This is an archive copy of a document originally located at http://www.nphp.gov.au/sigpah/gaa/index.htm. All copyright remains with the creator.

Part 1. The importance of physical activity in Australia (cont.)

1.3 Recent physical activity participation data for Australian adults

Summary

There is an urgent need for standardised approaches to measuring and monitoring physical activity participation at the State and National level in Australia. Some States have already conducted physical activity surveys, including New South Wales, South Australia and Western Australia. At the National level, representative population samples of Australian adults were surveyed about their physical activity patterns in November 1997, 1999 and 2000. Specific sub-groups that are more likely to be inactive can be identified in such data; these include women, older adults, people at some social disadvantage, and those from a non-English-speaking background. Trend analyses showed that there were increases in awareness of the Active Australia media campaign, and in understanding some aspects of the moderate physical activity message, however, the rates of participation appeared to fall between 1997 and 1999, and between 1999 and 2000. Further monitoring is needed to confirm these trends but it is likely that substantial efforts will be required to reverse the trend.

1.3.1. Population surveys of physical activity in Australia

There have been several attempts to measure physical activity in Australia since the late 1970s. Early efforts included physical activity questions in the National Heart Foundation Risk Factor Prevalence surveys in 1980, 1983 and 1989 (NHF 1989). There were also efforts to characterise physical activity participation comprehensively in a sport and recreation context between 1984 and 1987 (Bauman et al. 1990). The National Health Survey, collected by Australian Bureau of Statistics (ABS) also asked identical physical activity questions in 1989, 1995 and 2001. These data systems were not comparable, as the questions they asked were different, the sampling and survey methods were different, and the estimates of the proportion of the (adult) population that was 'active' varied widely. More recent surveys were conducted, particularly at the State level, during 1990s. Detailed analyses of NSW Health Survey questions have been reported from the 1994 State Health Survey (Bauman et al. 1996), and from the 1997 and 1998 NSW surveys (NSW Health, Public Health Division 2000). In addition, a specific NSW physical activity survey was conducted in 1996 (Bauman et al. 1999).

Other States have conducted specific physical activity surveys in more recent years. Data from Victoria were published in 1999 in the Active for Life Reports (Smith et al. 1999). Specific surveys to identify physical activity levels have been conducted in South Australia and Western Australia (summaries are provided in the text boxes), which provide local detailed information on physical activity levels, as well as information of specific determinants and factors associated with activity and inactivity. Several NSW surveys have been conducted, including State health surveys with a focus on physical activity, as well as several specific surveys in 1996, 1998 and 1999.

Nonetheless, despite the recent interest in monitoring physical activity, there is no standardised approach to measurement and, therefore, making comparisons among States or assessing trends over time is more difficult.

1.3.2. Recent trends in physical activity in Australia

This section describes three recent National surveys of physical activity, which were carried out in November 1997, 1999 and 2000, using identical survey methods and questions. Analyses of the trends of these National data have been reported elsewhere, by the Australian Institute of Health and Welfare for the first two surveys (Armstrong et al. 2000). State-level comparisons are shown briefly, only for the 1997 and 1999 surveys.

Methods

The data are from three National telephone surveys of participation in physical activity. The surveys were conducted by the Hunter Valley Research Foundation, and were jointly funded by the Australian Institute of Health and Welfare, the Department of Health and Aged Care, the Australian Sports Commission. Oversampling was carried out for some regions, and was funded by specific State and Territory health departments.

Each of the two surveys was conducted in November, the first in 1997 and the second survey in 1999. The sample size of the 1997 survey was 4,824, but was increased through oversampling from several States. Thus, the total sample for the 1997 survey was 6,803. The actual sample sizes were 3,841 adults for the 1999 survey and 3,590 for the 2000 survey.

The data for all three surveys were based on a simple random sample of telephone numbers, and a randomly sampled adult was selected within each eligible household. The data were entered into a computer-assisted telephone interview (CATI) system and analysed using SPSS and SAS software.

