A

wtiFctwataog©

K

1

acraoteu

1d

Available online at www.sciencedirect.com

Journal of Science and Medicine in Sport 13 (2010) 417–423

Original paper

Factors affecting walking activity of older people fromculturally diverse groups: An Australian experience

Stephen R. Bird a,∗, Harriet Radermacher b, Jane Sims b, Susan Feldman b,Colette Browning c, Shane Thomas d

a Health Innovations Research Institute (HIRi), School of Medical Sciences, RMIT University, Melbourne, Victoria 3083, Australiab Healthy Ageing Research Unit, Monash University, Melbourne, Australia

c Monash Institute of Health Services Research, Monash University, Melbourne, Australiad School of Primary Health Care, Monash University, Melbourne, Australia

Received 29 July 2008; received in revised form 31 May 2009; accepted 12 July 2009

bstract

This study sought to investigate the walking habits of older people from diverse cultural backgrounds, and to identify the factors associatedith their walking. Three hundred and thirty three people over the age of 60 years were recruited from seven culturally diverse groups from

he Western suburbs of Melbourne, Australia. A survey questionnaire recording physical activity, and various factors related to activity, wasnterviewer-administered in the participants’ preferred language. Data were analysed using Kruskal–Wallis, χ2 and Mann–Whitney tests.orty-seven percent of the participants walked at least 150 min per week, with no significant difference in prevalence between genders orultural groups. Some cultural differences were found in relation to reasons and locations for walking, and women were more likely than meno report walking in the shopping mall, whilst men were more likely than women to report walking in the park and along walking trails. Thoseho attained >150 min of walking were more likely to report health and fitness as reasons for walking, to perceive their walking environment

s more pleasurable, to use walking trails, and to consider their environment safe and to facilitate social interaction. This study indicates thathe continued advocating of walking as a health promoting activity should be central to future campaigns to increase physical activity in this

ge group. The provision of locations that are accessible, safe, aesthetically pleasing, and encourage social engagement are likely to facilitatelder people’s participation in walking. For maximum effectiveness, however, strategies may benefit from being tailored to meet specificender and cultural preferences.

2009 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

tivity

3fiafa

mo

eywords: Older adults; Cultural diversity; Ethnicity; Walking; Physical ac

. Introduction

Australia has a growing population of older people,1 andssociated with this is an increasing prevalence of chroniconditions such as coronary heart disease and diabetes.2 Cur-ent national and international guidelines recommend thatll adults should do 30 min of moderate intensity exercisen most, preferably all, days of the week.3,4 Adoption of

he recommended physical activity levels by older adults5 isxpected to reduce risk of disease and loss of function byp to 50%.4 However, in Australia only 51% of men and

∗ Corresponding author.E-mail address: stephen.bird@rmit.edu.au (S.R. Bird).

nthSmi

440-2440/$ – see front matter © 2009 Sports Medicine Australia. Published by Eloi:10.1016/j.jsams.2009.07.002

7% of women aged between 60 and 75 years are classi-ed as sufficiently active to attain health benefits from theirctivity.6 Furthermore, there are concerns about older peoplerom particular cultural groups, where physical activity ratesre exceptionally low, placing them at greater health risk.2,7

In Metropolitan Melbourne, there are 32 Local Govern-ent Authorities (LGAs). To promote physical activity for

lder people, a number of these LGAs have identified theeed to be proactive and implement physical activity initia-ives and strategies to alleviate the impending burden on the

ealthcare system as part of their ‘Ageing Well Strategy’.8

ince walking is reported to be the exercise of choice forany older people from different cultural backgrounds, and

s considered to be an activity that is relatively inexpensive,

sevier Ltd. All rights reserved.

4 and Me

appT

waaaecpeso

wgtdea

2

BAmot(m5pa(Iw3

acpGii

ttidnp‘

itwhCrea

t

•

•

fapptatqtrm

sgapim

18 S.R. Bird et al. / Journal of Science

ccessible, convenient, and therefore amenable to increasedarticipation,9 three LGAs wished to implement specificolicies to increase walking amongst their older residents.his study was funded to inform these future health policies.

To design effective strategies to increase participation inalking, it is important to understand the factors associ-

ted with walking,10,11 which can include personal, socialnd environmental factors.12,13 The objectives of this studyre framed within an ecological perspective13,14 whichmphasises the dynamic interaction between biological, psy-hological, social and environmental factors in influencingatterns of health, well-being and physical activity. Given thethnic and cultural demographics of the region in which thetudy was located, we needed to incorporate the perspectivesf people from culturally diverse backgrounds.

