Short and long-term adherence to swimming and walking programs in older
women — The Sedentary Women Exercise Adherence Trial (SWEAT 2)
Kay L. Coxa,d,⁎, Valerie Burkea,d, Lawrence J. Beilina,d, Amanda J. Derbyshirea,d,
J. Robert Groveb, Brian A. Blanksbyb, Ian B. Puddeya,c,d
aSchool of Medicine and Pharmacology, University of Western Australia, Australia
bSchool of Human Movement and Exercise Science, University of Western Australia, Australia
cFaculty of Medicine, Dentistry and Health Sciences, University of Western Australia, Australia
dWestern Australian Institute for Medical Research, Perth, WA, Australia
Available online 26 January 2008
Objective. To examine in previously sedentary older women the effects of exercise mode and a behavioural intervention on short and long-term
retention and adherence.
Methods. Healthy, sedentary women aged 50–70 years (N=116) were randomly assigned to a supervised 6-month swimming or walking
program 3 sessions a week. They were further randomised to usual care or a behavioural intervention. The same program was further continued
unsupervised for 6 months. We assessed retention, adherence, stage of exercise behaviour and changes in fitness.
Results. One hundred women (86%) completed 6 months and 86 (74%) continued for 12 months. Retention rates were similar for both exercise
modes at 6 and 12 months. Adherence to swimming or walking was similar after 6 months (76.3 (95% CI: 69.5, 83.1)% vs. 74.3 (67.7, 80.9)%)
and 12 months (65.8 (57.9, 73.8)% vs. 62.2 (54.6, 70.0)%). The behavioural intervention did not enhance retention or adherence. Fitness improved
for both exercise modes after 6 months and was maintained at 12 months.
Conclusions. Either swimming or walking programs initiated with careful supervision over 6 months resulted in similar high retention and
adherence rates by highly motivated older women over 12 months. Behavioural intervention in this setting did not improve these rates further.
© 2008 Elsevier Inc. All rights reserved.
Keywords: Swimming; Walking; Adherence; Women; Physical activity; Behavioural intervention
Sedentary behaviour increases with age (Owen and Bauman,
1992) and is a major risk factor for disorders including heart
disease, obesity and diabetes (U.S. Department of Health and
Human Services, 1996). In developed countries 15% of the
population is at least 65 years old (McMurdo, 2000) and in
Australia the proportion is expected to reach 25% by 2051
(Australian Bureau of Statistics, 2005). The resulting greater
prevalence ofdisorders associatedwithinactivity makesinitiation
and maintenance of physical activity in older adults a priority.
Avaried exercise program may promote adherence (Pollock
et al., 1991). Swimming is popular in Australia (Booth et al.,
1995) but participation decreases with age (McCormack et al.,
2003). Whilst swimming may encourage activity by providing
an alternative to walking, cost and accessibility are potential
barriers, particularly in older individuals.
In reviewing physical activity programs for older adults, van
der Bij et al. (2002) found adherence to be unrelated to the type
and frequency of activity, although no study used swimming. To
date as far as known, the effectiveness of different modes of
exercise in promoting physical activity has not been adequately
Behavioural interventions targeting exercise have often lack-
eda theoretical base, their scientific rigor has varied (Pinto etal.,
1996), and effects on adherence are inconclusive (van der Bij
et al., 2002). However, behavioural interventions based on the
Available online at www.sciencedirect.com
Preventive Medicine 46 (2008) 511–517
⁎Corresponding author. University of Western Australia, School of Medicine
and Pharmacology, Royal Perth Hospital Unit, GPO Box X2213 Perth, Western
Australia 6001, Australia. Fax: +61 08 9224 0246.
E-mail address: email@example.com (K.L. Cox).
0091-7435/$ - see front matter © 2008 Elsevier Inc. All rights reserved.
