Worldwide, the number of people with chronic diseases and who are overweight or obese increases continuously [1
]. Previous research has shown that regular physical activity can prevent becoming overweight or obese [3
] and is related to a lower incidence of chronic diseases [5
]. However, most people do not meet physical activity recommendations [6
]. Therefore, there is an urgent need for interventions promoting physical activity to enhance population health [5
]. Previous research suggests that parks can play an important role in promoting physical activity among people of all ages as parks are present in most communities and are generally free to use [9
]. Parks can provide a suitable setting for organized and non-organized physical activity by providing a variety of physical activity facilities. In an urban environment, parks can also be a destination to visit by foot or bike [12
] and thereby have the potential to increase physical activity even if park users engage in sedentary behaviour after arriving at the park.
A substantial amount of research has demonstrated that people living closer to parks perform more leisure-time and park-based physical activity [11
]. Moreover, studies from Australia and the UK have shown that park users are more likely to comply to physical activity recommendations [19
]. However, there seems to be a difference in physical activity levels depending on the kind of park area or facilities used (i.e., wooded area, tennis court, trails and paths, meadows, open spaces) [21
]. For example, in the U.S., Besenyi and colleagues (2008) reported higher adult energy expenditure on paved trails and tennis courts than on open spaces, playgrounds and picnic areas. Among children, higher energy expenditure was observed in playgrounds compared to picnic shelters [21
Moreover, Bedimo-Rung and colleagues emphasized in their conceptual model, that not only structural characteristics of parks, but also individual user characteristics influence park visitation and park-based physical activity (e.g., women are less likely to visit a park compared to men) [9
]. Therefore, it would be useful to determine physical activity levels according to the specific park area used and to determine if differences exist based on gender and age of the park users. Park use and park-based physical activity can also differ between workweek and weekend days and according to the hour of the day (e.g., U.S. parks were used less frequently during the morning [11
]). However, knowledge is limited in regards to the temporal aspects of park use. Therefore, insight is needed into the time of the day and day of the week during which parks are used less often so that they can be targeted in interventions aimed at increasing park use.
Objective information on park users can be obtained through direct observation tools, of which the System for Observing Play and Recreation in Communities (SOPARC) is the most frequently used [25
]. Two recent systematic reviews of studies using SOPARC and other observational methods to measure park-based physical activity [26
] indicated that most research originated from North America and few studies were conducted in Australia [28
], South America (Brazil) [30
], Asia (Taiwan, China) [32
] and Europe (Denmark and Belgium) [34
]. One study used direct observation to determine the association between park-based physical activity levels and neighbourhood walkability and income in Belgian and U.S. parks. However, they did not report the specific areas of the parks related to physical activity nor the time of the day and day of the week when parks were used most often [34
]. Research on park visitation and park-based physical activity in Europe and Belgium is scarce and urban environments in Europe differ from those in other parts of the world. Accordingly, insight is needed on this topic in order to better understand park user characteristics, to define priorities for park renewal and construction, and to inform interventions to promote physical activity in parks.
Therefore, the first aim of this study was to describe the characteristics of park users in Ghent (Flanders, Belgium), the activity levels of park users and the types of activities performed. Secondly, this study aimed to examine the association between park areas, day and time of day, and the observed number of park users and physical activity levels of park users for males and females and children/adolescents and adults/seniors using direct observation methods.
Sufficient physical activity contributes to better health status across the lifespan, and parks have the potential to increase physical activity at the community level by providing opportunities for physical activity. This study described the characteristics of park users in Ghent (Flanders, Belgium), the activity levels of park users, the type of activities performed and the associations between park areas and temporal variables with observed physical activity levels. The majority of the observed visitors were males, adults and were engaged in vigorous-intensity physical activity with cycling, sitting and walking being the most popular activities. Seniors were the least represented age group and the smallest number of park visitors were observed in the morning and during weekends. Activity levels were higher on trails and lower in wooded areas compared to grassy areas with a playground.
