Accelerometry-measured physical activity and sedentary behaviour of preschoolers in Nova Scotia, Canada

Objective: To describe the levels and bouts of physical activity (PA) and sedentary behaviour (SB) among preschoolers in Nova Scotia, Canada, and the proportion meeting PA and step guidelines. Methods: Children (75 boys, 49 girls; mean age = 4.2 (range = 3-5 years)) participating in the Physical Literacy in the Early Years (PLEY) study provided data. Average time (minutes, % of day) spent sedentary, in light PA (LPA), moderate-to-vigorous PA (MVPA), and total PA (TPA), average frequency and duration of bouts of MVPA (≥ 5 minutes, ≥ 10 minutes) and sedentary bouts (≥ 10 minutes), per day were determined using published cut-points, using 15-s epochs. The proportion of children meeting PA and step guidelines was determined, and differences by sex explored. Results: Children spent the majority (70.8%) of their day active; nearly all (≥97%) met PA guidelines. Most children met step guidelines on a weekly basis, but not daily. Only LPA differed by sex (greater in girls; p=0.001). Little time was spent in sustained SB (bouts ≥10 minutes). Boys had greater and longer bouts of MVPA, and spent more of their day in these (p<0.05). Girls spent less time in sustained SB (p=0.009). Conclusion: This is the first study to describe the PA and SB of Nova Scotia preschoolers using accelerometry. Findings suggest preschoolers spend the majority of their day active, and that there are sex-related differences in PA and SB, warranting further examination. probability


Introduction
Physical activity (PA) during the early years (age 0-4 years) and in school-aged children and youth (age 5-17 years) is critical for the development and maintenance of physical, mental and social wellbeing.
Systematic reviews provide evidence that higher levels of PA are associated with more favourable measures of adiposity, bone and skeletal health, cardiometabolic health, motor skill development, psychosocial health and cognitive development in infants, toddlers and preschoolers (Carson et al. 2017a;Timmons et al. 2012). Moreover, moderate to vigorous physical activity (MVPA) and more PA in general, appears to be better for health in the early years (Carson et al. 2017a). Insufficient levels of PA and excess time sedentary among preschoolers has been linked with increased adiposity (LeBlanc et al. 2012) and blood pressure (Vale et al. 2015), poorer motor skills (Iivonen et al. 2013 D r a f t 5 continuum (Chaput et al. 2017). The 24-hour movement guidelines classify PA guideline adherence based on an average of all valid days of accelerometer wear. Previous reports (Colley et al. 2013;Garriguet et al. 2016) have operationalized PA guideline adherence differently, requiring that children meet recommended amounts of PA on every valid day of accelerometer wear. These earlier reports also use varying classifications for the amount of PA required to meet guideline adherence, which consequently, impacts the percentage of children achieving guidelines. Together, these findings highlight the importance of clearly defining how PA guidelines are operationalized when indicating the proportion of preschoolers achieving guidelines. Furthermore, these findings indicate the utility of investigating guideline adherence in a variety of ways, to determine whether preschoolers are sufficiently active on a daily versus on an average basis.
Several other studies have used accelerometers to provide a direct, objective assessment of PA and SB in Canadian preschoolers (Borkhoff et al. 2015;Carson et al. 2017b;Colley et al. 2013;Garriguet et al. 2016;Kuzik et al. 2015;Kuzik et al. 2016), contributing to the growing international body of evidence (Aguilar-Farias et al. 2015;Andersen et al. 2017;Berglind and Tynelias 2017;Gutiérrez-Hervás et al. 2017;Moller et al. 2017). The results of these studies vary due to inconsistencies in accelerometers used to measure PA and SB, and accelerometer data reduction decisions (e.g. wear and non-wear time, number of valid days necessary for inclusion in data analysis, intensity cut-points, etc.). Furthermore, the PA and SB profiles of Canadian preschoolers are mostly based on National samples; less is known about the behaviours of regional/local Canadian preschoolers. While National PA and SB data provide a larger, more diverse and representative sample of preschoolers, data from large population-level studies (e.g. Canadian Health Measures Survey) are not typically compared by region in resulting publications (Chaput et al. 2017), leaving a gap in the understanding of whether the PA and SB of Canadian preschoolers differ significantly by region. Studies that provide regional or local-level data, albeit on smaller, possibly more homogenous populations, can be compared to National data to explore D r a f t 6 whether these preschoolers' PA and SB patterns appear to be similar or different to National averages.
Regional-level PA and SB data on preschoolers can also be linked to other regional-level data (e.g. childcare/school programming and policies for outdoor play provision, quality of childcare/school outdoor play environments) to determine whether any relationships exist. This information can spark a greater conversation, regionally, about potential changes that need to be made, to better support preschoolers' opportunities to be physically active in major settings and places (e.g. childcare, school, home, greater community).
Few studies have examined less traditional measures of PA (i.e. sustained bouts of MVPA) and SB (sedentary bouts) in preschoolers (Andersen et al. 2017;. Sustained bouts of MVPA have been associated with more favourable health outcomes (adiposity, cardiometabolic health) in school-aged children and youth (Poitras et al. 2016), with 5-and 10-minute bouts of MVPA having similar associations with cardiometabolic health (Holman et al. 2011); but there is little evidence of health benefits of sustained MVPA in the early years (Carson et al. 2017a). Similarly, there is evidence to suggest that the manner in which sedentary time is broken up is significantly and independently associated with markers of cardiometabolic health, independent of sedentary time and physical activity, in children (Saunders et al. 2013) and adults (Healy et al. 2008;Healy et al. 2011); however, there is less evidence of the health benefits of breaking up sedentary time in the early years (Poitras et al. 2017). A study by Kuzik et al. (2015) provided an estimate of the amount of sedentary bouts Alberta preschoolers typically accumulate while in childcare; findings lead the authors to conclude that promoting SB in short bouts during the preschool years might be important for the primary prevention of overweight and obesity. More evidence around the amount (and duration) of sustained bouts of MVPA, and sedentary bouts, that preschoolers typically accumulate during the day is needed.