The response rates were defined as the proportion of people who completed an interview divided by the total eligible to do so. The household response rates were 61 per cent in the 1997 survey, 65 per cent in 1999, and 76 per cent in 2000. Of the individuals eligible to be surveyed, the individual-level response rates were over 80 per cent for all three surveys. Response rates were similar among States and Territories, although rates in the ACT, South Australia and Tasmania were slightly higher than in other regions.

Measurements

The surveys used a questionnaire, which sought information about responders' participation in physical activity in the previous week, as well as questions about knowledge and understanding of the moderate intensity physical activity message, intention to be more active, and also recall of Active Australia messages and campaign themes (Armstrong et al. 2000).

The primary measures of physical activity were derived from questions that asked about participation in walking, vigorous activity and moderate activity in the previous week, and incorporated information on the total number of sessions during the week, and the total time in minutes. A definition of sufficient physical activity for health was developed, and was based on the accrual of at least 150 minutes of physical activity during the previous week. Minutes spent in vigorous activity were 'weighted' to count for twice those spent in moderate activity, as described in Armstrong et al. (2000). This is consistent with the National Physical Activity Guidelines for Australians (DHAC, 1999), and also with the United States Surgeon Generals Report on physical activity (USDHHS 1996).

Questions were asked about recall of any general messages about exercise or physical activity in the media, and specific message recall about two Active Australia campaigns. These were the 1998 campaign, ‘Exercise. You only have to take it regularly, not seriously’, and the 1999 International Year of Older People ‘Rusty’ campaign, which portrayed a tin man who shed his armour and became more active. Recall of these messages was asked in both unprompted and prompted ways, but the prompted recall is described in more detail in this chapter. The same questions were asked in 1997, 1999 and 2000, so trends in these perceptions and beliefs and behaviors were the focus of this analysis.

Weighting of the data

The data were weighted by age and sex, in proportion to the National population (Armstrong et al. 2000). For these analyses, weighted sample data were used, to provide National estimates. The data were down-weighted to an effective sample size of 2,500 or 3,000 for each survey. Although some over-sampling occurred in some States, this was not consistent across all surveys and, hence, State-level comparisons were not possible using these data. The limited State-level data reported are based on unweighted data. For the smaller States, such as ACT, South Australia, Tasmania and NT, the sample sizes were too small to provide reliable State-level or trend data, and caution should be used in interpreting State-level comparisons (State-level trends are summarised in the text box). For these reasons, primarily National weighted data are described.

Research questions

Five research questions are asked of these three National physical activity surveys.

1. Was there an increase in overall awareness of physical activity messages among adult Australians, and, specifically, greater awareness of physical activity campaign message 'Exercise. You only have to take it regularly, not seriously'?
2. Did people increase or change their understanding of the moderate activity message? (Responses to four knowledge statements about activity)
3. Did people change their intention to be more active?
4. Did physical activity change in terms of:
   1. reported minutes of activity in the previous week?
   2. reported proportion of the population 'sufficiently' active for health benefit?
   3. proportion inactive (completely sedentary)?
5. From a range of data sources, (including these surveys) what sub-groups of the population are least active?

Results

Message recall and understanding of the moderate physical activity message

Reported recall of any message about physical activity in the previous month was quite high - reported by nearly 64 per cent of adults in 1997 and 1999 and 66 per cent in 2000 (see Figure 2). Specific recall of the Active Australia tagline increased markedly from 14.4 per cent in 1997 to 42 per cent in 1999, and remained high in 2000.

Figure 2. Awareness of any physical activity mesage and Active Australia

Understanding of the moderate physical activity message

Understanding of the moderate physical activity message was assessed by responses to four statements asked at each survey. The questions were coded on a five-point Likert scale from 'strongly agree' to 'strongly disagree'. The proportions agreeing with the four statements are shown in Table 1.

Table 1. Agreement (per cent combined strongly agree and agree) with knowledge statements, 1997, 1999 and 2000 surveys

There was an increase in agreement with the first and fourth statement between 1997 and 1999, with no subsequent change to 2000. This suggests agreement with the notions of generally being more active and with ‘accumulating activity in bouts of at least ten minutes’ having a positive effect on health. The ‘half an hour per day’ question was generally well understood at all three surveys - over 90 per cent agreed with the statement. There was a slight favourable decline in the proportion agreeing with the ‘vigorous activity 3 x 20 essential’ statement - this decreased from 62 per cent in 1997 to 59.6 per cent in 2000.