Overall, the parks in this study had low levels of visitation (58% of observation moments, parks were empty), which could possibly be caused by the small amount of features and amenities present in the parks (according to the audit that was performed). However, these parks were selected because they did not have atypical features (for Flanders) and had not been renovated recently. Moreover, organized activities were limited and supervision and provision of equipment was lacking. This indicates that there are many opportunities to improve the attractiveness, use and physical activity within these parks. When developing programs to increase park visitation, it is important to determine which specific target groups are currently underrepresented and could benefit from extra attention, and which park areas are associated with higher activity levels. In addition, associations of park areas and temporal variables with the number of park users and physical activity levels were examined according to gender and age group.
Most park users in this study were male (58.7%), which is consistent with previous research using observational measurements in Belgium [34
] and with the results of a review by Evenson and colleagues [27
]. However, this gender difference was only present for children, adolescents and seniors, and not for adults, where the distribution was more equal between males and females. For children and adolescents, this gender difference may be attributed to higher independent mobility among boys compared to girls [45
] or by the nature of the activities boys and girls participate in (i.e., girls like to go shopping, whereas boys prefer to do sports [46
]). The age distribution of the park users was comparable with those from observational research from the U.S., with the majority of park users being adults and only a minority being seniors. In this study, more adolescents were observed than in U.S. parks [11
] and Australian parks [28
], but less adolescents than in previous Belgian research [34
]. The relatively low percentage of seniors observed in the parks (9.7% of the park users were older than 60, compared to 21.7% of the Ghent population [38
]) indicates that extra effort may be needed to encourage seniors to use parks more often. The city of Ghent has many parks spread throughout the city and they should therefore be readily accessible or within close proximity to home for most residents. However, the current park features/amenities may not encourage seniors to visit the parks or access to the parks may be difficult (i.e., uneven footpaths [49
]) and discourage visitation. Future studies need to explore which park features are important for park use among seniors. Of the park users observed, 82% were Caucasian, which is representative of the Ghent population and similar to previous research [34
In this study, almost half of the observed park users were observed engaging in vigorous-intensity physical activity, whereas in previous observational studies in the U.S., most park users were sedentary [11
]. In Australia, most park users were standing or moderately active [28
]. Sedentary activities in parks can have social benefits [50
] and even people who use parks for sedentary activities may travel to the park using active transport such as by foot or bike [12
]. Previous research in adolescents indicated that adolescents prefer to alternate between active and sedentary activities [51
]. Therefore, parks should be designed so that they support both physical activity and sedentary activities and provide sufficient infrastructure to support active travel to the park. The high number of vigorously active users in this study could possibly be attributed to the high number of cyclists observed in the parks (38% of park users were cycling). However, it must be noted that most cyclists were only passing through the park to travel elsewhere suggesting that the parks offered an alternative, convenient and/or safe cycling route. Policy makers should take this into account, as cycling is often forbidden in parks. At least one cycle lane in each park could be a useful strategy for future park development. The other activities that were frequently observed in the parks in this study were sitting (23% of users), walking (15% of users) and playing active games such as hide and seek (7% of users). This could indicate that apart from using a park to cycle through, parks in Flanders are used mostly as a place to relax rather than to engage in intensive sports or to work out, which is consistent with the conclusion by Van Dyck et al. (2013) [34
]. However, the qualitative results of a study conducted by Cohen and colleagues (2007) revealed that U.S. parks were used for sedentary activities such as picnics (22% of users), as well as for more active purposes such as playing basketball (15% of users) or as spectators of an organized activity (13% of users) [11
]. This indicates that cultural differences may exist in park activities between Europe and the U.S.
Less participants were observed during morning periods and during weekends. However, the activity levels were lower during the day (for females and all age groups) and higher on the weekend for males. This may indicate that when less people are present in the park, park users are more likely to engage in physical activities instead of sitting behaviours, or park users may have different intentions when they visit a park in the morning (e.g., they are cycling to school/work) compared to during the day or during the weekend compared to a weekday. Other studies have also shown park visitation to be lower in the morning [11
]. However, an Australian study on park use in metropolitan parks reported more visitors during the weekends [28
] than on weekdays, which is contradictory to our results. These contradicting results could be due to differences in size and location of the parks included in the Australian study: large metropolitan parks outside of the city (=long travel time) compared to small parks located in urban areas. If the findings of the current study are confirmed in future research for small parks, future interventions promoting park use in small urban parks could possibly target morning and weekends as these are the moments small urban parks are currently used least.
