Introduction

Worldwide, overall levels of physical activity (PA) among school-age children are low [1]. There have been international and national efforts to advocate for higher PA participation in this population group, and school environment has been highlighted. A Global Action Plan on Physical Activity, launched by the World Health Organization (WHO) in 2018, recommended that member states should ensure provision of opportunities for PA at education institutions [2]. From a public health perspective, school is an ideal setting where a majority of children spend a substantial amount of their weekday time and can therefore be reached regardless of their race and religion. School is a strategic setting not only for the provision or promotion of PA at the population level, but also for increasing and sustaining it. With regard to this significant role of school, the school environment, i.e. policies and practices, can potentially influence PA behaviors in school-age children [3].

Policy refers to a formal written statement of intent that defines priority and the parameters for decisions and actions required to achieve desirable outcomes. A school policy can be mandated at different levels from national, state/regional, district, and the school itself [3]. Once established and implemented, policies should shape school practices and delivery of PA programs in a sustainable manner. PA programs can be provided through either specific PA policies or holistic health promoting policies. Specific PA policies involve a provision of opportunities for students to be physically active commonly through physical education (PE), recess, extracurricular activities or activities outside school hours, active transport to/from school (i.e., walk or bike), and provision of school PA facilities [4]. PA programs can also be provided through a holistic health promoting policy; e.g., WHO Health Promoting School (HPS) framework which incorporates health policy as an integral element [5].

Overall, review of current evidence suggests that the association between school PA policy and students’ participation in PA is complex, and the knowledge base in this area remains incomplete [3]. Available evidence mostly conducted in developed countries suggests that some specific policies, such as a provision of PE class [6], and recess [7] have no impacts on student’s overall PA levels. However, other specific policies seem to positively influence students’ participation in PA. For example, when schools have policies to provide extracurricular activities for students to be physically active, there appears to be positive impacts on students’ PA levels [6,8]. Nonetheless, evidence on contributions of these policies on children’s overall and particularly domain specific PA is scarce and warrants further investigation.

School practices are referred to decisions and actions guided by policies, and in this case, related to a provision of school PA facilities and programs [9]. The practices differ greatly between schools as they are multifaceted and involve multiple concerned parties. Frequency and duration of PE class and a provision of PE specialists were frequently examined as concrete practices following adopted PE policies. Previous evidence showed that the frequency and duration of PE class consistently had no association with students’ PA [6], whereas a provision of a PE specialist showed a positive association with class time moderate to vigorous PA [10]. The delivery of PE class in a majority of countries have legal requirements for the implementation and consequently all schools across the country have to comply [11]. In Thailand, PE is bundled with health education (HE) and is compulsory as it is mandated by the Ministry of Education. The national education curriculum clearly specifies the maximum class time allocation for PE/HE per academic year (i.e., 80 hrs for grades 1–9, and 120 hrs for grades 10–12) [12]. In practice, Thai schools typically provide <60 minutes of PE/week.

Activities during PE class and active play have received significant recognition from international agencies. PE has been highlighted in the Kazan Action Plan of UNESCO as the most important means to ensure lifelong participation in PA [13], while active play is the right of the child and necessary for their healthy growth and development [14]. Active play (defined as spontaneous, freely chosen, self-directed, unstructured, and unorganized [15]) is a relatively new area of study and active play policies are rarely examined independently [16]. They are typically investigated indirectly through policies providing time for free play [17], recess [18], and non-curricular activities [7]. Moreover, active play has continuously been identified as an alarming issue by a regular national assessment and raised as a national campaign in Thailand since 2016 [19]. Active transport has been promoted as a potential means for increasing PA in people’s daily life, and offers promising opportunities for countries to concurrently achieve many sustainable development goals [2]. Schools that have a policy to promote active transport to/from school report a larger proportion of students who travel actively and a higher number of active trips made by students, as well as a larger proportion of students who meet PA guidelines [20–22].

International literature on the prevalence of school PA policies and their influence on children’ PA levels is growing gradually. However, there is a dearth of evidence from less developed countries. The body of knowledge is lacking particularly in the contributions of school PA policies and practices on children’ domain specific activity. Importantly, it is timely for Thailand to obtain baseline insights into school PA contexts as “Active Children” was one of the three policies recently adopted by the National Committee of the National Action Plan on PA [23]. This present study investigates the prevalence of school policies and practices related to PA and their association with Thai children’s overall and domain specific PA. The information is timely to help plan for national school-related policies in Thailand and will be beneficial for other countries.