Transtheoretical Model (TTM) or Stages of Change Model have
been efficacious (Marcus et al., 1992, 1994). Previously, a be-
havioural intervention based on the TTM was developed and
utilised, but the study design prevented evaluation relative to
‘usual care’ (Cox et al., 2003). In the current study this inter-
vention was compared with ‘usual care’ using 6-months super-
vised centre-based initiation followed by 6-months unsupervised
moderate-intensity exercise (Cox et al., 2003). It was hypothe-
sized that,inthe short andlonger-term,a walking program would
and that the behavioural intervention would enhance adherence.
Healthy, sedentary women aged 50–70 years were recruited from the com-
munity by research assistants via media advertising. Fig. 1 shows the flow of
participants through the study. Entry criteria included being non-smokers, sed-
entary (less than 30 min/week of moderate activity) and having BMIb34 kg/m2
(Cox et al., 2006). Exclusion criteria included being unable to float in deep
water, musculo-skeletal disorders, mental incapacity, difficulty in communicat-
ing in English, diabetes, cardiac, respiratory or other chronic illness. During
1999–2001, 116 eligible women entered a 6-week run-in period. After all
measurements were completed and at the end of the run-in women were randomly
assigned via computer-generated random numbers (by the statistician) to a closely
supervised moderate walking or swimming program for an initial 6 months and
were further randomly assigned to a behavioural intervention (SWEAT) or ‘usual
in blocks of 8. After 6 months participants further continued the same exercise
program unsupervised for 6 months. In this type of study it was not practical to
blind the participants or the research staff to the group assignment.
The study was approved by the University of Western Australia Committee
for Human Rights. All participants gave informed written consent.
Assessments were made at baseline, 6 and 12 months. Exercise stage was
assessed using the Stage of Change Instrument (SCI) (Marcus et al., 1992)
exercise history was elicited by questionnaire. Fitness was measured using two
sub-maximal field tests (Cox et al., 2006), the 1.6 km walk(Kline et al., 1987)
and the 12-minute swim (Cooper, 1977). Illness and injury were recorded during
the supervised and unsupervised phases of the intervention.
During the first 6 months the walking and swimming groups attended the
same centre free of charge 3 times/week for supervised sessions. Sessions
comprised 10 min warm-up, 5 min stretching, 30 min of swimming, or walking
at a moderate intensity (60–70% of heart rate reserve (HRRes)) then 10 min
cooling down and 5 min stretching (Cox et al., 2006). Heart rate was measured
by a heart rate monitor (Polar Edge, Polar Electro Oy Kempele Finland). The
Borg Rate of Perceived Exertion (RPE) scale (Borg, 1982) assessed exercise
intensity. At all sessions a trained supervisor recorded attendance, heart rates and
In the second 6 months participants were asked to continue the same
program at the same intensity, unsupervised. They measured their own heart
rates manually, a skill learned throughout the supervised period. Standardized
exercise logs were returned fortnightly by each participant. Swimmers received
a booklet of swimming workouts for the unsupervised period. Adherence
was defined as the percentage of sessions completed relative to the number
In the initial 6 months all participants were given information sheets about
the program requirements, exercise techniques, safety, and a report on fitness at
6 months. Nine newsletters provided reinforcement during the intervention.
Fig. 1. The SWEAT 2 study in older women in Western Australia during 1999–2001. Participant flow from recruitment to the end of the 12 months intervention.
512 K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517
The behavioural intervention (Cox et al., 2003), presented in 12 worksheets,
included strategies such as goal setting, time management and overcoming
barriers. Those assigned to the behavioural intervention had the ‘SWEAT
package’ delivered via ‘mini-workshops’ by the same trained facilitator before
an exercise session (approximately fortnightly) and received worksheets to
complete at home. These were returned the following week then briefly
reviewed in the next ‘mini-workshop’. During the second 6 months all parti-
cipants received newsletters but no other behavioural change material.
Results are expressed as mean (SD) at baseline and as mean (95% con-
fidence intervals) otherwise. Chi square tests were used for categorical data.
Between-group comparisons of continuous variables at baseline used one-way
analysis of variance. Within-group comparisons used t-tests with Bonferroni
adjustment. Generalised Linear Models (GLM) (SAS Institute, Cary, NC.) with
adjustment for baseline values were used to examine treatment effects on
exercise adherence, intensity, and fitness. Analysis of adherence used intention
Pb0.05 was considered significant. The study had 90% power at α=0.05 to
detect 3% difference in adherence related to exercise mode or the behavioural
Baseline characteristics did not differ significantly between
the 4 intervention groups (Table 1).