Intention to be more active

Figure 3 shows the trends in the proportions that did not intend to be more active, or intended to do so in the next month or six months. There was no increase in intention to be more active between 1997 and 1999, and a small short-term (one month) increase in intention to be more active in 2000.

Figure 3. Intention to be more active (per cent agree) 1997-2000

Table 2. Any intention to be more active (per cent agree)

Overall, any intention to be more active is shown in Table 2. There were increases in intention for females (1997 to 1999), and for middle-aged adults (45-49 years, 1999 to 2000).

Participation in physical activity

The proportion of the population that met the criterion of sufficient activity for health benefit is shown in Table 3. There was a significant decline from 62 per cent in 1997 to 57 per cent in 1999 and 2000. This decline was noted for men and women, and for all age groups except older adults who showed a small (non- significant) increase in the proportion sufficiently active for health.

Table 3. Trends in ‘sufficient’ levels* of physical activity, 1997, 1999 and 2000 National samples (Australian estimates)

HSC, Higher School Certificate.

* 'Sufficient' time is 150 minutes total of all moderate walking, including vigorous physical activity minutes weighted by two, according to the method reported in Armstrong et al. (2000).

Education gradients showed that the greatest decline was noted for those with tertiary education, declining from 72 per cent in 1997 to 62 per cent in subsequent surveys, representing a 13 per cent decline in the proportion in this group achieving health-related levels of activity. Declines were seen in those with lower educational levels but of small magnitude (8.8% and 6.6% declines, respectively).

Table 4. People reporting no physical activity (sedentary), 1997, 1999 and 2000

HSC, Higher School Certificate.

Table 5 shows the mean number of minutes spent in walking, moderate activities, vigorous activities and vigorous gardening in the previous week across the three surveys. These data also show declines in each of these dimensions of physical activity between 1997 and 1999 among adult Australians. These measures showed very little change between 1999 and 2000, which indicates that reported activity, across all four dimensions, remained unchanged during this period.

Table 5. Total time for physical activity during previous week, trends over the two time periods 1999 and 2000 (showing mean minutes, 75th and 95th percentiles)

* Vigorous-intensity gardening and yard work.

State-level summary data from the first two surveys, in 1997 and 1999, are shown in the text box but no specific State-level estimates are provided because of the small samples sizes from some States and Territories. Generally, the trends were similar across States but the data suggest that increases in awareness were greater in NSW and the ACT than elsewhere in Australia. Declines in participation occurred everywhere but were less marked in NSW and WA.

Groups identified as being at greater risk of inactivity

Analyses of representative data from Australian adults have profiled several population groups that were less likely to be active (Armstrong et al. 2000; p 46). The purpose of identifying these groups is to understand the populations that are more likely to be inactive, so that programs can take account of these groups. Six factors were associated with inactivity:

• women were 20 per cent less likely to report leisure time physical activity (LTPA) than men;
• people in all age groups were less likely to be active than people aged 18-29;
• there was a middle-aged 'slump' with lowest rates of activity among adults in their 40s and 50s, compared to the immediate post-retirement group (aged 60-69) that was slightly more active;
• people who had attained tertiary education were more likely to report sufficient LTPA than people who were least educationally advantaged but there have been opposite trends in recent years - declines in activity among middle-aged people who are most educated - suggesting that work pressures and lack of time may be playing a part in the decline in activity levels of adult Australians;
• people who did not speak English at home were less likely to report activity, even after adjustment for other demographic differences - this is a generic difference because the sample within any specific ethnic group was too small to detect differences accurately; and
• married women, especially those with children, reported less LTPA than their age-matched single counterparts.

It is apparent from other research that Indigenous adults are less active (NSW 1996), although even large population surveys have included too few Indigenous people; the detailed National Aboriginal and Torres Strait Islander Health survey in 1994 did not collect physical activity data. Other determinants include social supports, perceived social environments for activity (Macdougall 1997), perceived barriers to being active (Owen 1992). The USSG report identified other factors that were associated consistently with activity, including the presence of social supports, enjoyment of the activity, and perceptions that activity was not vigorous (USDHHS 1996).

