Active commute to school: does distance from school or walkability of the home neighbourhood matter? A national cross-sectional study of children aged 10–11 years, Scotland, UK

Abstract

Strengths and limitations of this study► This study used a sample of children from across the whole of Scotland, and was weighted to ensure representativeness to the wider population of 10-11 years old living within Scotland. ► We included objectively measured home-to-school distance and walkability score. ► We explored how home-to-school distance moderates the effect of walkability on active travel. ► Travel mode (active vs non-active) was determined via self-report rather than objective measurement. AbStrACtObjectives To study the extent to which home-toschool distance and neighbourhood walkability were associated with self-reported active travel to school (ATS) (eg, walking, cycling), and to explore how distance moderates the effect of walkability on ATS, among 10-11 years old.Design Cross-sectional study.Setting Data were collected between May 2015 and May 2016 in partnership with the Growing Up in Scotland Study, a nationally representative longitudinal cohort study.Participants 713 children (male (n=330) and female (n=383) 10-11 years old) from Studying Physical Activity in Children's Environments across Scotland. Primary and secondary outcome measures Children who actively travelled to/from school categorised as active all (100% of ATS) and active 60%+ (at least 60% of ATS); home-to-school road/path network distance (<0.5 km, 0.5 to <1 km, 1 to <1.5 km, 1.5 to <2 km, 2 km+); home neighbourhood walkability (i.e., composite measure of road/path intersection density and dwelling density) (in quintiles). results Distance and walkability were both associated with ATS. The likelihood of ATS for all or most journeys decreased with increasing distance. Compared with 'most' walkable areas (Q1), the odds of active all were significantly lower within least walkable areas (Q5 OR 0.45, 95% CI 0.21 to 0.99), and odds of active 60%+ were significantly less in Q2-Q5 (lowest odds Q5 OR 0.20, 95% CI 0.07 to 0.47). Regarding walkability and distance interactions, for all distance categories, higher walkability increased the probability of ATS (for most school journeys).Conclusion Walkability was positively associated with ATS within all distance categories, with the relationship between walkability and ATS more complex than the clear-cut association between distance and ATS. A more walkable environment had a larger effect on the probability of reaching the 60% threshold of school journeys using ATS than the probability of always travelling in an active manner. Investment is needed in existing less walkable neighbourhoods to provide infrastructure to support opportunities for children's ATS.

Measuring the Built Environment in Studies of Child Health—A Meta-Narrative Review of Associations

“…Other BE attributes reported as significantly positively associated with self-reported ATS include residential density [38,74,76,77,79], land-use mix in adolescents [73], girls [77], and children [55,83], ground floor attractions and retail density [84], the presence of food outlets [74], and walkability [49,77,80,85,86]. Similarly, walking infrastructure attributes such as proximity to walking paths [87], sidewalk width [84], and streetscapes for active travel [86] were reported as associated with self-reported ATS.…”

“…Similarly, walking infrastructure attributes such as proximity to walking paths [87], sidewalk width [84], and streetscapes for active travel [86] were reported as associated with self-reported ATS. Proximity to school was found to be the main variable associated with self-reported ATS for children and adolescents [73,74,76,78,85,88,89] and with objective ATS [26,81,90]. Likewise, objective and reported ATS were associated with accessibility to activity-promoting destinations [55,75,91] and the availability of open public spaces [77,86], greenery [92] [89], greenways [90], and street trees [93].…”

How can an agent-based model explore the impact of interventions on children's physical activity in an urban environment?

“…Distance and street walkability are central determinants of children's choice for mode of travel ( Oliver et al, 2015 ; Williams et al, 2018 ; Macdonald et al, 2019 ), thus affecting PA accumulated by active travel. In the model, roads are explicitly represented allowing agents to evaluate walkability and distance before every journey to decide on the preferred mode of travel.…”

How can an agent-based model explore the impact of interventions on children’s physical activity in an urban environment?

“…The frequency of walking to school is a function of distance to school and street walkability score. The travel mode frequency function is based on an ordinal logistic regression (see regression results in: (Almagor, 2021)) estimated from empirical data from 713 children living in Scotland who reported their mode of travel to school for a period of a week (Macdonald et al, 2019). When travelling to destinations other than school, agents calculate a probability to walk based on distance to destination, walkability score and number of cars in the household.…”