National physical activity recommendations for older Australians: Discussion Document

3. Physical activity patterns in older Australians

The National Ageing Research Institute was commissioned by The Department of Health and Ageing to review the evidence and develop physical activity recommendations for older people.

Page last updated: 01 February 2011

This Chapter reports on the prevalence rates for physical activity behaviour amongst older Australians, drawing upon the National Health Surveys and other population surveys and referring to patterns amongst the Aboriginal and Torres Strait Islander community and those from culturally and linguistically diverse backgrounds (CLDB). Comparative data from a number of other countries is also provided.

The proportion of Australians aged 65 years and older is set to increase from 13% (2001 figures) to around 18% by 2020 and by 2051, it could be as high as 29% (Australian Bureau of Statistics 2003). The absolute rates of sedentary behaviour increase with age. According to national data (Armstrong, Bauman et al. 2000), less than half of Australians aged 65 years and over do sufficient physical activity to produce health benefit (via the accumulation of 150 minutes or more of moderate and/or vigorous activity/week). Physical activity prevalence rates are lower in women than men. About a third of the ‘insufficiently active’ are completely sedentary (Bauman 2004). Adequate physical activity rates amongst those where English is a second language and older people with chronic conditions are lower than in the general population (Taylor, Baranowski et al. 1998). Given the absolute risk of ill health associated with inactivity in older people, there is much to gain at both an individual and societal level from older people being physically active.

Bauman et al. (2003) compared national physical activity levels in adults in 1997 and 1999 and reported a 6% decline in rates of those sufficiently active (63% to 57%). The authors suggested that this might reflect the decrease in available leisure time, with more Australians working, and working longer hours, during this period. The decline was greater than that reported for other countries e.g. United Kingdom (UK), New Zealand, United States of America (US), and the Netherlands (Hillsdon, Cavill et al. 2001; Bull, Bauman et al. 2004). The decline was smaller in New South Wales (NSW) and the Australian Capital Territory (ACT) where a specific media campaign and inter-agency collaboration had occurred. Recognition of the campaign message and knowledge about physical activity was also greater in NSW and ACT. The national decline was not observed amongst older adults. This is encouraging particularly given an older people targeted media campaign in NSW. However interpretation is limited by the sample's age cut-off of 75 years. The most recent data from the 2004-2005 National Health Survey confirms the prevailing trends reported in the literature: moderate and vigorous activity was most common in younger adults. Walking for exercise was most common (around 54%) in the younger old age groups, i.e. 55-64 and 65-74 years, but lowest in those aged 75 years and over (35.7%)(Australian Bureau of Statistics 2006). Compared to the 2001 figures, the proportion of those 65 years and above who are sedentary has increased. Table 3.1 shows the prevalence of sedentary behaviour in older Australians. For both men and women an increasing proportion of those 75 years and over are sedentary. Amongst women aged 75 years and over, the proportion categorised as engaged in moderate levels of physical activity has decreased from 17% in 1995 to 11.6% in 2004-2005 (figures not shown).

Table 3.1 The prevalence of sedentary behaviour in older Australians 1995 to 2005

Men (%)
Women (%)
65-74 years
75 and over
65-74 years
75 and over
Source: ABS National Health Survey Summary of Results 2004-2005 (Australian Bureau of Statistics 2006)

In 2001, Brownie surveyed over a thousand older Australians to determine their physical activity patterns (Brownie 2005). A questionnaire was mailed to a proportionately random sample, stratified by State and Territory from the 2000 Electoral Commission Roll. Responses were obtained from 1,263 people (641 males, 622 females, a response rate of 62%). The study has limitations: the author did not record intensity levels and excluded housework and gardening from the categorisation of activity frequency/duration. However, just over half reported doing at least 30 minutes exercise most days. Men were more likely to do so than women: they also reported a wider variety of activities. Individual characteristics (e.g., health status and motivation), influenced behaviour in the predicted direction. Sociocultural determinants of health were also predictive of physical activity level, but no regression analyses were conducted. Gym training, Tai chi and yoga were the least frequently reported activity types, with walking being the commonest form of activity reported. The findings were grouped into no activity; low (at least 30 minutes 1-2 times per week); medium (at least 30 minutes 3-4 times per week) and high (at least 30 minutes 5+ times per week). Twenty six percent fell into the ‘no activity’ category and just over half into the ‘high activity category’, with 10 and 11% in the low and medium categories respectively.