Specifically, the aims of the study were to investigate thealking habits of older people from diverse cultural back-rounds in the Western suburbs of Melbourne, Australia, ando identify the factors associated with their walking. The datarawn on here were collected as part of a broader study thatxamined all types of physical activity, the results of whichre reported elsewhere.15

. Methods

The study was conducted in three LGAs (Maribyrnong,rimbank, and Melton) in the Western region of Melbourne,ustralia. The Western region has a population of approxi-ately 420,000, 26% (110,629) of whom are over the age

f 60 years.16 The percentage of people aged over 60 inhe three LGAs is 24% (15,288), 23% (38,967), and 13%10,263) respectively.17 The Western region is one of the mostulticultural regions in Australia, with 43% (Maribyrnong),

4% (Brimbank), and 24% (Melton) of people in the threearticipating LGAs speaking languages other than Englisht home.17 The three LGAs also rank 2nd (Brimbank), 3rdMaribyrnong), and 10th (Melton) on the Socio-Economicndexes for Areas (where 1st indicates most disadvantaged)ithin the Melbourne Statistical Division which comprises1 LGA areas.18

Older people were recruited via local community groupsnd Home and Community Care (HACC) services. Theultural groups identified by the LGAs as being the mostopulous groups in the area were Anglo-Celtic, Croatian,

reek, Italian, Macedonian, Maltese and Vietnamese. For

nclusion into the study, participants had to be over 60 years,dentify with one of the targeted cultural groups,1 and live

1 LGA staff informed potential participants at the time of recruitment thathey were seeking participants from particular cultural groups. Please notehat several participants in this study identified as ‘Macedonian’. Macedonias a region in northern Greece. The Republic of Macedonia is an indepen-ent state which was admitted to the United Nations in 1993 under theame FYROM [Former Yugoslav Republic of Macedonia]. This means thateople from both Greece and FYROM could have identified themselves asMacedonian’, however this information was not clarified in this study.

oprewItvwfa

dicine in Sport 13 (2010) 417–423

ndependently in one of the three participating LGAs. Fur-her details regarding the recruitment strategy employed, asell as a critical analysis of the challenges of recruitment,ave been reported elsewhere.19 The Human Research Ethicsommittee at Victoria University, Melbourne, approved this

esearch project, and all participants were provided with anxplanatory statement in their preferred language and signedconsent form.

The survey instrument included established and interna-ionally validated instruments:

The International Physical Activity Questionnaire(IPAQ).20 This records physical activity within a 7-dayperiod under the categories of: walking, moderate inten-sity activity and vigorous activity. It was selected as thepreferred tool, as it has been widely used internationally,enabling comparisons with normative data. The minuteswalked per day were truncated at 180 in accordance withthe IPAQ recommendation. In order to avoid errors dueto over-reporting, any reported duration that exceeded28 h (1680 min) per week were truncated back to 28 h perweek.The Neighbourhood Environment Walkability Scale(NEWS).21 This includes questions about streets and facil-ities within the neighbourhood, safety and perceptions ofcrime. Items are scored on a 4 point Likert Scale.

The survey was administered in each participant’s pre-erred language, by experienced multilingual interviewers,t a location convenient to the participant. Interviews tooklace at a local cultural or community centre, or in thearticipant’s home. The survey took approximately 60 mino complete. The interviewers, who were fluent in Englishnd in their native language, attended a training sessionhat included a detailed explanation of the nuances of theuestions, in addition to instructions regarding the collec-ion, translation and recording of the data. Interviewers alsoeceived ongoing monitoring and supervision by the projectanager.Data were initially analysed using descriptive and chi-

quare (χ2) statistics to compare differences between culturalroups and gender (Table 1). Using the national guideline oft least 150 min physical activity a week as a benchmark,articipants were divided into those who reported achiev-ng: (i) less than 150 min walking per week, and (ii) 150 or

ore minutes walking per week. Additionally, since somef the questions included in our questionnaire required thearticipants to have undertaken some walking, such as thoseeferring to ‘perceptions’ whilst walking, it was deemed nec-ssary to exclude those who had completed no walking or noalking of a sustained duration. This was achieved using the

PAQ question that required the participants to state ‘whetherhey had walked for more than ten minutes at a time in the pre-

ious week’. And on this basis the <150 min per week groupere subdivided into those who reported not having walked

or at least 10 min at a time in the previous week (‘<10 minat and <150 min pw’), and those who did report walking

S.R. Bird et al. / Journal of Science and Medicine in Sport 13 (2010) 417–423 419

Table 1Basic demographics of sample and population (mean ± s.d.).

