previous week:,
  • moderate exercise (apart from walking) for fitness, recreation or sport during the previous two weeks:
  • moderate exercise (apart from walking) for fitness, recreation or sport during the previous week:
  • vigorous exercise for fitness, recreation or sport during the previous two weeks; and
  • vigorous exercise for fitness, recreation or sport during the previous week.

For each of these categories of exercise, respondents were asked:
  • the number of times they had done that exercise in the previous two weeks;
  • the number of days they had done that exercise in the last week;
  • the total amount of time spent (hours and minutes) doing that exercise over the previous two weeks; and
  • the total amount of time spent (hours and minutes) doing that exercise in the last week.

The two-week time period was maintained in 2007-08 to allow data to be compared to previous surveys. The move to the one-week reference period allows some comparability over time, as well as aiding the calculation of whether the respondent met physical activity guidelines.

Respondents who answered these questions were also asked about their level of activity compared to 12 months ago.

From the information recorded about the frequency, duration and intensity of exercise undertaken for fitness, recreation or sport, an exercise level was derived for each respondent. The aim was to produce a descriptor of relative overall exercise level, and to indicate the quality of the activities undertaken in terms of maintaining heart, lung and muscle fitness. Whether a person has met physical activity guidelines is calculated using the following formula:
      No. of times activity undertaken (in last week/two weeks) x Average time per session (minutes) x Intensity

where intensity, or metabolic equivalent of task (MET), is a measure of the energy expenditure required to carry out the exercise, expressed as a multiple of the resting metabolic rate (RMR). As the survey did not collect details of the types of activities undertaken, an intensity value was estimated for each of the three categories of exercise identified in the survey, as follows:
  • 3.5 for walking;
  • 5.0 for moderate exercise; and
  • 7.5 for vigorous exercise.

A score was derived for each of the three categories of exercise and then summed to provide a total for the respondent for that period. Respondents were grouped into exercise levels according to their score. For the two week period, score ranges were grouped and labelled as follows:

Exercise level Criteria

Sedentary Scores less than 100 (includes no exercise)
Low Scores of 100 to less than 1600
Moderate Scores of 1600 to 3200, or more than 3200 but less than 2 hours vigorous exercise
High Scores greater than 3200 and 2 hours or more of vigorous exercise



After the specific exercise questions, all respondents were asked whether they had walked in the previous week for periods of 10 minutes or more, for the purpose of going from place to place (i.e. for transport, not for fitness, recreation or sport). Those who had done so were asked the number of times they had walked for transport in the last week and the total time walked.

Respondents who were employed were asked whether their usual activity at work during a typical work day was mostly sitting, mostly standing, mostly walking, or heavy labour. Full-time employees were asked to report the amount of time (hours and minutes) they spent sitting at work during a usual work day, and the amount of time (hours and minutes) they spent sitting watching television and using the computer before and after work.

All other respondents were asked to report the time (hours and minutes) they spent sitting while watching television and using the computer on a usual week day, and all respondents were then asked to report the amount of time (hours and minutes) they spent sitting in other leisure time on a usual work/week day. Data is not available separately for time spent in leisure due to the potential overlap of responses (e.g. persons may have reported the same activity for 'time spent sitting at the computer' and 'time spent sitting in other leisure time'). There may be other activities which were not included by respondents, so this data should be used with caution.

Data from the questions on walking for transport, level of activity at work and time spent sitting do not contribute to the calculations of exercise level.


Population

Information was collected for all persons aged 15 years and over.


Data items

Output categories for the data items used in the 2007-08 survey are available from the data item list available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).