The odds of at least one park user being present were higher on trails and lower in wooded areas compared to grassy areas with a playground. No difference was found in the number of children/adolescents observed on trails and in wooded areas compared to grassy areas with a playground. This is surprising as it could be hypothesized that children and adolescents would use a grassy area with a playground more often than a trail [21
]. This may be the result of analysing children and adolescents together, thereby age-dependent differences may have been overlooked. Future research should allow for analyses for each age group separately by including more parks. Furthermore, the activity level of all user groups was higher on trails compared to grassy areas with a playground. This could possibly be due to the high number of cyclists on the trails. This supports previous research from Canada and Denmark that revealed a significant association between the presence of a walking/cycling path and park-based physical activity in adults [53
] and research from Besenyi et al. (2013) that revealed the highest energy expenditure among adults on paved trails in U.S. parks [21
]. A possible approach for future urban planning may be to link existing urban parks with each other using trails to encourage walking and cycling [53
]. This could be a suitable strategy for European historical cities such as Ghent, which are often densely built with many individual small urban parks.
The higher activity level of children and adolescents on trails compared to grassy areas with a playground could be explained by the nature of the playgrounds. In the studied parks, the playgrounds consisted of components that primarily facilitate sedentary behaviour (such as a sand pit) rather than encourage physical activity. Furthermore, playgrounds are often designed for younger children [51
]. Hence, playgrounds that include equipment such as climbing structures or basketball hoops that encourage physical activity should be chosen over playgrounds that include equipment that promotes sedentary behaviour [56
]. Previous qualitative research in adolescents indicated that adolescents often visit public open spaces with younger siblings [51
]. Therefore, it is also important that playgrounds are designed to cater for the needs of multiple age groups.
A possible explanation for higher activity levels of females on trails compared to grassy areas with a playground could be that when females are present at a grassy area with a playground, they often engage in standing and sedentary activities to supervise children at the playground [11
]. Cohen and colleagues (2007) have recommended the provision of equipment that provides opportunities for carers of children to be active near playgrounds while supervising their children (e.g., walking paths around the playgrounds, or adult fitness stations) [11
]. This could be a valuable strategy to increase physical activity levels in parks among carers of children.
Earlier findings highlight the importance of the presence of trees in parks as this is associated with physical activity [54
], mental health [58
] and the provision of shade [59
]. However, in this study, the wooded areas had the lowest number of users (55% lower than trails). This low number of users may be due to the density of the wooded area (in this study the wooded area had a high density of trees) which may have a negative influence on park users’ perceptions of safety, since secluded areas may create feelings of insecurity [57
]. Therefore, park designers should ensure the presence of trees but avoid creating densely wooded areas.
This study used a valid and reliable observation tool (designed specifically for parks) to assess park use, park-based physical activity and characteristics of the park users [25
]. All observers received training to use this tool and interrater reliability was confirmed before starting the observations. Additionally, the same observers performed the observations in both parks at exactly the same time and on the same days. Cohen et al. (2011) recommended at least four days of observation to obtain robust measures of park use and user characteristics [41
]. In this study, observations were performed on nine days.
This study only comprised two parks in Ghent and the findings should be considered exploratory. The audit of park characteristics was not validated by a second observer, and this should be considered as a limitation of the study. In the future, park audits should include an assessment of the park areas separately and an assessment of the quality and condition of the park features and amenities. In addition, to provide a better understanding of the results, future research could include qualitative methods to study the perceptions of park visitors. The observations only took place during neutral to good weather, therefore no conclusions could be drawn about the association between weather and park use. In addition, the observations were performed in summer and beginning of autumn (July to October) so no seasonal effect could be examined and results may be different in other seasons. The observations took place at fixed moments in time on nine days and other days and moments could possibly provide different results. Further, no additional information (such as living close to the park, income or transport mode to the park) was collected from the park users. A final methodological limitation is the estimation of the park users’ age which may have led to the misclassification of park users into different age groups.