The latest Exercise, Recreation and Sport (ERASS) survey, conducted by the Australian Sports Commission, indicates that the majority of respondents aged 65 years and over were involved in organised and/or non-organised physical activity (71.6%), with similar figures in both men and women (72.5% and 70.8% respectively) (Standing Committee on Recreation and Sport - Australia 2005). Non-organised activity was the most common type of physical activity, but people did not necessarily participate in one type to the exclusion of the other. The absolute participation rates (any form of physical activity) decreased with age across the sample (from 91.7% in 15 to 24 year olds), but the proportions for organised versus non-organised physical activity were similar across all age groups. Walking was the commonest form of activity that older people participated in (47.6%), followed by aerobics/fitness programs (12.1%) and golf (8.6%). Older people participated in fewer types of activity (1.6 different types on average). Participation in organised physical activity was lowest amongst older people (30.7%, compared to 66.1% in the 15-24 year age group). The mean number of weekly sessions of activity was 4.6 (4.5 and 4.8 for men and women respectively), with a median of 3.9.

The findings from the Brownie and ERASS surveys contrast with those produced from the population based National Health Surveys. The figures are higher than in the multipurpose national surveys because they relate to the physical activity patterns of those responding to a specific physical activity survey, who are unlikely to representative of the overall older population. All these data sources are limited by the fact that they are based upon self reports in cross sectional surveys. Discussion of methodological issues such as this are beyond the scope of the current document. An overview of measurement considerations is provided in Appendix 4.
Top of page

Physical activity prevalence amongst Aboriginal and Torres Strait Islanders

The Aboriginal and Torres Strait Islander peoples’ prevalence of physical activity is low. Data from 2001 reported physical inactivity amongst 42% of Aboriginal and Torres Strait Islander women and 38% of Aboriginal and Torres Strait Islander men (AIHW 2001). The 2001 National Health Survey noted that, similar to non-Indigenous Australians 70% of the sample reported no or low levels of exercise in the fortnight prior to interview (ABS, 2002). The 2004-05 National Aboriginal and Torres Strait Islander Health Survey collected information about the frequency, intensity and duration of exercise undertaken by Indigenous Australians living in non-remote areas (Australian Bureau of Statistics 2006). The proportion of Indigenous people in non-remote areas who were sedentary or engaged in low level exercise in the two weeks prior to interview was higher than in 2001 (75% vs. 68% respectively) . About 60% of adult Aboriginal and Torres Strait Islander respondents to the Queensland Well Person’s Health Check Survey (1998-2000) had inadequate levels of physical activity in the preceding week (Miller, McDermott et al. 2002).

People from culturally and linguistically diverse backgrounds

In 1996, 17.8% of the older population were from CLDB, with a further 13.1% of the older population being overseas-born people from the main English-speaking countries (Gibson, Braun et al. 2001). The proportion of CLDB older people is projected to increase, reaching 22.8% by 2016. Amongst those aged 80 years and over, the proportion of people from CLDB was 13.2% in 1996 and is set to rise to 25.2% in 2026. Thus, one in four people aged 80 or over will be from CLDB in 20 years time. The older, overseas born CLDB population will increase more rapidly over this period than the Australian born older group.

People from culturally and linguistically diverse backgrounds (CLDB) are at greater risk for the health consequences of physical inactivity, given their higher rates of sedentary behaviour (Wilcox, Castro et al. 2000). There are cultural and ethnic differences in physical activity patterns. Unfortunately, much of the evidence emanates from the US and thus has limited relevance with regard to the different CLDB communities in Australia. Some of the more generic findings are salient and will be discussed in Chapter 5 when considering the determinants of physical activity behaviour in older people.

A study by Haralambous and colleagues (2003) examined rates of participation in physical activity amongst older people from four cultural backgrounds: Chinese, Croatian, Italian and Australian born. The groups held unique beliefs about physical activity participation. The Chinese respondents reported high levels of physical activity participation and a low incidence of functional limitations.
Top of page

Global comparisons

How do Australian trends compare with those of older populations internationally? This section gives a brief overview of recent statistics from overseas.

The 1998 Health Survey for England used a representative national sample and asked about frequency, duration and intensity of activities in the 4 weeks pre interview (UK Department of Health 2001). (The 2000 survey sampled only older people living in residential care). Respondents aged 65 years and over numbered 3270. As reported in other countries, walking was the most popular type of exercise, although 3.9% of over 65 year olds reported being unable to walk. Amongst those 85 years and over, this proportion increased to 13%. Whilst 88% of 65-69 year olds walked for at least 5 minutes in the previous 4 weeks, this prevalence decreased to 51.5% in the 85 years plus group. A significant age decline was also seen for walking at least 15 minutes: from 80.4% to 57.3%. This means that the proportion of older English residents who met the physical activity recommendations (at that time 150 minutes of moderate physical activity per week, based on the US Surgeon General’s report) was low and declined with age. Two-thirds participated in less than one 30-minute bout of physical activity per week and were categorised as inactive (UK Department of Health 2001).