Other intra-individual factors influence activity; for example, situation-specific confidence (self-efficacy), and beliefs about the benefits of activity may contribute to an individual's likelihood of engaging in LTPA. There are also larger-scale economic and structural barriers, such as access to facilities, or being able to afford necessary equipment or programs, and other aspects of the social and physical environment are discussed further in chapter 2.6.

1.3.3. Conclusions

This chapter has identified various measures used to assess physical activity in Australian population surveys over recent decades. The main focus was to describe trends based upon recent population surveys of physical activity in Australia. Only one-half of the adult population achieves the recommended 150 minutes of at least moderate activity each week. This is a more attainable threshold than earlier recommendations for 'aerobic' levels of activity but large proportions of Australian adults do not achieve this level. The three identical and seasonally-matched surveys conducted in 1997, 1999 and 2000 showed that there were National increases in recall of the Active Australia message, and in some aspects of understanding the new moderate physical activity message, but declines in total participation in physical activity.

There were no seasonal or sample differences that might have explained this observation, and additional surveys are needed to confirm this trend. However, the interpretation is that at best, no change in activity participation has occurred. Moreover, the decline seems to have been smaller in NSW and the ACT, regions with stronger community Active Australia campaigns to increase community understanding about activity during 1998 and 1999.

Population data are also used to describe the sub-groups that are less active, so that equity-focused and targeted interventions and programs can be developed. Attention to these trend data, and to the groups identified for interventions, should be part of a best-practice approach to promoting physical activity.

References

Armstrong T, Bauman A, Davies J. (2000). Physical activity patterns of Australian adults. AIHW Catalogue CVD 10. Canberra: Australian Institute of Health and Welfare.

Bauman A, Owen N, Rushworth RL. (1990). Recent trends and socio-demographic determinants of exercise participation in Australia. Community Health Studies 14:19-26.

Bauman A, Bellew B, Booth M et al (1996). Towards best practice for the promotion of physical activity in the areas of NSW. NSW Health Department.

Bauman A, Brown W, Bellew B, et al. (1999). NSW 1996 physical activity survey: summary of major findings. NSW Health Department. www.health.nsw.gov.au/public-health/health-promotion.

Bull F, Milligan R, Rosenberg M, et al. (2000). Physical activity levels of Western Australians, 1999. Perth: Health Department of Western Australia.

Department of Human Services. (1999). South Australian Physical Activity Survey 1998. Adelaide: Department of Human Services, Government of South Australia.

Department of Health and Aged Care (DHAC). (1999). National Physical Activity Guidelines for Australians, Canberra.

Macdougall C, Cooke R, Owen N, et al. (1997). Relating physical activity to health status, social connections and community facilities. Australian and New Zealand Journal of Public Health 21:631-7

National Heart Foundation of Australia (NHF). (1989). Risk factor prevalence study No. 3. Canberra: NHF

NSW Health, Public Health Division (2000), The health of the people of New South Wales-Report of the Chief Health Officer. NSW Health Department: Sydney.

Owen N, Bauman A. (1992). The descriptive epidemiology of a sedentary lifestyle in Australia. International Journal of Epidemiology 21:305-10.

Smith J, Owen N, Leslie E, et al. (1999). Active for life physical activity patterns in Victoria 1998. Melbourne: Victorian Department of Human Services.

United States Department of Health and Human Services (USDHHS). (1996). Physical activity and health: a report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion.

Next: Part 1 (cont.) - 1.4. Towards best practice: evidence and policy implications

Previous: Part 1 - 1.2 The 'burden of disease' and the costs of physical inactivity in Australia

This is an archive copy of a document originally located at http://www.nphp.gov.au/sigpah/gaa/index.htm. All copyright remains with the creator.

The National Sport Information Centre Web Archive provides access to archive copies of materials in electronic forms. There is no inference that these materials necessarily reflect the current policy of any of the institutions or agencies that created them.