Stages of change
Fig. 2 shows the distribution of the stages at 0, 6 and
12 months. After 6 months 83% of swimmers and 86% of
walkers had moved to the Action stage; 6% of both swimmers
and walkers were in Maintenance; 8% of swimmers and 6% of
walkers were still in Contemplation; and 4% and 2% of swim-
7% of swimmers and walkers were in Contemplation; 11% and
14% respectively were in Preparation; and 7% of swimmers and
walkers were in the Action stage.
Flow through the Stages of Change was similar in the usual
care and SWEAT groups. For those in Contemplation at base-
line (N=102) 11% withdrew in the first 6 months and 12% in
the second 6 months. For those initially in Preparation (N=11)
36% withdrew in the first 6 months; the rest remained for the
Retention rates and exercise adherence
After 6 months 100 women remained in the study (Fig. 1).
There was no significant difference in retention (the number re-
maining) between the swimming (85.7%) and walking (86.7%)
groups. Adherence was similar (Fig. 3) in swimmers (76.3 (69.5,
Baseline demographic characteristics of older Western Australian women in the 4 study groups of the SWEAT 2 study 1999–2001
Usual care (N=29)‘SWEAT’ package (N=31) Usual care (N=29)‘SWEAT’ package (N=27)
BMI (kg m−2)
Post menopausal (N, % of group)
Married (N, % of group)
Number of children (N)
Number of children at home (N)
Full-time paid employment (N, % of group)
Part-time paid employment (N, % of group)
Distance travelled to centre (km)
Years since vigorous exercise
Attempted physical activity in past 12 months
(N, (% of group))
Contemplation stage (N, (% of group))
Preparation stage (N, (% of group))
Values are mean and (SD). There were no significant differences between groups at baseline.
Fig. 2. The SWEAT 2 study for older Western Australian women during
1999–2001. The stages of change of exercise behaviour for all participants at
baseline ■ 6 months and 12 months □.
513 K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517
83.1)%) and walkers (74.3 (67.7, 80.9)%). Retention in the usual
care and the ‘SWEAT’ groups was the same (86.2%) and adher-
ence was similar (75.5 (68.8, 82.2)% vs. 75.1 (68.4, 81.8)%
(Pb0.05). During the supervised program both groups exercised
(swimming (60.9 (58.9, 62.8)% HRRes); walking (59.7 (57.9,
61.6)% HRRes)). The RPE for both groups was similar and in the
moderate range (swimming 12.6 (12.4, 12.8), walking 12.4 (12.2,
The number of women who withdrew during this period did
not differ significantly between the swim (N=4) and walk
groups (N=10) or the usual care (N=6) and SWEAT groups
(55.3 (45.1, 65.6)%) and swimmers (50.2 (40.3, 60.2)%). With
analysis limited to those who started the unsupervised period,
adherence was 64.7 (54.9, 74.7)% for swimmers and 58.4 (48.9,
SWEAT groups (49.4 (39.2, 59.5)%). Being older was associated
with higher adherence and fulltime employment was associated
with lower adherence (Pb0.05).
The exercise intensity in the unsupervised period was 66.4
(63.7, 69.1)% HRResfor swimmers and 63.4 (60.8, 66.1)%
HRResfor walkers. The RPE was unchanged at 12.3 (11.9, 12.6)
and 12.3 (11.9, 12.5) for the swim and walk groups respectively.
Overall 12 months period
After 12 months 86 (74%) remained in the program; the
retention rate was similar for both the swim and walk (79% vs.
70%) and the usual care (76%) and behavioural intervention
(72%). The adherence rate was similar for swimmers (65.8
(57.9, 73.8)%) and walkers (62.2 (54.6, 70.0)%) and for the
usual care (65.7 (57.4, 73.9)%) and behavioural intervention
groups (62.3 (54.5, 69.7)%). Being older was associated with
long-term retention and adherence (Pb0.05) whereas fulltime
employment was not. Fig. 3 shows the weekly adherence over
the 48 weeks. There was a steady decline in adherence over the
first 24 weeks with a plateau at about week 32, for both modes
of exercise and both behavioural intervention groups.
