The high prevalence of overweight and obesity in children under the age of five years is a global public health issue [1
]. Insufficient physical activity contributes to childhood overweight and obesity [5
]. The current Australian physical activity guidelines recommend 1- to 5-year-old children should be physically active for at least three hours per day [9
]. The Australian Health Survey 2011–2012 showed that 28% of children aged between 2 to 4 years did not meet these guidelines [10
]. Thus, promoting physical activity becomes an important strategy to prevent early onset of childhood overweight and obesity. Besides weight control, regular physical activity, especially outdoor play, can also help to promote children’s physical health as well as self-regulation, cognitive, and socio-emotional capacities [11
]. Understanding the factors associated with young children’s physical activity can inform the development of early interventions to increase physical activity.
The socio-ecological model emphasises that individual health behaviours are influenced by different levels of environmental factors, including social-economic, physical, and policy environment [14
]. Neighbourhoods are the primary setting where people’s daily activities take place. Hence, the features of the social or physical environment in neighbourhoods where families live may influence people’s physical activity. Substantial evidence shows that the neighbourhood environment plays an important role in physical activity engagement in children and adults [16
]. Social disorder and the built environment in neighbourhood have been the main focus of previous studies. The most commonly examined features of the environment include safety (social disorder aspect), availability or accessibility of recreation facilities, road safety, walkability, residential density, and mixed land use. However, little is known regarding the influence of neighbourhood environmental characteristics on the physical activity of children under 5 years of age.
The nature of physical activity in young children is different from that in older school-age children or adults. While walking, sports and structured physical activities are common physical activities for older children and adults, free play or outdoor play is the primary form of physical activity for young children [19
]. For children between the ages of 2 to 3.5 years, outdoor play is usually free play without goals set by adults, but under parental or carer’s supervision [20
]. It can be expected that the neighbourhood environment influences the physical activity of young children in different ways. In contrast to older children who have more autonomy to arrange their daily activities such as physical activity in leisure time, young children are more dependent on their parents for such activities. It is possible that parents’ perceptions of their neighbourhood environment have greater influences on young children’s physical activities than the actual environment. Therefore, Davison and Lawson suggested that research should include both subjectively perceived and objectively measured characteristics of the environment [16
]. In addition, the type of accommodation and the number of vehicles in a household, which to some extent, reflect a family’s socio-economic circumstances, are also expected to have an impact on young children’s physical activity [21
]. Yet, studies regarding the associations of neighbourhood environments with the physical activity of young children are sparse and the results are inconsistent [22
]. For example, while some studies found no association between neighbourhood environmental factors, such as walkability, road safety, and yard space, with children’s physical activity [23
], other studies have found traffic safety, public housing, and neighbourhood physical disorder were associated with children’s outdoor play [24
]. Hence, there is a need to further investigate the associations between the neighbourhood environment and young children’s physical activity.
Outdoor play is strongly related to young children’s physical activity [28
]. It is often used as a surrogate measure of young children’s physical activity [28
]. Given the potential influence of the environment on young children’s physical activity, the present study aims to investigate whether mothers’ perceived neighbourhood environmental factors, the type of accommodation, the number of vehicles in a household, walkability, and population density are associated with the outdoor playtime of 2 to 3.5-year-old children.
The characteristics of the mothers and children at ages 2 and 3.5 years are shown in Table 1
. With the exception of annual household income, there were no significant differences in mothers and children’s socio-demographics at 2 and 3.5 years. Compared to the participants at 2 years, significantly, more participants with higher income retained at 3.5 years. Mothers who were lost to follow-up were typically young, unmarried, had less education, were unemployed, and had a lower household income.
The results of descriptive analyses of children’s outdoor playtime and neighbourhood environment factors are shown in Table 2
. The percentage of children having 2 or more than 2 h outdoor playtime per day increased from 63% to 67%. However, this increase was not statistically significant. There were no significant changes in most neighbourhood environment factors from 2 to 3.5 years of age. However, the percentage of mothers who thought it was not safe for children to play outdoors and there was heavy traffic in the neighbourhood significantly increased from 26% to 42% and from 19% to 26%, respectively.