Anglo-Celtic Croatian Greek Italian Macedonian Maltese Vietnamese Total

Men (n) 15 8 21 13 27 18 34 136Women (n) 37 21 21 40 17 31 30 197Age (years) 76 ± 8a 73 ± 5 70 ± 6 72 ± 6 70 ± 5b 73 ± 7 72 ± 6 72 ± 7Length of time in

Australia (years)61 ± 21c 38 ± 13 44 ± 8 47 ± 8 34 ± 9 48 ± 9 17 ± 4d 41 ± 18

Population aged 65 yearsand over in Westernregione (no./%)

20,872 (39%) 1,620 (3%) 2,969 (6%) 6,831 (13%) 873 (2%) 2,930 (5%) 1,688 (3%) 53,403 (100%)

a The Anglo-Celtic group were older than all the others except the Croatians (p < 0.05).b (p < 0.0

an the oe other006): A

fatda1p‘awnwsgcuaaK

3

ivt3r

agdg

Ktau71au(c1I(pptp

po

TD

C

T

The Macedonians were younger than all other groups except the Greeksc The Anglo-Celtic group had been in Australia for significantly longer thd The Vietnamese had been in Australia for significantly less time than the Percentages are derived from country of birth data in the ABS Census (2

or 10 or minutes at a time in the previous week (‘≥10 minat but <150 min pw’). This also had the effect of dividinghe <150 min walking per week group into 2 groups that alsoiffered in their total volume of walking. With the ‘<10 minat and <150 min pw’ group reporting a mean of less thanmin (range 0–10 min) and the ‘≥10 min aat but <150 minw’ reporting a mean of 79 min (range 15–140 min). The<10 min aat and <150 min pw’ group were not included in thenalyses of questions that required the participation in somealking outside of the home environment, as it was deemedot applicable for them. All those who attained ≥150 min aeek reported walking for at least 10 min on at least 1 occa-

ion in the previous week. The cultural composition of theseroups is presented in Table 2. Associations between gender,ultural group and walking activity level were investigatedsing χ2 analysis. Perceptions of neighbourhood data weressessed for normality using Kolmogorov–Smirnov tests, ands not normally distributed, groups were compared usingruskal–Wallis and Mann–Whitney tests.

. Results

Of the 408 older people who initially expressed an interestn participating and signed a consent form, 362 were inter-

iewed. Of these, 29 were excluded from the analyses, dueo their not meeting the inclusion criteria. This resulted in33 participants, (137 men and 196 women) whose data areeported here. The mean age of participants was 72 ± 7 years,

5wt4

able 2istribution of participants in each cultural group according to time walked per we

‘<10 min aat and <150 min pw’

ultural group Croatian 8Greek 9Macedonian 0Maltese 18Italian 14Vietnamese 6Anglo-Celtic 10

otal 65

5).ther groups (p < 0.001).

groups (p < 0.001).nglo-Celtic group comprises those stating Great Britain or Australia.

nd the number of participants in each of the seven culturalroups ranged from 29 to 64 (Table 2). The age of participantsid not differ between men and women within each culturalroup.

Minutes walked per week were assessed using aolmogorov–Smirnov test, and shown not to be normally dis-

ributed (p < 0.001). The data below are therefore describeds medians and inter-quartile ranges (IQR). The median min-tes of walking per week for the whole group (n = 333) was0, (IQR 0–280 min), with 47% of respondents achieving50 min or more per week, in line with the national physicalctivity guidelines.3 There was no gender difference in min-tes walked per week: males 60 (0–300); females 90 (0–300),Mann–Whitney test, p = 0.333). Nor were there signifi-ant differences between the cultural groups: Anglo-Celtic20 (15–300); Croatian 90 (0–110); Greek 150 (40–420);talian 140 (0–525); Macedonian 90 (60–250); Maltese 700–300); Vietnamese 140 (60–280); Kruskal–Wallis test,= 0.225. There were no significant differences in theroportion of participants achieving >150 min across cul-ural groups (χ2 = 5.60, p = 0.470) or gender (χ2 = 0.14,= 0.738).