Table 2 shows walk time and distance swum at baseline, 6
and 12 months by group. After 6 months, there were significant
within-group changes in walk time for both modes of exercise
(Pb0.001). There was a significant difference (Pb0.001) in the
reduction in walk time between the walking and swimming
groups after 6 months (6.5 (7.9, 5.4)% vs. 3.8 (4.9, 2.8)%).
After 6 months the distance swum in the swim test increased by
29 (23.8, 34.9)% (Pb0.001) in the swim group but not the walk
group (−1.0 (−2.7, 1.5)%). The swimming group swam signi-
ficantly farther than the walk group (Pb0.001).
Baseline, 6 and 12 month walk and swim fitness results of Western Australian women in the 4 study groups of the SWEAT 2 study 1999–2001
Usual care‘SWEAT’ packageUsual care‘SWEAT’ package
6 months (N)
12 months (N)
1.6 km walk time (min)
Swim distance (m)
Values are mean and (SD). There were no significant differences between groups at baseline.⁎Pb0.05. Both swim groups were significantly higher than each walk
group at 6 and 12 months.
Fig. 3. Exercise adherence for the women in the SWEAT 2 study conducted in
Western Australia during 1999–2001. Mean weekly adherence (±SEM) to the
prescribed exercise sessions over 48 weeks of intervention in the walk (
and the swim groups ( ). There was no significant difference in
adherence between the modes of exercise.
514 K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517
After 12 months both groups maintained the improvements
in walk time achieved in the initial 6 months (6.1 (7.4, 4.8)%
and 4.0 (5.3, 2.7)% in the walk and swim groups respectively;
Pb0.05). Changes in swim distance seen in the first 6 months
were maintained (1.9 (−3.3, 7.3)% for walkers vs 32.8 (27.6,
37.9)% for swimmers Pb0.001). Interpretation was unchanged
after adjustment for age.
Reasons for withdrawal
In the 16 women who withdrew, “no time” was the most
frequent reason, followed by injury (Fig. 1). Two walkers and 1
swimmer developed injuries, but the walkers' injuries were
unrelated to exercise and the swimmer had symptoms from a
pre-existing shoulder injury. Two participants in the swimming
group had a fear of water (un-revealed at start of the study) and
withdrew after 6 weeks. Two women stopped exercising for
medical reasons and 1 relocated interstate.
14 women further withdrew during this period. Medical
reasons were the most frequent reason given (29%),followed by
aggravation of a pre-existing injury (21%). Two more swim-
mers experienced shoulder problems and a walker had ankle
soreness. Having “no time” accounted for only 14% of the
withdrawals. Death of a spouse and work and family commit-
ments accounted for the remainder.
Retention and withdrawals
Table 3 summarises the baseline characteristics of the parti-
cipants who remained in the study and those who withdrew in
the 12 months period. Participants who remained were signi-
ficantly older (Pb0.05), as seen in the first 6 months (Pb0.05)
(did not withdraw 55.91 (54.97, 56.85) years; withdrew 53.13
(51.85, 55.07) years). Further, the women who withdrew were
almost twice as likely to be in fulltime employment (Pb0.05),
this factor was significant in the unsupervised and long-term but
not the supervised period of the study.
During the group-based sessions there was no significant
difference between the swim and walk groups in the distance
travelled and time taken to get from home to the exercise venue
or the estimated weekly travel cost of attending the sessions. For
the swim group the mean distance, time travelled and cost of
travel was 17.83 (14.72, 20.94) km, 25.50 (20.73, 30.28) min,
and 11.94 (9.72, 14.16) $/week respectively. The corresponding
results for the walk group were, 16.22 (13.32, 19.11) km, 24.35
(19.90, 28.79) min, and 9.76 (7.70, 11.82) $/week.
