  |
|
|
||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
04-April-2008, Vol 121 No 1271
|
||||||||||||||||||||||||||||||||||||||||||||||
|
Walking to school: frequency and predictors among
primary school children in Dunedin, New Zealand
Sofie Yelavich, Cindy Towns, Richard Burt, Kent Chow, Roana
Donohue, Haji S H Sani, Keryn Taylor, Andrew Gray, Jason Eberhart-Phillips,
Anthony I Reeder
Childhood obesity is a global problem that has increased
dramatically over the last 25 years.1–6
The New Zealand Children’s Nutrition Survey shows that 21% of New Zealand
children are overweight and 10% are
obese.7
Childhood obesity has numerous effects on both individuals
and communities. Health consequences include diseases of the cardiovascular,
gastrointestinal, respiratory, and endocrine systems. Neurological, orthopaedic,
and psychological sequelae are also
possible.8–11
The direct cost of obesity had been estimated to be between 2 and 7% of a
country’s annual health budget.12
Although the increase in childhood obesity is
multifactorial,10,11 declining participation in
physical activity is a major contributor. Attractive sedentary alternatives have
drawn children away from discretionary physical activity, while the need for
incidental physical activity has been reduced by energy saving devices such as
remote controls.13–16
Safety concerns are also restricting children’s
freedom to play and exercise outside the
home.17 In New Zealand, more than one-third of
children fail to achieve the recommended level of physical
activity.18 Similar
trends have been observed in other developed
countries.19,20
Walking to school provides children with a regular and
convenient form of physical activity. This study seeks to estimate the
prevalence of walking to and from school among Dunedin primary school children
and identify major predictors of this activity.
MethodsSample selection—Schools were
identified from the Ministry of Education
database.21 All 46 primary schools (Years
1–6 [primer 1 to standard 4]; ages 5–11) within a 10 km radius of
the Dunedin city centre were contacted by telephone and invited to participate.
Special schools (those catering solely for children with physical or
intellectual disabilities) were excluded. Thirty-nine schools (85%) agreed to
participate in the study.
Two year-levels were randomly selected at each
participating school: one from Years 1–3 (age 5–8) and one from
Years 4–6 (age 8–11). All children from the selected year levels at
each school were surveyed by members of the research group. ‘Model’
classes (those with a mix of students from all year levels in one classroom)
were excluded.
Procedures—Children were asked
to raise their hands if they had either walked or been driven to school, or got
to school by any other means of transportation on Tuesday 30 November 2004.
Children were then asked how they expected to get home after school that day. In
each case, the numbers of children were recorded on a data sheet. In addition,
every child received a questionnaire to take home for their caregivers to
complete and return to school. Extra copies were left for any children who were
absent on the sampling day. The questionnaires were followed by reminder letters
5–6 days later. Completed questionnaires were collected from the schools
approximately 10 days after distribution.
Instrument—The questionnaire
requested demographic data and information about the number of days the child
walked to/from school in a typical week. Questions included sex, age, and
ethnicity of the child; number of adults in household; number of other children
in household; distance from school (less than 1 km, 1–2 km, 2–3 km,
more than 3 km); hours of physical activity per week (less than 1 hour,
1–2 hours, 2–3 hours, 4 hours or more); and whether the parent had
walked to school as a child. Parents were asked to specify the number of days
(out of 5) that the child walked to school and from school in a typical
week.
Analysis—Data were analysed
using SPSS 12.0 and SAS 8 software. A logistic regression model was used to
predict walking on a given day/time with a random effect for schools (taking
into account the cluster effects of schools) and accounting for the correlated
nature of the 10 possible walking periods for each child.
Ethical approval—Departmental
ethical approval was obtained, in accordance with the University of Otago Human
Ethics Committee guidelines.
ResultsOn the sampling day, 1524 children were present and all
participated in the classroom survey. Overall, 1703 questionnaires were
distributed to caregivers; 1157 were returned, thus giving a response rate of
67.9%.
Proportion of children walking to
school—From the classroom survey, 34.5% of children had walked to
school and 36.8% anticipated walking home on the study day. Parents’
reports of how often their child walked to school in a typical week are
presented in Table 1.
Table 1. Number of walking episodes per week
and percentages of children walking to or from school in a typical week
(caregivers’ reports)
Predictors of walking behaviour—A
logistic regression model was used to predict walking on a given day/time with a
random effect for schools (taking into account cluster effects of schools) and
accounting for the correlated nature of the 10 possible walking periods for each
child. Of the 1157 caregivers’ reports, 1006 provided full data and were
included in the model. The model included sex, school year group, school decile
group, distance to school group, ethnicity, number of adults in household,
number of cars relative to adults (no cars, fewer cars than adults, same or more
cars than adults), number of other children in household, physical activity
group, whether the parent walked to school as a child, and presence of a walking
school bus at school.
All variables were statistically significant (p<0.05)
except for the number of other children in the household and the existence of a
walking school bus. These results are summarised in Table 2.