The following analysis is based on the data from 268articipants who reported walking for at least 10 min atne time during the last seven days. Of these participants,

7% walked to stay healthy and keep fit, and this reasonas more commonly reported by the ‘≥150 min pw’ group

han the ‘≥10 min aat but <150 min pw’ group (65% vs6%, χ2 = 9.78, p = 0.002). Thirty-one percent of those who

ek (χ2 for all participants p < 0.001).

‘≥10 min aat but <150 min pw’ ‘≥150 min pw’ Total

10 11 2910 23 4228 16 4411 20 4912 27 5325 33 6417 25 52

113 155 333

4 and Me

rowmχ

fbppwbχ

ma(qwg((

Aaifiipg<t(twwadrasfttnt

tpmχ

ipbnr

lip(

a‘touttppgtχ

mm<‘

iih(attdwts

wpwp

wgmmha

frWg<

20 S.R. Bird et al. / Journal of Science

eported walking more than 10 min at a time in the previ-us week listed walking to or around the shops as a reason,ith the ‘≥10 min aat but <150 min pw’ group reporting itore commonly than the ‘≥150 min pw’ group (44% vs 21%,2 = 17.15, p = 0.000). Thirty-one percent reported walking

or recreation or pleasure or to socialise, and this did not differetween the ‘≥10 min aat but <150 min pw’ and ‘≥150 minw’ groups (31% vs 30%, χ2 = 0.05, p = 0.820). Thirteenercent reported housework or gardening as a reason foralking, again with no difference between the ‘≥10 min aatut <150 min pw’ and ‘≥150 min pw’ groups (11% vs 14%,2 = 0.75, p = 0.385).

Thirty-seven percent of participants who reported walkingore than 10 min (both the ‘≥10 min aat but <150 min pw’

nd ‘≥150 min pw’ groups) said that the natural environmente.g., gardens, trees, parks, flowers, and general peace anduiet) was what they most liked about the places where theyalked. Other key factors included the opportunities walkingave them to engage with friends, neighbours and family29%), and the general neighbourhood environment (25%)e.g., clean streets, looking at houses).

Cultural groups differed in their reasons walking.mongst those who reported walking for more than 10 min

t a time in the previous week, there were cultural differencesn the prevalence of stating ‘health and fitness’ as a reasonor walking (χ2 = 31.37, p < 0.001). However, when subdivid-ng the participants this difference only reached significancen the ‘≥10 min aat but <150 min pw’ group (χ2 = 28.49,< 0.001), and was not significant in the ‘≥150 min pw’roup (χ2 = 10.65, p = 0.1). Within the ‘≥10 min aat but150 min pw’ group, the Italians (92%) were most likely

o report health as a reason for walking, and the Croatian0%) and Macedonians (21%) least likely. Significant cul-ural group differences were also found in the proportion whoalked to shop (χ2 = 53.75; p < 0.0001), the proportion whoalked for recreation and pleasure (χ2 = 17.79; p = 0.007),

nd the proportion who walked around their house and gar-en (χ2 = 17.75; p = 0.007). The Italians were most likely toeport walking to stay fit and healthy than any other group,nd they were least likely to walk for recreation and plea-ure. The Macedonians were most likely to report walkingor shopping, recreation and pleasure. The Croatians werehe least likely to report walking to stay fit and healthy, andhe Greeks were least likely to walk for shopping. The Viet-amese group were the most likely to report walking aroundhe house or garden.

For those who reported walking at least 10 min at aime in the previous week (‘≥10 min aat but <150 minw’ and ‘≥150 min pw’ groups combined), women wereore likely than men to walk for shopping (36% vs 24%,2 = 4.13; p = 0.042). When the group was split, this signif-

cant gender difference was still evident in the ‘≥150 min

w’ group (χ2 = 6.66; p = 0.01) but not in the ‘≥10 min aatut <150 min pw’ group (χ2 = 0.23; p = 0.635). There wereo gender differences in prevalence of walking for health,ecreation/pleasure or housework and gardening.

ptn“

dicine in Sport 13 (2010) 417–423

Three hundred (90%) of the 333 participants reported ateast one place in which they walked most of the time. Thesencluded neighbourhood streets (71%), shopping mall (58%),ark (37%), walking trails (17%), ‘other’ places e.g., at home12%), and the treadmill (5%).