Interpretation

Points to be considered when interpreting data relating to exercise for fitness, recreation or sport include the following:
  • The topic conceptually excludes physical activity undertaken for reasons other than fitness, recreation or sport (e.g. household duties). As a result the data should not be interpreted as necessarily indicative of overall activity levels of persons, or of their fitness.
  • Although respondents were only asked to provide details of exercise undertaken for fitness, recreation or sport, some respondents may have reported details of activities at work which may have contributed to some very high levels of exercise reported. However, because information was not available to confirm this had occurred, the data were not amended and so remain as reported. The possibility that this has occurred in some cases should be considered in interpreting the data.
  • The information is ‘as reported’ by respondents and reflects the respondent’s perception of the activity undertaken, the intensity of their participation, their level of fitness, etc. Information about exercise undertaken by persons aged 15 to 17 years may have been reported by an adult within the household, usually a parent. The child may or may not have been consulted. As a result, data for this age group should be interpreted with particular care.
  • In general, the use of both a two-week and one-week reference period was not considered to pose significant recall problems for respondents. For many people, participation in exercise is regular and/or for a set period each session. However, to the extent that persons undertook exercise in less formal circumstances or that the reference period was atypical of usual exercise patterns, the accuracy of the information provided may have been affected.
  • Research on exercise or physical activity has recently moved away from the use of MET values in deriving exercise level, placing more emphasis on time exercised as a key indicator. In the 2007-08 NHS, retention of the exercise level approach described above was primarily for the purpose of consistency and comparability with data from previous NHSs. In keeping with the move towards time exercised as a key indicator, an item has been derived to indicate whether or not the respondent had met the recommended guidelines for exercise in the week prior to the interview that uses the total time spent walking for exercise and transport, as well as the total time spent on moderate or vigorous exercise and the number of days the respondent exercised. However, this data should be used with caution due to its reliance on self-reported information and the fact that the derivation does not take into account all types of physical activity.

Points to be considered when interpreting data relating to walking for transport include the following:
  • Walking for transport is a difficult concept to measure and define in a way which is meaningful to both respondents and users of the data. Testing before the survey showed significant recall and reporting problems for respondents, however it was not possible within the constraints of the survey to adequately address this issue. The data from this topic, therefore, is considered to be of poor quality, and should be interpreted with caution.
  • In this survey, walking for fitness, recreation or sport and walking for transport are conceptually separate activities, and occasions should be recorded as of either type, not both. Respondents may, however, have reported the same occasions of walking in both sections, as, for example, they may have chosen to walk to work for the exercise rather than take the bus. The order of the questionnaire and instructions to interviewers were aimed at giving priority to recording such occasions as 'walking for fitness, recreation or sport', however, the possibility that there may have been some dual reporting of data should be kept in mind.
  • Walking for transport conceptually excludes walking done at work. Interviewers were asked to exclude these cases where they became aware that respondents had included walking at work. However, testing indicated the likelihood that some respondents will have reported walking at work in response to this question in the final survey.
  • The 10 minute threshold (per occasion) is based on advice that this is the minimum time required before some benefits to health accrue from walking. It also provided a cue to respondents about the occasions of walking they should include. However, it is clear from some responses recorded that this threshold was not consistently applied by respondents, and this has impacted both reporting of occasions of walking for transport and the total time reported.


Comparability with 2004-05

The majority of the data on exercise for fitness, recreation and sport were collected in the 2007-08 NHS with the same methodology and questions used in the 2004-05 survey, and therefore most results are considered directly comparable. The following changes, however, should be noted:
  • the 2007-08 NHS collected information for physical activity in both the two weeks prior and the one week prior to the interview, while the 2004-05 NHS collected this information for the two weeks prior to the interview only. Care should be taken to ensure the correct time frames are used in any comparison of exercise levels between the two surveys;
  • the reference period for the 'walking for transport' questions changed from 'yesterday' in the 2004-05 survey to 'the previous week' in the 2007-08 survey. Care should be taken when comparing these items between the two surveys; and
  • the question wording for questions on walking was re-ordered from 'sport, recreation or fitness' in 2004-05 to 'fitness, recreation or sport' in 2007-08.

Over recent years there has been an increasing focus by governments and media on health and lifestyle issues around obesity and physical activity. While such attention is likely to influence the levels of activity in the community, it may also have an impact on reporting behaviour; for example, creating a tendency to report what is perceived to be a desirable level of activity rather than actual activity. This should be considered in interpreting changes between results from 2007-08 and 2004-05.