Methods

Results

Demographic data and descriptive statistics of the student samples are provided in Table 1. The proportions of students that participated in this study by sex and age categories were almost equal, and almost 30% of them resided in the north of Thailand. Half the students had normal weight (50.3%) although less than one quarter of students (21%) met the current global PA guidelines, and only 18% of students reported active play in ≥2 free periods at school across a school week. Almost all students (92.4%) reported PE classes in the current semester, and just over half (52.6%) traveled actively to/from school.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Descriptive information of the student samples.

https://doi.org/10.1371/journal.pone.0245906.t001

The prevalence of different school policies and practices promoting PA are shown in Table 2. All schools reported a PE policy, whereas only one third of schools reported a policy on recess (35.3%). Most schools (87.5%) provided <60 min of PE class/week and a high proportion of schools (77.2%–94.1%) reported many practices supporting PA promotion. Table 2 also presents results from univariable regressions of school policies and practices potentially associated with students’ overall PA levels. The policy on shared use of school PA facilities with the community had a slightly inverse relationship with students’ overall PA levels (OR = 0.75, 95%CI: 0.56–1.0, p = 0.048). However, after accounting for sex, age, BMI, region, and other school variables in the multiple regressions, no school variables had a significant relationship with students’ overall PA levels.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Univariable regression results of school variables associated with students’ meeting PA guidelines (n school = 136, n student = 5,830)^†.

https://doi.org/10.1371/journal.pone.0245906.t002

Table 3 shows regression results of school policies and practices associated with students’ participation in PE class. The multivariable regression results revealed that students were almost three times more likely (OR = 3.54, 95%CI: 1.39–9.01, p = 0.008) to participate in PE class when schools reported other policies related to PA promotion, compared to those from schools that reported no other policies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Regression results of school variables associated with students’ participation in PE class.

https://doi.org/10.1371/journal.pone.0245906.t003

Results of univariable regressions of school policies and practices related to students’ active play during four free periods at schools are shown in Table 4. Although other policies related to PA promotion (OR = 0.49, 95%CI: 0.31–0.78, p = 0.002) and having a PE specialist at school (OR = 0.54, 95%CI: 0.38–0.76, p<0.001) seemed to have an inverse association with students’ active play of ≥2 free periods at school in the univariable regressions, no factors were significantly associated with students’ active play when accounted for potential confounders and other variables in the multivariable regressions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Univariable regression results of school variables associated with students’ participation in active play^†.

https://doi.org/10.1371/journal.pone.0245906.t004

Students’ participation in active transport to/from school was significantly associated with the existence of a school active transport policy in both univariable and multivariable regressions. When accounting for potential confounders and other factors in the model, students were 40% (OR = 1.40, 95%CI: 1.09–1.82 p = 0.01) more likely to travel actively, compared with their counterparts from schools without this policy, as seen in Table 5.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Regression results of school variables associated with students’ active travel to/from school.

https://doi.org/10.1371/journal.pone.0245906.t005

Discussion

This present study revealed some interesting results. Contrasted with low PA levels among Thai children, a high prevalence (>80%) of Thai schools reported a policy promoting PA in many different areas (e.g., PE, community shared-use of school PA facilities, and HPS). Also, a very high prevalence (>90%) of schools reported practices supporting policies and PA promotion (i.e. provision of annual fitness test, organization of extracurricular activities, and student access to PA facilities outside school hours). Nevertheless, almost none of the school PA policies, practices, and characteristics examined in this present study was associated with students’ overall and domain specific activities, except the policy promoting active transport to/from school. The results suggested that school PA policy might be ineffective in influencing overall PA amongst Thai children, or that school policies might not be constructed to support PA, or their implementation might not be translated into supportive actions.

Counterintuitively, our study showed that no school PA policies, practices, and characteristics had an association with students’ overall PA levels. The results contrasted with some previous studies which concluded that a policy promoting active play [31], recess [7], extracurricular activities [6,8,32], and general health through the HPS framework [33] was positively related to students’ overall PA. Additionally, other studies also showed a positive relationship between students’ total PA and multiple school PA policies (measured by a composite index) [34,35], and allowed student access to PA facilities outside school hours [36]. Nevertheless, our findings are consistent with some studies which demonstrated that school policies (measured by a composite index) were unrelated to students’ self-reported [37] or accelerometry-assessed PA [38]. Beets and colleagues found provocative results in their study as written school policy (promoting PA in general) had an inverse association with students’ PA [39]. These inconsistent findings might be affected by different measures used to assess policy, for example, some used a single policy question and some used a composite index. A standardized policy measurement will need to be developed and agreed among researchers in this field to help better clarify the influence of school policies on student’s PA.

PE policy is believed to contribute greatly to children’s PA [13], but the results from our study showed that PE policy had no association with students’ PA levels. Similar findings concluded that PE policy or practices in schools might not contribute to students’ overall PA [6,37,40]. In our case, school policy and practices concerning PE are unlikely to be major factors influencing students’ PA, due to the high level of PE offered by schools and high level of participation recalled by students across all areas. However, unlike Thailand, most previous studies did not observe high levels of PE delivery by schools and high levels of attendance by students, and this raises important questions for future research. Firstly, on the level of implementation of PE policies including quality and fidelity of PE class delivery in Thailand. Secondly, on the potential overall contributions that PE can make to students’ PA levels with one class of <60 min/week, and whether it is better to offer more PE.