When the three levels of walking activity (‘<10 min aatnd <150 min pw’, ‘≥10 min aat but <150 min pw’, and≥150 min pw’) were compared for their use of walking loca-ions, they cited the shopping mall with a similar prevalencef 54%, 54% and 62% respectively (χ2 = 2.04; p = 0.360). Notnexpectedly the ‘<10 min aat and <150 min pw’ group citedhe park as a place where they walked far less prevalentlyhan either the ‘≥10 min aat but <150 min pw’ or ‘≥150 minw’ groups (14%, 41% and 39% respectively, χ2 = 8.84;= 0.012). Similarly the ‘<10 min aat and <150 min pw’roup cited their neighbourhood streets far less prevalentlyhan the other two groups (37%, 74% and 76% respectively,2 = 21.68; p < 0.0001). Additionally the group who walkedost (‘≥150 min pw’) stated that they used walking trailsore often than the other two groups (‘<10 min aat and150 min pw’ 11%, ‘≥10 min aat but <150 min pw’ 7% and≥150 min pw’ 25%, χ2 = 15.46; p < 0.001).

Cultural groups differed in the places they reported walk-ng most of the time. Significant group differences were foundn the proportion who walked most of the time in neighbour-ood streets (χ2 = 22.30; p < 0.001), in the shopping mallχ2 = 33.90; p < 0.001), in the park (χ2 = 48.58; p < 0.001),nd on a treadmill (χ2 = 32.40; p < 0.001). Maltese and Croa-ian participants were least likely to report walking most ofhe time in neighbourhood streets, while the Italian and Mace-onian participants were most likely. The Italian participantsere also most likely to report walking on a treadmill, and

he Macedonians were least likely to report walking in thehopping mall.

Women (68%) were more likely than men (44%) to reportalking most of the time in the shopping mall (χ2 = 17.01;< 0.001), whilst men were more likely than women to reportalking in the park (men 44%, women 32%: χ2 = 4.26;= 0.039).

Interestingly, only 33% of all participants reported usingalking trails and these were predominantly in the >150 minroup. Men (45%) were significantly (χ2 = 14.12; p < 0.001)ore likely to use walking trails than women (25%). Theost common reason for not using walking trails were poor

ealth, trail location (too far away and/or isolated location),nd lack of awareness about trail locations.

Overall, 62% of the 268 participants who reported walkingor at least 10 min at one time during the last seven days,eported feeling extremely safe or quite safe while walking.

hen the perceptions of safety were compared across the 3roups according to the amount of walking (‘<10 min aat and150 min pw’, ‘≥10 min aat but <150 min pw’, ‘≥150 min

w’) the ‘<10 min aat and <150 min pw’ group showed arend to be more concerned about safety and crime in theireighbourhood, with them more frequently perceiving it asunsafe to walk at night (χ2 = 4.75; p = 0.093), and “having

and Me

aattp

wCM(s

ils<‘tci(pt(p

4

wwaptrnctb

malrwasw(Mfgthi

cbtl

psetbapftwwFepkao

<whaiwt<pwsgpp

(pwelwgtgthco

S.R. Bird et al. / Journal of Science

high crime rate” (χ2 = 5.07; p = 0.079). Furthermore, whensked about the safety of their neighbourhood during the day,he ‘≥150 min pw’ group showed a trend to perceive it as saferhan the ‘≥10 min aat but <150 min pw’ group (χ2 = 3.19;= 0.074).

There were significant differences in perceptions of safetyhile walking between cultural groups (Anglo-Celtic 81%;roatian 67%; Greek 42%; Italian 67%; Macedonian 76%;altese 69%; Vietnamese 67%: χ2 = 14.14; p = 0.028). Men

78%) were more likely than women (60%) to report feelingafe while out walking (χ2 = 9.23; p = 0.002).

When comparing the <150 min and ≥150 min walk-ng groups, the group that walked ≥150 min was moreikely to perceive that “there were many attractive naturalights in their neighbourhood” (Mann–Whitney Mean ranks:150 min = 157, ≥150 min = 177; p = 0.048). And when the≥10 min aat but <150 min pw’ group was compared withhe ‘≥150 min pw’ group, the ‘≥150 min pw’ group moreommonly reported that “people in their homes can eas-ly see walkers and cyclists on the streets and paths”Kruskal–Wallis Mean ranks: ‘≥10 min aat but <150 minw’ = 123, ‘≥150 min pw’ = 140; p = 0.049), as well as arend to report “the streets as being well lit at night”Kruskal–Wallis Mean ranks: ‘≥10 min aat but <150 minw’ = 125, ‘≥150 min pw’ = 140; p = 0.095).