New items for the 2007-08 survey include:
  • exercise level in the last week;
  • number of days exercised in the last week;
  • whether met physical activity guidelines;
  • level of activity at work;
  • time spent sitting at work; and
  • time spent sitting at leisure.


BODY MASS

Definition

This topic refers to:
  • the height and weight of respondents as reported and measured during interview;
  • the waist and hip measurement of respondents as measured during interview;
  • derived self-reported and measured body mass;
  • waist circumference; and
  • derived waist to hip ratios and related risk categories.


Methodology

Self-reported height and weight

Respondents were first asked whether they considered themselves to be underweight, an acceptable weight or overweight. Women who identified that they were pregnant at the time of the interview were sequenced out of the module at this point. Remaining respondents were asked whether their weight had increased, decreased or stayed the same since 12 months ago. They were then asked to report their weight and height without shoes. Answers provided in imperial measurements were recorded by interviewers and converted into metric measurements. If respondents rounded their weight or height (e.g. 'about 6 feet') interviewers prompted for a more exact measure where possible.

Physical measurements

Physical measurements were taken towards the end of the survey. All physical measurements were voluntary, and women who had identified they were pregnant were not measured. Interviewers used digital scales to measure weight, a stadiometer to measure height, and a metal tape measure (which avoided the risk of the tape stretching) to measure waist and hip circumference. Thorough interviewer training identified the points at which hips and waists were to be measured (as recommended by Australian government health agencies), as well as how to take the measurements with the least amount of interviewer and respondent discomfort (either holding the end of the tape at the appropriate point and asking the respondent to turn around until the tape met, or asking the respondent to hold the end of the tape and walking around them until the tape met).

Interviewers encouraged respondents to remove their shoes and any heavy clothing, e.g. jumpers, before they took measurements, however, this was voluntary, and may not have occured in some cases. Interviewers were not required to record if they thought clothing may have impacted significantly on measurements. Weight was recorded in kilos to one decimal point, and height, waist and hip measurements were recorded in centimetres to two decimal points. If a respondent's waist or hip measurement was more than two metres, it was recorded as 200.00.

Body mass index scores

Body mass index (BMI) scores were derived using Quetelet’s metric body mass index which is calculated as weight (kg) divided by height (m)2. BMI scores are commonly grouped for output. Although certain ethnic groups, including Asian and Indigenous people, have been shown to have an increased prevalence of disease at much lower BMIs than Europeans (Wood, 2007), the NHS is not able to differentiate for ethnicity, therefore BMI cut-off points are those established for people of European origin. The output classification for adults used for this survey is shown below:

BMI score

Underweight
Grade 3 thinness
Less than 16
Grade 2 thinness
16 to less than 17
Grade 1 thinness
17 to less than 18.5
Normal weight
(ABS time-series cut-off)
18.5 to less than 20
(ABS time series cut-off)
20 to less than 25
Overweight
Grade 1 overweight
25 to less than 30
Obese
Grade 2 overweight
30 to less than 40
Grade 3 overweight
40 or more



While the formula to calculate BMI scores is the same for adults and children, the classification of children's BMI is different to that of persons aged 18 years and over, and takes into account individual age and sex. BMI cut-off ranges for children 2 to 17 years of age are included in Appendix 5: Classification of BMI for children. The NHS uses the half-year cut-off points to calculate children's BMI scores for persons aged 5 to 17.

For more information on this classification please refer to the list of output data items available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).

Waist circumference

Waist circumference reflects mainly subcutaneous abdominal fat storage, and has been shown to positively correlate to disease risk (NHMRC, updated 12 March 2004). The scale used for determing risky waist circumference in the 2007-08 NHS is that recommended by the World Health Organisation, (See Obesity: preventing and managing the global epidemic. Report of a WHO Consultation, 2000). As with BMI, the cut-off points in this scale are best used for people of European origin, however, as ethnicity cannot be determined in the NHS, the same cut-off points are used for all respondents.