There was a U-shaped relationship between physical
activity and the odds of children walking to school. Children who exercised for
2 to 4 hours per week were less likely to walk to school than those who
exercised for less than 2 hours or more than 4 hours. Children with a parent who
had walked to school as a child were 35% more likely to also walk (OR 1.35, 95%
CI 1.16–1.58).
Table 2. Summary of logistic regression model
for predictors of walking to school
*Walking school bus (WSB) consisting of adult
volunteers walking a set route at a given time, collecting children along the
way.
DiscussionThis study indicates that 47.5% of Dunedin primary school
children walked to/from school less than three times a week. Surveys from other
cities in NZ and overseas have reported between 21% and 69% of children walk to
school.22–26 However, comparison between
studies is limited by area-specific variations that may impact on walking
patterns. These include: climate, season, availability of public transport,
topography, population density, and socioeconomic status.
Moreover, methodologies also vary considerably; sampling may
include one or several schools, response rates differ, and the questionnaires
themselves are often not comparable.
As with other studies in the
US,27 UK,28
and Australia,29 the distance from home to
school had the greatest impact on whether the child walked to school. This
finding may have implications for school zoning policies. Education policy in
New Zealand allows children to attend the school of their choice with few
restrictions.21
Recent closures of small schools may also encourage children
to use motorised transport as they no longer have the option of attending a
school within reasonable walking distance of their homes.
Boys were more likely to walk than girls. This may reflect a
gender difference in activity levels30 or
safety concerns related to a perception that girls are more vulnerable than
boys.
Children in Years 1 to 3 were less likely to walk than older
children; this finding is also likely to be related to safety issues. Caregivers
may be concerned that younger children have not fully developed the perceptual,
cognitive, and motor skills to successfully avoid hazards that may be
encountered on the journey to school.31 Indeed,
the New Zealand Police recommend that children in Years 1 and 2 should not cross
roads alone and (if possible) should walk to school accompanied by someone
older.31
Attendance at a low-decile school and lesser car ownership
were independently associated with a higher probability of walking. Indeed,
other studies have also reported an association between high walking rates and
low levels of a number of socioeconomic
indicators.24 This finding is likely to reflect
the relatively high cost of motorised transport currently further exacerbated by
high fuel prices.
Walking rates were lower among NZ European children when
compared to other ethnicities. Higher walking rates have been reported in
children from non-English speaking
backgrounds.24 This is of interest when placed
in the context of childhood obesity within New Zealand; Pacific Island, and
Māori have much higher rates of childhood
obesity,8 yet appear to be participating in
greater levels of incidental physical activity via walking to school.
There are several potential explanations for this apparent
contradiction. First, the higher rate of incidental physical activity in this
group may not be enough to outweigh other contributors to obesity, such as
nutritional factors (e.g. consuming relatively large amounts of snacks
containing high levels of fat and sugar).
Secondly, body mass index data were not collected, so we do
not know whether individual Māori and Pacific Island children in our study
were more overweight than their peers.
Finally, the observed effect may be due to confounding by
socioeconomic status. School decile and car access were used as proxy indicators
of socioeconomic status in this study, but these measures may be too crude to
adequately control for socioeconomic factors.
Cooper et al30 suggested
that children who walk to school are more active throughout the day. However,
this study has shown that a higher frequency of walking is associated with both
the highest and lower physical activity groups.
A walking school bus (WSB) consists of adult volunteers who
walk a set route at a given time, collecting children along the
way.32 This study showed that WSBs had no
effect on the proportion of children walking to school. This may be due to
relatively few WSB routes in Dunedin and too few children taking part in each
bus.
Identified constraints on the success of a WSB program
include lack of parent volunteers, loss of key enthusiasts on the school staff,
and loss of novelty value.33
ConclusionsThe increasing burden of obesity and the decline in physical
activity have been recognised as major public health
issues.12 Walking to school provides children
with a convenient and regular means of increasing energy expenditure.
Living a short distance from school was the strongest
positive, potentially modifiable predictor for walking to school. Male gender,
higher school year, non-NZ European ethnicity, and attending a low decile school
were all associated with a significantly increased rate of walking.
Results from this study could inform and guide the
development of health, transport, and education policies directed at increasing
the proportion of children walking to school.
Competing interests: None known.
Author information: SofieYelavich, Cindy
Towns, Richard Burt, Kent Chow, Roana Donohue, Haji S H Sani, Keryn Taylor
(Trainee Interns); Andrew Gray (Biostatistician); Jason Eberhart-Phillips
(Senior Lecturer); Anthony I Reeder (Director, Social & Behavioural Research
in Cancer Unit); Department of Preventive & Social Medicine, Dunedin School
of Medicine, University of Otago, Dunedin
Acknowledgements: Dr Reeder and the Social
& Behavioural Research in Cancer Group receive support from the Cancer
Society of New Zealand Inc and the University of Otago.
Correspondence: Dr Tony Reeder,
Social & Behavioural Research in Cancer Unit,
Department of Preventive & Social Medicine,
Dunedin School of Medicine, University of
Otago, PO Box 913, Dunedin 9054, New Zealand.
Fax: +64 (0)3 4797298; email: tony.reeder@ otago.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|