The results of bivariate and multiple analyses are shown in Table 3
and Table 4
. Most of the statistically significant associations from bivariate analyses remained significant after adjusting for allocation of intervention group and other confounding factors. At 2 years, mothers who perceived that their neighbourhoods is a good place to bring up children were more likely to have their children playing outdoor for ≥2 h/day with AOR 1.87 (95% CI 1.13–3.07) on weekdays and 1.91 (95% CI 1.12–3.27) at weekends; mothers who perceived that it is safe for children to play outside were more likely to have their children playing outdoor for ≥2 h/day with AOR 2.06 (95% CI 1.29–3.30) on weekdays, and 2.47 (95% CI 1.46–4.19) at weekends; mothers who perceived that there are good parks or playgrounds in the neighbourhood were more likely to have their children playing outdoors for ≥2 h/day with AOR 1.86 (95% CI 1.09–3.18) on weekdays, and AOR 1.83 (95% CI 1.03–3.25) at weekends. The association between the number of vehicles in a household and outdoor play was only statistically significant on weekdays (Table 4
At age 3.5 years, children who lived in a free standing house were more likely to play outdoors for ≥2 h/day on both weekdays (AOR 2.03; 95% CI 1.17–3.51) and weekends (AOR 2.23; 95% CI 1.09–4.55); mothers who perceived that their neighbourhood is a good place to bring up children (AOR 2.96; 95% CI 1.42–6.17) and it is safe for children to play outside (AOR 1.94; 95% CI 1.02–3.70) were more likely to have their children playing outdoor for more than 2 h/day at weekends. Other neighbourhood environment factors, such as mothers’ perceived safe neighbourhood, traffic, suburb level walkability, and population density were not associated with children’s outdoor playtime at 2 and 3.5 years (Table 4
Multiple logistic regression models also showed that from ages 2 to 3.5 years, children with Australian born mothers were more likely to play outdoors for more than 2 h per day during both weekdays and weekends. When compared to mothers who had a less formal education (i.e., completed primary school to School Certificate), mothers who had achieved a higher educational level (i.e., Higher School Certificate, Technical and Further Education, and university degree) were less likely to allow their children to play outdoors for more than 2 h per day on weekdays. However, mother’s own physical activity was not significantly associated with children’s outdoor playtime.
The present study indicates that some factors in the neighbourhood environment could be associated with children’s outdoor playtime at ages 2 and 3.5 years. Mothers’ perceptions that the neighbourhood is a good place to bring up children and that it is safe to play outdoors were associated with children’s higher likelihood of playing outdoors. Mothers’ perception that there are good parks or playgrounds in neighbourhood was associated with children’s higher likelihood of playing outdoors at 2 years. Children living in free-standing house were more likely to play outdoors at 3.5 years.
Mothers’ perceptions that the neighbourhood is a good place to bring up children can be seen as a general view of their neighbourhood. It may reflect that the neighbourhood is safe (i.e., low crime rate), easy to access to recreational facilities, easy to travel, and shopping, etc. Similarly, mothers’ perceptions that it is safe for children to play outdoors can represent physical or social aspects of safety in neighbourhood, such as road safety, or high community cohesion and lower crime rate. There could be some overlap among these perceptions. Heavy traffic on streets or roads is one of indicators regarding road safety and may reflect mother’s concern about their children’s safety, especially when they play outside. An American study found that 5–10 year old children had a lower physical activity level when their parents were concerned about neighbourhood safety regarding both social-disorder and road safety [43
]. Another American population study found that five year old children spent more hours playing outdoors and had more trips to a park or playground when their mother had higher perceptions of neighbourhood collective efficacy (i.e., mutual trust between neighbours who look out for one another) [24
]. Findings from our study echoed this evidence.
However, in our study, mothers’ general view of neighbourhood safety and traffic situations were not associated with young children’s outdoor playtime. An Australian study also found that parental perceptions of neighbourhood safety (i.e., road safety, incivilities, and personal safety) was not associated with young children’s moderate-to-vigorous physical activity, but was associated with adolescents’ moderate-to-vigorous physical activity [44
]. In our study sample, the proportion of mothers who perceived that outdoor play is safe and the traffic is not heavy were significantly less at 3.5 years than that at 2 years. This may show that mothers’ safety concerns change as children grow and become more mobile. A Netherlands study found that traffic volume and speed was not significantly related to outdoor play in 4 to 12 year old children, while other aspects of road safety, such as presence of zebra crossings, traffic lights, and roundabouts, were associated with outdoor play in children [25
]. This indicated that other features of neighbourhood environment might have a stronger influence than traffic volume or speed, or possibly, mothers’ natural protection of young children weakens the influence of traffic.