Statistical analyses
All statistical analyses were performed using SPSS (Statistical Package for the Social Sciences for Windows, Version 24, IBM, Inc., Chicago, USA). Descriptive statistics (mean, range) for PA and SB variables were calculated. Independent samples t-tests were conducted to examine sex-related differences in PA and SB, and the proportion of children meeting PA and step guidelines. Statistical significance was set as p<0.05.

Results
Participant characteristics are provided in Table 1. Of the 192 children who provided accelerometer data, a total of 140 (73%) had valid accelerometer data. Of those, 124 participants also had valid descriptive data (e.g. age/sex/height/weight). Analyses are therefore based on a sample of 124 children (75 boys, 49 girls). Children were on average 4.2 years old (SD =±0.6 years), with 60.5% of the sample being boys. Children wore accelerometers, on average, for 8.3 days (range = 4.0 to 9.0); average accelerometer wear time per day was 660.2 minutes (11.0 hours; range = 7.9 to 14.3). Demographic characteristics and accelerometer wear time did not differ significantly by sex.
D r a f t  Table 2). Significant sex differences emerged for sustained bouts of MVPA and SB. Boys had more, and longer, bouts of MVPA (≥5 minutes and ≥10 minutes) and spent more of their overall day engaged in sustained bouts of MVPA (p<0.05). Girls spent less of their day engaged in sustained SB than boys (p=0.009; see Table 2). D r a f t However, significantly fewer (21%) accumulated ≥6000 steps/day on each day. The proportion of children meeting guidelines did not differ significantly by sex.