The Zutphen Elderly Study in the Netherlands reported a stable five-year prevalence of sedentary behaviour of about 28% amongst their cohort (Bijnen, Caspersen et al. 1998). Total physical activity decreased for the majority of participants (75%) across their ten-year follow-up period. There was a 33% decrease in mean total physical activity, equivalent to 28 minutes per day.

Time-series cohorts with the North Karelia and Kuopio adult populations in East Finland highlight trends in physical activity behaviour over a 25 year period (Barengo, Nissinen et al. 2002). Prevalence of inactivity decreased, but low levels remained relatively stable at around 8%. Prevalence of high (intensity) leisure time physical activity has slightly increased for both men and women. In seeking to account for this, the authors note that working hours have not decreased. Occupational and commuting physical activity has decreased: this may have led to the increase in leisure time physical activity. The increase may also be due to greater health awareness and/or a cohort effect, with more recent cohorts having had greater exposure to physical activity opportunities. The study benefits from use of the same assessment tools at all time points, but the activity categories (leisure time, occupation and commuting) use different domains of physical activity volume.

In 2002/3 the International Physical Activity Questionnaire (IPAQ) was administered by telephone with several thousand Australian and New Zealand adults aged 18 to 65 years (McLean, Smith et al. 2005). Response rates of 55% and 42% respectively were achieved. Both populations appeared active, with 63.7% of New Zealanders categorised as sufficiently active, compared to 58.8% Australians. It was only for the 50-65 year old group that the New Zealanders were significantly more active than their Australian counterparts (63.2% vs. 50.3%). There were also gender differences, with women being less active in both countries. The majority (74.4%) of New Zealand men were sufficiently active compared to 65.1% of Australian men. There are societal and structural explanations for the trans-Tasman differences. New Zealanders reported fewer physical environment barriers. The long-term government initiatives (eg Push Play and Green Prescription) have targeted men and the older population. This is therefore level IV evidence that the New Zealand strategies are having some impact on physical activity behaviour.
Top of page
In Canada, around 50% of those 65 years and over are sedentary, with the prevalence rising to almost two thirds amongst those over 75 years (National Advisory Council on Ageing 2003). Statistics from the US suggest that older people remain the least active sector, with a greater proportion inactive than in Australia (US Department of Health and Human Services 2002). At the time of the landmark Surgeon General’s physical activity recommendations, 80% of over 65 year old US citizens were not sufficiently physically active (Pate, Pratt et al. 1995) and the figures have not changed markedly in the subsequent decade (Agency for Healthcare Research and Quality and the Centers for Disease Control 2002). In 2000, only 31% of Americans aged 65 to 74 years old and 23% of those 75 years and over reported participating in 20 minutes of moderate physical activity on three or more days per week (US Department of Health and Human Services 2000). Even fewer reported 30 minutes of moderate activity five or more days per week (16% and 12% of the respective aged groups). Recent US figures report an estimated 25% prevalence of inactivity across US adults (Centres for Disease Control 2004). Amongst older Americans, those aged 65-74 spent only 5.2% of their daily energy in leisure time physical activity and 35% on household related activity, with the figures being even lower in the 75+ group (Dong, Block et al. 2004). The United States Behavioural Risk Factor Surveillance System (BRFSS) data for 2004 shows that when asked ‘during the past month have you participated in any physical activity?’ 22.9% of respondents answered ‘no’. However, BRFSS data show that walking for leisure rates in those aged 65 and above increased (by 3.5%) from 1987 to 2000. Increases across a range of activities is needed. Dishman (1994) reported at around this time that 50% of these older people had no intention to begin regular exercise and the statistics would certainly seem to bear this out. Clearly, strategies to modify behaviour in this sector of the population have been limited in impact and there is much scope for improvement.

Table 3.2 Physical activity patterns in three Western countries

65-74 year olds
75+ year olds
Prevalence measure reported
no leisure time physical activity a
physical inactivity b
15 min brisk walking past 4 weeks c
sedentary d

Source: a) Healthy People 2010 (US Department of Health and Human Services 2000) b) Third NHANES data (Christmas and Andersen 2000) c) UK Department of Heath 2002 d) Interim report card: Seniors in Canada 2003 (National Advisory Council on Ageing 2003)


The proportion of older Australians being sufficiently active, that is accumulating 150 or more minutes per week of physical activity, remains low and has remained relatively stable in more recent surveys. There appears to be an increase in the proportion of those who are insufficiently active who are classified as sedentary. Although difficult to draw direct comparisons between international studies due to different definitions and methodologies, there is an indication that the proportion of older Australians classified as sedentary is slightly lower than countries such as Canada and the United States of America. Consistent with international studies, walking is the most frequent type of physical activity across the older age groups. There is a clear need for more detailed Australian physical activity data from representative samples.
Top of page

Document download

This publication is available as a downloadable document.

National physical activity recommendations for older Australians: Discussion Document(PDF 1160 KB)