. Discussion

The majority of participants (80%) in this study reportedalking at least 10 min at one time in the previous week,ith 47% of respondents achieving the national physical

ctivity guidelines’22 recommendation of 150 min or moreer week, solely through walking. This proportion reflectshe amount of overall physical activity amongst older peopleeported in national surveys.2,7 There was no statistically sig-ificant difference in the amount of walking reported acrossultural groups, which confirms findings from other studieshat walking is the exercise of choice regardless of culturalackground.11,23

It was noticeable that the group who reported walkingore than 150 min a week was more likely to report health

nd fitness as a reason for walking than those who attainedess than 150 min. Whilst this finding does not prove a causalelationship, the association may suggest that promotingalking for health has had a positive impact on participation,

nd one that should be further promoted. From a cultural per-pective it is also of interest to note that within the group thatas undertaking some walking, but not attaining 150 min

‘≥10 min aat but <150 min pw’ group), the Croatians andacedonians were less likely to report health as a reason

or walking. While the small number of participants in these

roups is acknowledged, the finding may still suggest poten-ial benefits of specifically targeting particular groups with theealth message. Although, it must be noted that some researchndicates that actual knowledge of health and exercise is not

amsw

dicine in Sport 13 (2010) 417–423 421

onsistently found to be associated with physical activityehaviour.24 If this is the case then other socio-cultural fac-ors, not explored in this study, may have contributed to theack of participation in walking.

The second most common reason for walking was shop-ing, and for many of the participants it represented aubstantial proportion of their weekly walking activity. Forxample, some (∼35%) of the ‘≥150 min pw’ group attainedhe advocated >150 min without citing health as a reason,ut typically indicated a combination of shopping, recreationnd pleasure. However, since the ‘≥10 min aat but <150 minw’ group were more likely to report shopping as a reasonor walking, but less likely to report health and fitness, thanhe ‘≥150 min pw’ group, it may be surmised that whilstalking while shopping should be encouraged, on its own itas commonly not sufficient to attain the target of >150 min.uture strategies may therefore need to consider putting anmphasis on making healthy lifestyle choices, such as pur-oseful walking for its own sake, and opting to walk in thenowledge of the health benefits when having a choice, if thedvocated >150 min is to be attained by a greater proportionf the population.

For those who did very little walking (‘<10 min aat and150 min pw’), a fear of crime and concerns about safetyere issues. These findings concur with other studies thatave identified personal safety as a factor influencing physicalctivity.11,25,26 However there were no apparent differencesn the prevalence of these concerns between those who didalk for more than 10 min at a time, regardless of whether

hey attained 150 min or not in a week (‘≥10 min aat but150 min pw’ and ‘≥150 min pw’). This suggests that it wasrimarily an issue for those who did little walking. Howeverhether the association was causal, and whether allaying

uch fears would promote more walking in this inactiveroup is unknown. Furthermore since these fears were morerevalent in women and the Greek participants, it indicates aossible need for strategies to target particular groups.

With respect to walking locations, the level of walking‘<10 min aat and <150 min pw’, ‘≥10 min aat but <150 minw’, and ‘≥150 min pw’) did not affect the prevalence ofalking in the local shopping mall. However, not unexpect-

dly the ‘<10 min aat and <150 min pw’ group were lessikely to walk in the park or their local streets than thoseho walked more than 10 min at a time. Interestingly, theroup who walked for more than 150 min were more likelyo use walking trails than either of the other two groups. Thisroup’s increased use of these facilities may be linked toheir undertaking walking for the purpose of promoting theirealth and fitness, which as previously stated, was a moreommon reason for walking in this group compared with thethers.

Differences were also found in reported walking locations

cross gender and cultural groups. Men reported walkingore often in the park and on trails, and not surprisingly,

ince women were more likely to cite shopping as a reason foralking, women cited the shopping mall as a walking location

4 and Me

mfecawwi

rahtatimtstg

cnrtlttmtwgaips

5

stawistwttiatR

tt

A

sMrtfGv

ibfarow

R

22 S.R. Bird et al. / Journal of Science

ore prevalently than men. Cultural group differences wereound across all walking locations. These findings providevidence to support the important consideration and signifi-ance of location in developing strategies to increase walkingctivity. For example, if women are more likely to reportalking to shop, then strategies that explicitly encourageomen to walk further during this activity, may be effective

n enabling them to attain the recommended amount.In summary, the findings indicate that the locations and

easons for walking varied by level of walking activity, gendernd cultural group. This warrants further attention, as it mayave implications for the choice and effectiveness of interven-ions to increase walking behaviour. Being properly informedbout the benefits of exercise and creating more opportuni-ies for safe walking, such as walking groups, and linkingt to activities such as shopping, may be effective in pro-

oting walking activity.27,28 However, this study indicateshat whilst a generalist approach may have some benefits,trategies may be more effective when specifically targeted tohe needs, perceptions and preferences of particular culturalroups, each gender and current level of physical activity.