Discussion
The purpose of this paper was to describe the PA and SB profiles of a sample of preschoolers aged 3 to 5 years living in Nova Scotia, Canada, and the proportion meeting age-appropriate PA and step guidelines. To the authors' knowledge, this is the first study to provide an account of the time Nova Scotia preschoolers spend physically active and sedentary, and how many meet the recommended guidelines, using objective measures (accelerometry). Our sample of preschoolers (n=124)  Our analyses provide novel insight into sustained bouts of MVPA, and SB, among preschoolers, of which little information exists (Andersen et al. 2017;Kuzik et al. 2015). Children accumulated anywhere from 5 to 25 bouts of MVPA lasting 5 minutes or more, and 1 to 12 bouts D r a f t lasting 10 minutes or more, throughout the day, with the highest bout lasting 17.8 minutes. These contributed to anywhere from 12.2 minutes, to over 3.5 hours, of accumulated MVPA per day. Boys had significantly greater, and longer, bouts of MVPA than girls, which may indicate differences in the play habits of boys and girls in our sample. As mentioned previously, boys are more likely to engage in play types of higher intensities than girls (Logue and Harvey 2009;Storli and Sandseter 2015), which may lead to the accumulation of more, and longer, bouts of MVPA throughout the day. Our data also revealed that boys accumulated significantly more, and longer, bouts of SB than girls, findings that contradict previous research indicating no significant sex differences in preschoolers' bouts of SB (Kuzik et al. 2015). Unfortunately, behavioural observations of preschoolers' PA and SB were not conducted, and therefore we cannot determine the types of PA and SB that preschoolers were engaging in, making it difficult to provide explanations for these differences. Furthermore, the context in which bouts of MVPA and SB were occurring (e.g. indoor or outdoor childcare environment) was not measured in our study. Previous work indicates that indoor environments offer a broader range of affordances for domestic play themes than rough and tumble play (e.g. play fighting and play chasing) (Logue and Harvey 2009;Pellegrini and Smith 1998;Storli and Sandseter 2015), the latter of which is more likely to be banned, particularly in indoor environments, and ultimately restricting boys and girls from engaging in physically active co-play (Storli and Sandseter 2015). Unfortunately, we do not have data on the play types encouraged/discouraged in participating childcare centres' indoor and outdoor settings. Previous literature has shown that preschool teachers are more likely to afford children with a greater license for independent mobility during free outdoor play than free indoor play, which could result in more sustained bouts of MVPA in outdoor environments. Allowing rough and tumble play and providing environmental supports for this type of play in childcare centres (particularly indoors where preschoolers spend the vast majority of their day) could lead to more sustained bouts of MVPA for both sexes (particularly girls), which could be associated with greater health benefits. Future work is needed to explore Nova Scotia early childhood educators' perspectives on supporting a range of play ; this philosophy is also integrated into the 24-hour movement guidelines (Tremblay et al. 2017b). Other research points to the many health benefits associated with short and intermittent bouts of physical activity, suggesting that the accumulation of physical activity in a highly sporadic and disordered way should enhance children's metabolic flexibility (Stone et al. 2009). Like Kuzik and colleagues (2015), our findings suggest that our sample of preschoolers seems to be exhibiting typically healthy patterns of physical activity. with those used in an analysis of Alberta preschoolers' PA and SB (Carson et. al. 2017b).
Inconsistencies between the two datasets still emerged: Alberta preschoolers aged 30 to 59 months (n=79) spent 48.6% of their day sedentary, 38.6% in LPA, and 12.8% in MVPA, whereas Nova Scotia preschoolers spent 29.2% of their day sedentary, 38.4% in LPA, and 32.4% in MVPA. These inconsistencies could be attributed to numerous confounding variables (e.g. demographic, biological, social and/or cultural factors) that influence young children's physical activity behaviour (Bingham et al. 2016) that might differ between the two respective samples. 91.8%, and 87.1%, of 3-4 year olds, and 5 year olds, respectively, met these same guidelines on an average basis, yet just 45% met these on a daily basis. In this sample (Colley et al. 2003), average steps per day (age 3-4: 9764; age 5: 10202) were slightly higher than the present study (age 3-5: 7602).

Strengths
D r a f t

16
The results presented in this paper provide the first estimate of the PA and SB of a sample of Nova Scotia preschoolers aged 3 to 5 years, an under-studied population, using objective measures, and the proportion meeting age-appropriate PA and step guidelines. Our paper is strengthened by the use of a leading, scientifically validated accelerometer (ActiGraph wGT3X+) to describe Nova Scotia preschoolers' PA and SB, and the use of accelerometer wear and data reduction decisions that are consistent with the analysis of Alberta preschoolers' PA and SB (Carson et al. 2017b). Findings are based on baseline data of a randomized, mixed methods, control study design, and probability sampling.

Limitations
The results of our study are based upon a relatively small sample of Nova Scotia preschoolers (n=124) and modest participation rate, which could indicate selection bias; results therefore cannot be generalized to Nova Scotia preschoolers. Furthermore, various correlates of preschoolers' PA and SB were not investigated. For example: several systematic reviews have explored behavioural, social, cultural, and physical environment variables that could have an influence on PA in the early years (Bingham et al. 2016;De Craemer et al. 2012;Hinkley et al. 2008), some of which include ethnicity, time spent outdoors, and parent behaviour. Future research is needed to determine whether these correlates significantly impact Nova Scotia preschoolers' PA behaviour.
The findings in this study are based upon an average of preschoolers' PA and SB for all valid days of accelerometer wear. Daily and hourly behaviours were not investigated, and we did not examine whether PA and SB varied across the week (i.e., between weekdays and weekends) or during specific periods of the day (i.e., during childcare hours and out-of-childcare hours). These analyses will be explored in future.
D r a f t Finally, we are not able to determine the types of PA and SB that preschoolers were participating in, or whether PA and SB occurred in indoor or outdoor environments. Future work is needed to understand the types of PA and SB that Nova Scotia preschoolers are participating in, and the environments in which these are occurring (e.g. indoors vs. outdoors; home vs. childcare).

Conclusion
Our sample  and for school-aged children and youth (Tremblay et al. 2016). The majority also met the step guidelines, but not on a daily basis. Future research is needed to explore possible reasons for sex-based PA differences uncovered in our analyses. More attention is also needed to explore how Nova Scotia D r a f t the project. Finally, the authors express their graduate to all study participants, including children, early childcare educators, and families, who supported data collection.

Conflict of interest
The authors have no conflicts of interest to report. D r a f t   D r a f t 1