In the unsupervised phase, swimmers travelled an average
37.3 (26.9, 47.7) km/week to their venue compared with 11.0
(0.7, 21.2) km/week for walkers (Pb0.001). Timetakentotravel
to the venue was significantly greater (Pb0.001) for swimmers
(13.6(10.6, 16.6) min) than walkers (5.6 (2.7, 8.5) min). Whilst
there was no weekly entry fee for the walkers the swimmers paid
significant difference in the perceived cost of transport to the
the walkers $1.
In this study of previously sedentary older women, retention
and adherence to a swimming or walking program in the short
and longer-term were similar with both modes of exercise and
were not enhanced by a behavioural intervention program, con-
trary to the hypotheses. The retention rate for the centre-based
phase of the program, with 86% after 6 months and 74% after
12 months, although similar to other reports (Dunn et al., 1999)
for similar time frames, was lower than the previous study in
which retention rates were 97% at 6 months and 94% after
12 months (Cox et al., 2003). One possible explanation may be
compared to the previous study (mean age 48 years), with older
current study achieved a 75% adherence rate in the first 6 months
compared to 84% in the previous study. In the second 6 months
where both groups were home-based, the adherence rate was
similar (approximately 64%). In the short-term these results were
consistent with othergroup-based programs. At 12months adher-
ence was in the upper range for home-based programs and the
lower range for centre-based studies (van der Bij et al., 2002).
An inverse relationship between adherence and length of the
intervention has been reported, greater in home-based than
group-based programs (van der Bij et al., 2002). In the current
study the decline in adherence in the long-term could relate to
the length of the intervention or the change to a home-based
program. This study design was chosen to enhance the adher-
ence rate based on previous experience (Cox et al., 2003) and
did not allow evaluation of this aspect. A group-based program
Baseline characteristics of the older women in the exercise intervention who
withdrew from the study conducted in Western Australia 1999–2001 and those
who remained in the study for 12 months
BMI (kg m−2)
Post menopausal (N, % of group)
Married (N, % of group)
Number of children (N)
Full-time paid employment (% of group)
Part-time paid employment (% of group)
Distance travelled to centre (km)
Years since vigorous exercise
Attempted physical activity in past 12 months
(% of group)
Contemplation stage (% of group)
Preparation stage (% of group)
1.6 km walk time (min)
Swim distance (m)
Values are mean and (SD).⁎Pb0.05, age was significantly lower and a higher
percentage were employed fulltime in the group who withdrew from the
program compared to those who remained.
515K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517
in the second 6 months may have produced greater adherence
rate, as suggested by reviews (van der Bij et al., 2002), given
that 93% of participants said at 12 months that they would have
continued to attend sessions. However, 50% perceived that their
adherence would be similar to the unsupervised stage.
The effect of mode of exercise on adherence has not been
systematically evaluated. As physical ability, barriers, and ac-
cessibility may be important in adherence, these factors were
controlled in the first 6 months and supervised instruction was
using a centre-based walk program or a stretching program for
(McAuley et al., 1999) and another using an aerobic vs. a stretch
and flexibility program (Brassington et al., 2002).
In the unsupervised period more barriers were predicted for
the swim group such as finding a suitable swimming pool,
travelling to the venue, extra cost in time and money, paying an
entry fee and engaging in a newly acquired activity unsuper-
vised. Distance from a fitness facility has been inversely asso-
ciated with the uptake of programs (Berke et al., 2006) and
with activity level (Booth et al., 1997). Despite these potential
barriers there was no significant between-group difference in
Satisfaction in acquiring a new skill may have motivated the
reported that fitness outcomes during initiation were associated
with subsequent adherence, suggesting that perceived improve-
ments in fitness and appearance may lead to an increase in
confidence to continue. In the current study fitness improved in
both groups, but more in the swim group. These predominantly
novice swimmers received feedback on their progress at every
session in the first 6 months as every task completed was a new
achievement. This increase in perceived mastery may explain
the swim group's ability to overcome potential barriers, con-
sistent with the view that past mastery increases self-efficacy to
enhance future adherence and mastery experiences (Bandura,
The finding that women who withdrew were younger, is
consistent with other reports even though their participation
rates were greater in women under 60 years (van der Bij et al.,
2002). It is possible that in the current study older women were
more likely to be retired from fulltime work and have more time
to participate. The results of this study suggest that in different
situations age and employment may have different influences
on retention and adherence. Being in fulltime employment,
whilst not a risk for retention and adherence during the super-
vised group program it was a significant factor for both in the
unsupervised period and for long-term retention. Employment
status should be an important consideration when considering
physical activity promotion strategies for women particularly if
they are home-based or self managed.