Whilst the investigators endeavoured to recruit from aross-section of the population by working with commu-ity groups and HACC services, it is inevitable that someecruitment bias will occur. Similarly, voluntary participa-ion may have the same effect. Consequently some of theeast active, and least engaged members of society are likelyo be missed by the recruitment process. The choice of theools used represents a compromise between being able to

ake some comparison with international data, and limita-ions with respect to proven reliability and validity when usedith older people from a variety of ethnic and language back-rounds. The use of more accurate assessments of walkingctivity, such as accelerometers, would have required a morenvasive and time consuming commitment on the part of thearticipants and researchers, which would have limited theample size, but may be considered in future studies.

. Conclusions

This study indicates that a location that is accessible,afe and encourages social engagement with others is likelyo facilitate older people’s participation in walking. Thesere in accordance with the themes related to adherence thatere identified in Chiang et al.’s29 study of physical activ-

ty participation in ethnic groups, which included: location,ocializing and support between participants, and percep-ions of health benefits. Based on these findings, setting upalking groups in a local shopping mall may provide a poten-

ially effective strategy, to enable some older people to attainhe recommended levels of physical activity for maintain-

ng good health. However, based on the differences foundcross gender and cultural groups in this study, it is unlikelyhat a one-size fits all approach will be an effective strategy.ather, it may be necessary to develop tailored approaches

dicine in Sport 13 (2010) 417–423

hat are sensitive to particular preferences for why and whereo walk.

Practical implications

• Increased participation in walking could beachieved through targeting specific culturalgroups, where there may be a lack of aware-ness of the health benefits of regular walking.

• Walking in a social environment may pro-mote compliance, through enjoyment andsocial commitment.

• Group walking may also increase participa-tion through alleviating concerns regardingfear of injury and general safety.

• Policy makers, program developers and plan-ners must be aware of the potential groupdifferences and adapt strategies accordingly,since a ‘one size fits all approach’ may notbe most efficacious. Tailoring strategies thatare sensitive to different groups’ preferredwalking locations, reasons for walking, andperception of safety may enhance walkingactivity.

cknowledgements

The authors gratefully acknowledge the funding andupport provided by VicHealth, Brimbank, Melton and

aribyrnong Aged Care Services for this project. Theesearchers would also like to acknowledge the valuable con-ribution of all the participants, interviewers and communityacilitators, particularly Cuc Lam, Karolina Trigueros, Mariaarzo, Simone Cremona, Anita Milicevic and Ordan Andree-iski.

Disclosures: The authors have no competing interestsn relation to this study and its findings. VicHealth, Brim-ank, Melton and Maribyrnong Aged Care Services jointlyunded this study. Representatives from Brimbank, Meltonnd Maribyrnong Aged Care Services were part of theesearch team, acting as collaborators and advisors through-ut the study’s duration. These representatives also assistedith participant recruitment.

eferences

1. Australian Institute of Health and Welfare (AIHW). Australia’s welfare2007. Cat. no. AUS 93. Canberra: AIHW; 2007.

2. Australian Bureau of Statistics (ABS). National Health Survey: sum-mary of results 2004–2005. Cat. no. 4364.0. Canberra: AustralianBureau of Statistics; 2006.

and Me

S.R. Bird et al. / Journal of Science

3. Department of Health and Aged Care (DHAC). National physical activ-ity guidelines for Australians. Canberra: Commonwealth Departmentof Health and Aged Care; 1999.

4. Paterson D, Jones G, Rice C. Ageing and physical activity: evidence todevelop exercise recommendations for older adults. Appl Physiol NutrMetab 2007;32:S69–108, doi:10.1139/H07-111.

5. Sims J, Hill K, Hunt S, Haralambous B, Brown A, Engel L, et al.National physical activity recommendations for older Australians: dis-cussion document. A report to the department of health and ageing.Melbourne: National Ageing Research Institute; 2006.