While general PA policies may have limited impact on overall PA, some specific policies, such as those related to active transport were powerful in influencing students’ domain specific PA behavior. The school active transport policy showed a positive association with students’ participation in active travel to/from school, although the actual contributions of school active travel to overall PA among Thai children is unknown. Nonetheless, the relationship identified in our study was consistent with a previous Czech Republic study [21], where students made active travel more frequently when their schools had a policy promoting walking and cycling to/from school. The school active transport policy in Thailand, like many other countries, is not mandated by any government authorities. This policy is usually developed at the school level through the school executive board and any policies developed from schools’ own interests might reflect a strong commitment to its implementation, and potentially the alignment of necessary resources required to achieve the desirable outcome behavior [20,21]. Unfortunately, details of the development, specific strategies and implementation of school active transport policies were not gathered as part of this study. The results however have important implications, as extensive evidence shows that students’ participation in active travel to/from school significantly contributes to their greater overall PA [41]. This present study suggests that specific policies targeting specific PA behavior like the school active transport policy is beneficial and could bring about desirable outcome behavior. Further research on the development of school active transport policy including details of implementation is needed.

Conversely, a policy specifically promoting active play at school had no relationship with students’ PA during school free periods, although this policy is initiated at the school level, similar to the active transport policy. Policies concerning recess frequency or duration and provision of PA facilities and equipment including relevant practices, which seemed to be related to students’ activities during school free time, neither showed an association. These findings were consistent with a Canadian study [42]. However, our results conflicted with those of Ridgers and colleagues who found that written policy (promoting PA in general) had a positive association with PA during recess periods [7]. The lack of an association in this present study might be due to several reasons. First, children’ active play was a less obvious outcome behavior and measured indirectly through opportunities during school free periods e.g. recess, and lunchtime. To date, there are no valid and reliable measurement methodologies and instruments to accurately measure children’s active play (Tremblay et al., 2016). Second, the active play policy might actually have a limited effect on students’ choices for their activities during school free time. Multiple factors, apart from school policy, e.g., students’ preference for play space, safety concerns, and recreational environment, may have stronger influence on students’ engagement in active play (Lee et al., 2015). Finally, the implementation of this policy might play a significant role. Details of the implementation, for instance availability of guidance for implementation, school surveillance or supervision during school free periods, and school physical environmental support towards children’ active play would be helpful in explaining the lack of an association. Future studies need to unfold this missing information and studies with a robust measurement of active play are definitely warranted to clarify actual contributions of the policy promoting active play on children’ activities during school free time.

In light of students’ participation in PE class, other PA-related policies showed a positive association to this domain specific activity. It is possible that additional exposure to PA, other than PE class, through other PA-related policies, might indirectly reflect the importance of PE and lead to increased participation in PE class [43]. However, it is still questionable in the current study as the details of other PA-related policies adopted by schools were lacking. Further investigation is necessary in order to understand the identified positive relationship.

The key strengths of this study include the use of a previously validated instrument to collect Thai students’ PA data, and the use of nationally representative data of children and youth in a wide age spectrum of 6–17 years across the country. Despite the low response rate to the principals’ survey of 44.3%, this study has provided some baseline Thai data on school PA policies and practices. Though there is a potential selection bias through the principals’ response rate, a comparison between students’ participation in overall PA, PE, active play and transport revealed no significant differences between participating and non-participating schools in the principal survey. Data on school policies, practices, and characteristics including students’ participation in PA were based on self-reported measures. The school data might have been biased due to school principals’ social desirability of supporting students with opportunities for PA in school. Data on implementation of the policies reported which would allow more understanding of their influence on students’ participation in PA remain largely unknown. Additionally, the very high proportion of schools with PA-related policies made it difficult to compare between schools and student behaviors, and therefore, the overall influence of policies on student behaviors. Finally, a causal relationship between explanatory and outcome variables cannot be established from this cross-sectional study.

Conclusions

This study provides an important new finding on the contributions of school PA policies to children’s overall and domain specific PA. The finding supports the importance of establishing a formal policy position that should be specific to a targeted PA behavior. Despite a high prevalence of Thai schools reporting many policies and practices promoting PA in different areas, students reported low levels of overall PA. This study suggested that none of the school PA policies, practices, and characteristics was associated with students’ overall and domain specific activities, with active transport as the one exception. The policy promoting active transport to/from school seemed to positively influence students’ engagement in active travel. While this policy may need updating in the current situation, it remains a significant finding and helps underscore support for wider adoption of the policy promoting active transport to/from school which is designed for a specific outcome behavior amongst Thai schools. Intervention studies are necessary to confirm this finding. Our study also reflects that, at this stage, policy formulation might not be an issue among Thai schools. Concentration on strategies to translate the existing policies into effective implementation will be a priority, if greater impacts of PA policies are to be expected.

Acknowledgments

The authors appreciate Thailand Physical Activity Research Centre (PARC) for its administrative and coordinating support. We thank the Office of Basic Education Commission for its collaboration and institutional support. We highly appreciate assistance from all regional co-researchers and their staff who contributed greatly to the data collection. We feel gratitude to all school principals, teachers, and students who participated in this study.