Although behavioural interventions may enhance exercise
adherence (Estabrooks and Carron, 1999; McAuley et al.,
1994), this study found no such effect, consistent with a review
of randomised trials (van der Bij et al., 2002). Social cognitive
and TTM-based interventions can improve physical activity and
functional capacity but changes do not differ from controls
(Pinto et al., 2002; Basler et al., 2007). Possible reasons for a
lack of effect of the behavioural intervention include unrecog-
nised shortcomings in its delivery, or it may have been irrele-
vant for self-motivated women. The group effect, or support
from the group leader (Estabrooks et al., 2004) may also have
The current study design was chosen to allow systematic
evaluation of both treatments and standardize the delivery of
the program (Resnick et al., 2005). One limitation was that the
practical walk and swim sessions could not be separated ac-
cording to the behavioural intervention. Participants were
asked not to discuss the written materials in the practical
sessions but the possibility of some treatment diffusion cannot
Another limitation of the study was that it attracted predo-
minantly well-educated, Caucasian women; women from mi-
nority groups did not volunteer despite extensive recruitment.
However, the findings are relevant to community programs as
women similar to these participants are those likely to join an
exercise group. Although assessment of physical activity relied
on self-reports in the second 6 months, a strength of the study
was that, in the first 6 months, the exercise leader kept records
and objective measures of fitness. A further strength was that
the intervention was longer than most studies allowing evalu-
ation of short and longer-term effects.
In this group of previously sedentary older women, swim-
ming and walking programs did not differ in retention and
adherence in the short or long-term. A behavioural intervention
did not enhance adherence. Future studies should include sys-
tematic evaluation of interventions to ensure that rigorously
tested programs are selected for translation into the community.
This research was supported by a grant from the Western
Australian Health Promotion Foundation. Dr Burke was sup-
ported by the West Australian Institute for Medical Research
and a grant from the NHMRC for the Clinical Research Centre
for Excellence of Cardiovascular and Cerebrovascular Disease.
We thank Penny Rogers for her role in the interviews and data
collection for this project.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.ypmed.2008.01.010.
Australian Bureau of Statistics Cat 3222.0, 2005. Populations Projections 2004
to 2101. Australian Government.
Bandura, A., 2001. Social cognitive theory: an agentic perspective. Annu. Rev.
Psychol. 52, 1–26.
Basler, H.D., Bertalanffy, H., Quint, S., Wilke, A., Wolf, U., 2007. TTM-based
to activity recommendations in older adults with chronic low back pain — a
randomised controlled trial. Euro. J. Pain 11 (1), 31–37.
516 K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517
Berke,E.M.,Ackerman,R.T.,Lin,E.H.,etal.,2006.Distanceasabarriertousinga Download full-text
Booth, M., Owen, N., Bauman, A., Gore, C., 1995. Active and Inactive Aus-
tralians. Assessing and Understanding levels of Physical Activity. Depart-
ment of Environment, Sport and Territories, Commonwealth of Australia
Booth, M.L., Owen, N., Bauman, A., Gore, C., 1997. Physical activity pre-
ferences andsourcesof assistance andperceivedbarriersto increase physical
activity among physically inactive Australians. Prev. Med. 26, 131–137.
Borg, G.A.V., 1982. Psychological bases of physical exertion. Med. Sci. Sports
Exerc. 14, 377–381.
Brassington, G.S., Atienza, A.A., Perczek, R.E., DiLorenzo, T.M., King, A.C.,
2002. Intervention-related cognitive versus social mediators of exercise
adherence in the elderly. Am. J. Prev. Med. 23 (2S), 80–86.