With regard to recreational facilities, although using subjective measures of parks and playgrounds, our finding were in line with previous studies that used objective measures [22
]. Parks and playgrounds provide one opportunity for children’s outdoor play. It can be expected that the availability or accessibility of parks and playgrounds were associated with outdoor play in young children. A New Zealand study used objective measures of, and teenagers’ perceived access to, parks or playgrounds. The authors found that teenagers’ perceived access to parks or playgrounds was positively associated with their self-reported physical activity, but not the objectively measured total physical activity [46
]. The findings from our study and the New Zealand study might suggest that mothers’ or children’s perception of being closer to parks or playgrounds may have more influence on engaging children in outdoor play. A Netherlands study using objective measures of environment found that the number of formal outdoor play facilities per square kilometer was negatively related to outdoor play in young children, while the informal places such as sidewalks, parallel parking spaces, or grouped parking places was positively associated with outdoor play in young children [25
]. The author concluded that this difference may be due to researchers examining the availability of parks or playground at the neighbourhood level, rather than focusing on the individual accessibility of parks or playgrounds that may be a more proximal factor in relation to children’s outdoor play. Sidewalks or parallel parking spaces are usually closer to home and can be informal play spaces for children. Their findings suggested that proximity to play places have more influence on motivating children to play outdoors.
While many studies found that a higher walkability was positively associated with adults’ transformational or recreational walking and cycling [18
], results from studies for older children were inconsistent. Some studies found that a higher walkability was related to more physical activity, including active transportation to school in children [27
], some did not find such an association [46
], and others found a negative association. A Belgian study found that a lower walkability and longer distance to school was associated with more physical activity in adolescents [53
]. Low street connectivity (i.e., more cul-de-sacs) is a characteristic of a low walkable neighbourhood. Meanwhile, it is a characteristic of lower traffic and higher road safety. Many studies found that children living in cul-de-sac neighbourhoods were more active or spent more time playing in neighbourhoods [45
]. Our study found suburb level walk score was not associated with young children’s outdoor play. Partly, it might be a result of using of suburb level walkability rather than individual household walkability. The cluster nature of data significantly reduced the statistical power to detect the potential association between walkability and children’s outdoor play. On the other hand, neighbourhood walkability usually represents opportunities for walking in everyday life. It is more relevant to daily life of adults or adolescents rather than young children, as the primary form of physical activity of young children is play not walking. Therefore, the effect of neighbourhood walkability on adults may not be generalized to young children.
Population density is the population per unit of land. Intuitively, when the population density is high, the play space can be affected which in turn affect children’s outdoor play. However, our study did not support this hypothesis. Again, it is possible due to the using of suburb level population density, and the cluster nature of data reduced the statistical power to detect the effect of population density. Studies about the influence of population density on children’s physical activity or outdoor play are sparse. Inconsistent findings resulted from two American studies. One study found that a higher population density was associated with higher rates of walking and biking to school in school-age children [56
], and another study did not find such an association [57
Our study found that the type of accommodation children lived in (free standing house) was positively associated with outdoor playtime at 3.5 years of age. For children who live in a free standing house, it is possible that they are more likely to play in the front or back yard of their house. In Australia, there is an increasing urban trend towards high-rise apartment housing due to population increases. Future neighbourhood design should include sufficient green and outdoor play spaces in high housing density areas.
To some extent, the number of cars in a household may reflect the social-economic status of a family, which could influence children’s physical activity. One study found a higher family social-economic status was associated with higher physical activity levels in children younger than 5 years of age [26
]. Children from households that had two or more vehicles were less likely playing outdoors for more than 2 h per day on weekdays at 2 years of age. A UK study revealed that more household cars in use were associated with lower physical activity level in seven year old children [21
]. It is possible that having two or more vehicles in a household indicates both parents are working, which could affect outdoor playtime of their children. Further research is needed to explore this phenomenon.
The differences in findings between 2 and 3.5 years of age, and between weekdays and weekend days, indicate that different features or aspects of neighbourhood environments are associated with outdoor play of children at different ages and times of the week. As discussed above, mothers’ concerns and perceptions of neighbourhood environment can change along with children’s growth, which may also contribute to the differences.
The findings of the present study offered some insight into the associations of the neighbourhood environment with outdoor playtime of young children. Neighbourhood design should consider safety, and good parks or playgrounds for young children’s outdoor play. However, they should be viewed in the light of some limitations. For example, a loss to follow-up may cause selection bias. The cross-sectional analysis precludes the attribution of causality. Recall bias might be caused by the nature of mothers’ reporting their child’s outdoor playtime, especially for those children who attended child-care services or those mothers who were employed full-time, such self-reports may affect the estimates of outdoor playtime on weekdays. Although questions regarding neighbourhood environments were extracted from Growing up in Australia: The Longitudinal Study of Australian Children and were pre-tested, no reliability or validity scores are available. All environment variables were dichotomized due to non-normal distributions, which may have caused some information to be lost. In addition, some neighbourhood or home physical environment factors were not included in this study, such as weather or seasons, crime, or yard size. It is also worth noting that the study was conducted in South-western Sydney, Australia, an area with a relatively low socio-economic level, which could limit the generalizability of the study.