physical activity, factors other than the inter-
vention might have contributed to the project
results. In addition, selection bias is a concern.
Only women with their household telephone
numbers listed in the White Pages who were
willing to be interviewed had a chance of being
included in this study, although this was similar
for both surveys.
Comparisons between the two independent
samples were made in the study. No significant
differences in demographic factors were found
across the samples. Therefore, any changes occur-
ring between the pre- and post- surveys were not
due to sample differences. Furthermore, given
an Australian trend towards an increase in the
prevalence of sedentary people in recent years
(Armstrong et al., 2000) and taking into account
results of the qualitative study undertaken at the
same time, it is our conclusion that the inter-
vention has played a substantial role in these
changes.
The finding of a shift from sedentary to low
activity in the present study population was en-
couraging for community-wide health promotion
interventions generally. It has been argued that
the greatest benefits of public health will accrue
if, rather than persuading those who are already
active to be more active, campaigns are developed
to encourage those who are very inactive to
participate in some modest amount of activity
(Blair et al., 1989).
This study provided an opportunity to assess
the impact of a health promotion initiative
encouraging women to participate in physical
activity, particularly in walking. It was a collab-
oration with a local council supporting the
community ‘events’ for walking and developing a
walking group. Intersectoral action for health
improvement has been the focus of much recent
research and this project was a successful prac-
tical application of intersectoral action at a local
level (Harris et al., 1995). Community events
were seen as the most visible and successful
aspects of the project, from the point of view of
both the collaboration with the council, and the
participation of the community. The community
events demonstrated that there are existing com-
munity networks which can be readily mobilized.
Concord Council and the Health Promotion Unit
worked well together, as was evidenced by the col-
laborative efforts in implementing and sustaining
the project activities and the publicity generated
by the community activities. The involvement of
the community in the project and the use of the