6. Armstrong T, Bauman A, Davies J. Physical activity patterns of Aus-tralian adults: results of the 1999 national physical activity survey.Canberra: Australian Institute of Health and Welfare; 2000. AIHW Cat.no. CVD 10.

7. Bauman A, Ford I, Armstrong T. Trends in population levels of reportedphysical activity in Australia, 1997, 1999 and 2000. Canberra: Aus-tralian Sports Commission; 2001.

8. Jenny Ashby and Associates Pty Ltd. Ageing well strategy 2004–2016.Victoria, Australia: City of Maribyrnong, City of Brimbank: Shire ofMelton; 2004.

9. Hardman AE, Stensel D. Physical activity and health: the evidenceexplained. London: Routledge; 2003.

10. Dunton G, Schneider M. Perceived barriers to walking for physicalactivity. Prev Chronic Dis 2006:A116.

11. Belza B, Walwick J, Shiu-Thornton S, Schwartz S, Taylor M, LoGerfoJ. Older adult perspectives on physical activity and exercise: voicesfrom multiple cultures. Prev Chronic Dis 2004;1(4):A09.

12. King AC, Castro C, Wilcox S, Eyler AA, Sallis JF, Brownson RC.Personal and environmental factors associated with physical inactivityamong different racial–ethnic groups of U.S. middle-aged and older-aged women. Health Psychol 2000;19(4):354–64.

13. Sallis J, Owen N. Physical activity and behavioral medicine. ThousandOaks, CA: Sage; 1999.

14. Sallis J, Bauman A, Pratt M. Environmental and policy interventionsto promote physical activity. Am J Prev Med 1998;15(4):379–97.

15. Bird S, Radermacher H, Feldman S, Sims J, Kurowski W, Browning

C, et al. Factors influencing the physical activity levels of older peoplefrom culturally diverse communities: an Australian experience. AgeingSoc; in press.

16. Australian Bureau of Statistics. 2006 census community profile series:western Melbourne (statistical subdivision). Canberra: ABS; 2006.

dicine in Sport 13 (2010) 417–423 423

17. Australian Bureau of Statistics. 2006 census of population and housing.Canberra: ABS; 2006.

18. Australian Bureau of Statistics. Census of Population and Housing:Socio-Economic Indexes for Areas (SEIFA). Cat. no. 2033.0.55.001.Canberra: Commonwealth of Australia; 2006.

19. Feldman S, Radermacher H, Bird S, Browning C, Thomas S. Challengesof recruitment and retention of older people from culturally diversebackgrounds in research. Ageing Soc 2008;28:473–93.

20. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML,Ainsworth BE, et al. International Physical Activity Question-naire: 12-country reliability and validity. Med Sci Sports Exerc2003;35(8):1381–95.

21. Saelens BE, Sallis JF, Black JB, Chen D. Neighborhood-based differ-ences in physical activity: an environment scale evaluation. Am J PublicHealth 2003;93(9):1552–8.

22. Fiatrone Singh M. Exercise in the oldest old: some new insightsand unanswered questions. J Am Geriatrics Soc 2002;50(12):2089–91.

23. Giles-Corti B, Donovan R. Relative influences of individual, socialenvironmental, and physical environmental correlates of walking. AmJ Public Health 2003;93(9):1583–9.

24. Trost SG, Owen N, Bauman AE, Sallis JF, Brown W. Correlates ofadults’ participation in physical activity: review and update. Med SciSports Exerc 2002;34(12):1996–2001.

25. Li F, Fisher KJ, Brownson RC, Bosworth M. Multilevel mod-elling of built environment characteristics related to neighbourhoodwalking activity in older adults. J Epidemiol Community Health2005;59(7):558–64.

26. Dawson J, Hillsdon M, Boller I, Foster C. Perceived barriers to walk-ing in the neighbourhood environment and change in physical activitylevels over 12 months. Br J Sports Med 2007, bjsm.2006.033340.

27. Addy CL, Wilson DK, Kirtland KA, Ainsworth BE. Associationsof perceived social and physical environmental supports with phys-ical activity and walking behavior. Am J Public Health 2004;94(3):440–3.

28. Michael Y, Beard T, Dongseok C, Farquhar S, Carlson N. Measuring

the influence of built neighborhood environments on walking in olderadults. J Aging Phys Activity 2006–2007;14(3):302–12.

29. Chiang K, Seman L, Belza B, Tsai J. “It is our exercise family”: expere-inces of ethnic older adults in a group-based exercise program. PrevChronic Dis 2008;5:1–12.