Waist circumference risk indicator - adults
Not at risk
Increased risk
Substantially increased risk

Males
Less than 94cm
94cm or more
102cm or more
Females
Less than 80cm
80cm or more
88cm or more



Waist to hip ratio

The waist to hip ratio (WHR) is a simple measure of central obesity. The score from the WHR predicts the risk of developing several conditions associated with excess abdominal fat. Excess abdominal fat distribution is indicated by a WHR greater than 0.8 for women and 0.9 for men.


Population

Self-reported information was collected from persons aged 15 years and over. Physical measures were obtained for persons aged 5 years and over.


Data items

Output categories for the data items used in the 2007-08 survey are available from the data item list available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).

Waist and hip measurements, and self-reported and measured height, weight and body mass index scores are stored on the data file and can be grouped in output to suit individual user needs.


Interpretation

When interpreting data for this topic, users should bear in mind that:
  • Self-reported and/or measured height and/or weight are not available for some respondents, which prevents a BMI score being calculated for them. It cannot be assumed that BMI patterns for these people are the same as those for people who reported their height and weight or were measured.
  • Separate underweight and normal weight categories for children aged 5-17 years were not able to be incorporated into tables used for the National Health Survey: Summary of Results, 2007-08 (cat. no. 4364.0). This data is now available separately for NHS 2007-08.
  • While BMI is a useful tool to assess and monitor changes in body mass at the population level, it may be an inappropriate measure of the body fatness of certain populations and certain individuals; for example, those whose high body mass is due to muscle rather than fat, those with osteoporosis who have lower than usual BMI, or those of non-European background, whose risk levels are not accurately reflected in the BMI cut-off points used.


Comparability with 2004-05

Data collected on self-reported height and weight in the 2007-08 NHS used the same methodology and questions as the 2004-05 survey and are therefore directly comparable with 2004-05 results. There are, however, a number of significant differences for Body Mass items in 2007-08:
  • Measured height and weight were collected in 2007-08. Separate measured height, weight and BMI items are included in output. Previous NHSs collected only self-reported height and weight data, therefore care must be exercised in comparison of any data to ensure the correct items are being compared.
  • Measured height and weight data was collected for children aged 5 years and over in 2007-08. No data for children was collected in previous National Health Surveys, however measured height and weight data for children aged 2 years and over was collected in the 1995 National Nutrition Survey (NNS). For further information on NNS data, please refer to the National Nutrition Survey: Users' Guide, 1995 (cat. no. 4801.0). Information from the 1995 NNS was used in the BMI time series tables in the National Health Survey: Summary of Results, 2007-08 (cat. no. 4364.0).
  • In 2007-08, BMI scores for persons aged 15-17 were based on different cut-off points to those used in the 2004-05 publication (which used the adult cut-off points rather than the children's cut-off points). Comparisons between surveys for self-reported BMI for this age group should be undertaken with caution.

New items for 2007-08 include:
  • Measured height and weight;
  • Measured BMI;
  • Waist circumference;
  • Hip measurements; and
  • Waist to hip ratio.


DIETARY BEHAVIOURS

Definition

This topic covers selected dietary indicators relating to type of milk consumed and usual daily intake of fruit and vegetables. National dietary guidelines recommend a minimum number of serves of fruit and vegetables according to age.


Methodology

Respondents were asked to report the main type of milk they usually consumed, categorised as follows:
  • cow's milk;
  • soy milk;
  • evaporated or sweetened condensed milk; and
  • other type of milk (specify)

The fat content of milk usually consumed (i.e, whole milk, reduced-fat, skim) was then reported for persons who drank milk. Interviewers were able to access fat content for each milk type on their screens to assist in classification.

Respondents were then asked to report the number of serves of vegetables and of fruit they usually eat each day. For the purposes of this survey:
  • A serve of vegetables was defined as a half a cup of cooked vegetables, one medium potato, or one cup of salad vegetables - approximately equivalent to 75 grams. All types of vegetables were included, but legumes were excluded. Tomatoes were included as a vegetable rather than a fruit.
  • A serve of fruit was defined as one medium piece or two small pieces of fruit, or one cup of diced fruit, or quarter of a cup of sultanas, or four dried apricot halves - approximately 150 grams of fresh fruit or 50 grams of dried fruits.
  • Fruit and vegetable juices were excluded.