Cooper, K., 1977. The Aerobics Way. Bantam Books, New York, pp. 86–91.
Cox, K.L., Burke, V., Beilin, L.J., Grove, J.R., Blanksby, B.A., Puddey, I.B.,
2006. Blood pressure rise with swimming versus walking in older women:
Cox, K.L., Burke, V., Gorely, T.J., Beilin, L.J., Puddey, I.B., 2003. Determinants
of retention and adherence in a stage-matched exercise intervention in
women aged 40–65 years: the S.W.E.A.T. Study (The Sedentary Womens'
Exercise Adherence Trial). Prev. Med. 36, 17–29.
1999.Comparison of lifestyleandstructuredinterventionsto increase physical
Estabrooks, P.A., Carron, A.V., 1999. Group cohesion in older exercisers:
prediction and intervention effects. J. Behav. Med. 22 (6), 575–588.
Estabrooks, P.A., Munroe, K.J., Fox, E.H., et al., 2004. Leadership in physical
activity groups for older adults: a qualitative analysis. J. Aging Phys. Act.
Kline, G.M., Porcari, J.P., Hintermeister, R., et al., 1987. Estimation of VO2max
from a one-mile track walk, gender, age, and body weight. Med. Sci. Sports
Exerc. 19, 253–259.
Marcus, B.H., Banspach, S.W., Lefebvre, R.C., Rossi, J.S., Carleton, R.A.,
Abrams, D.B., 1992. Using the stages of change model to increase the
adoption of physical activity among community participants. Am. J. Health
Promot. 6, 424–429.
Marcus, B.H., Emmons, K.M., Simkin, L.R., Taylor, E.R., Linnan, L., Rossi, J.S.,
et al., 1994. Comparing stage-matched versus standard care physical activity
interventions in the workplace. Ann. Behav. Med. 16, S305.
adherence in middle-aged males and females. Prev. Med. 23, 498–506.
McAuley, E., Katula, J., Mihalko, S.L., et al., 1999. Mode of physical activity
and self-efficacy in older adults: a latent growth curve analysis. J. Gerontol.
54B (5), 283–292.
Levels of Western Australian Adults 2002: Results from the Adult Physical
Activity Survey and Pedometer Study. Western Australian Government, Perth,
McMurdo, M.E.T., 2000. A healthy old age: realistic or futile goal? Br. Med. J.
O'Brien Cousins, S., 1997. Exercise, Aging, and Health: Overcoming Barriers
to an Active Old Age. Taylor and Francis, Philadelphia.
Owen, N., Bauman, A., 1992. The descriptive epidemiology of a sedentary
lifestyle in adult Australians. Int. J. Epidemiol. 21, 305–310.
2002. Effects of a computer-based, telephone-counselling system on physical
activity. Am. J. Prev. Med. 23 (2), 113–120.
Pinto, B.M., Marcus, B.H., Clark, M.M., 1996. Promoting physical activity in
women: the new challenges. Am. J. Prev. Med. 12 (5), 395–400.
Pollock, M.L., Carroll, J.F., Graves, J.E., et al., 1991. Injuries and adherence to
walk/jog and resistance training programs in the elderly. Med. Sci. Sports
Exerc. 23, 1194–1200.
Resnick, B., Bellg, A.J., Borelli, B., et al., 2005. Examples of implementation
and evaluation of treatment fidelity in the BCC studies: where we are and
where we need to go. Ann. Behav. Med. 29, 46–54.
U.S. Department of Health and Human Services, 1996. Physical Activity and
Health: A Report of the Surgeon General. Department of Health and Human
Services, Centers for Disease Control and Prevention, National Center for
Chronic Disease Prevention and Health Promotion., Atlanta GA, U.S.
van der Bij, A.K., Laurent, M.G.H., Wensing, M., 2002. Effectiveness of physical
activity interventions for older adults. Am. J. Prev. Med. 22 (2), 120–133.
517 K.L. Cox et al. / Preventive Medicine 46 (2008) 511–517