Prompt cards were used to assist respondents in understanding the concept of a serve, showing pictorial representations as used by the State of Western Australia, 2007. One prompt card showed three pictorial examples of single serves of different vegetables and another card showed three pictorial examples of single serves of fruit. If respondents had difficulty in reporting, interviewers were encouraged to prompt in terms of asking respondents about their usual consumption of vegetables and fruit at breakfast, lunch and dinner, and for snacks.

Respondents were also asked whether their vegetable and fruit consumption had increased, decreased or stayed the same since this time last year.


Population

Information was collected for persons aged 5 years and over. Please note that this differs from the 2004-05 NHS where these questions were asked of persons 12 years and over.


Data items

Output categories for the data items used in the 2007-08 survey are available from the data item list available for download from the National Health Survey: Users' Guide, 2007-08, (cat. no. 4363.0.55.001) and the National Health Survey: Data Reference Package, 2007-08 (cat no. 4363.0.55.002).


Interpretation

Points to be considered in interpreting data for this topic include the following:
  • Data recorded on type of milk usually consumed and the fat content of the main type of milk consumed is based on the information provided by respondents against a defined classification of milk type and fat content categories. The variety of milk products available, and the various terminologies used to label milk products may have led to some misreporting and incorrect classification.
  • Questions on intake of fruit and vegetables are based on short questions used in the 1995 National Nutrition Survey (NNS). The questions, however, are complex, as respondents needed to understand and apply the inclusions/exclusions, understand the concept of a serve and assess their consumption levels accordingly, and think about their total consumption in what would constitute a usual day. Interviewers were instructed to prompt/assist respondents in a standard way if necessary.
  • Inadequate fruit or vegetable consumption was derived in the 2007-08 NHS to assist users to determine whether vegetable and fruit consumption met the recommended guidelines. Respondents who did not meet the recommended guidlines for either fruit or vegetables were considered to have inadequate fruit or vegetable consumption. This item should be used with caution as it is based on self-reported data.
  • Overall, it is considered that the indicators of vegetable and fruit intake from the 2007-08 NHS are of a lower quality than most other items from the survey, but are considered sufficiently reliable for the purposes of assessing broad intake levels for population groups, and comparisons between population groups. Use of the data for other purposes should be undertaken with care.
  • Data for persons aged 5 to 14 years, and 36% of those aged 15 to 17 years, was provided by a proxy, usually a parent. As a result the data reflects the parent's knowledge of the child's consumption. This is likely to be less accurate for usual consumption of fruit than for type of milk and usual consumption of vegetables.
  • A comparison of results from the 2001 NHS with those obtained in the 1995 National Nutrition Survey was published by the ABS in the information paper, Measuring Dietary Habits in the 2001 National Health Survey, Australia (ABS cat. no. 4814.0.55.001).


Comparability with 2004-05

Many dietary indicator questions used in the 2007-08 NHS were similar to those used in the 2004-05 NHS and the data are considered broadly comparable. However, as outlined above, information for the 2007-08 survey was collected for persons aged 5 years and over, whereas in the 2004-05 NHS, information was collected for persons aged 12 years and over.

There were significant differences between the prompt cards used in the two surveys to assist respondents in determining the size of a serve of fruit or vegetables. This may have had some impact on the comparability of the data.

Items collected in 2004-05 that were not collected in 2007-08:
  • food security in the last 12 months; and
  • whether went without meals.

New items for 2007-08 include:
  • fat content of milk usually consumed;
  • vegetable consumption level compared to 12 months ago; and
  • fruit consumption level compared to 12 months ago.


Bookmark and Share. Opens in a new window

Commonwealth of Australia 2014

Unless otherwise noted, content on this website is licensed under a Creative Commons Attribution 2.5 Australia Licence together with any terms, conditions and exclusions as set out in the website Copyright notice. For permission to do anything beyond the scope of this licence and copyright terms contact us.