Physical activity is beneficial for young children’s health. Positive associations between cardiovascular risk factors (26), bone health (16), motor development (9), and children’s physical activity have been reported in the literature, and recent evidence suggests that physical activity behaviors established during early childhood may track into later childhood (15) and even adolescence (17), highlighting the need to focus on physical activity promotion from a young age. Furthermore, the main form of physical activity for young children is active play, which has been shown to improve social, emotional, and cognitive skills (4,10,18). As such, several governments in countries around the world have recently developed physical activity recommendations for children under age 5 (1,29). Although these guidelines differ slightly in their recommendations, both encourage participation in 180 min of physical activity every day, including light-, moderate-, and vigorous-intensity activity.
It is a common belief among parents and caregivers that young children are naturally active and that most will easily meet physical activity recommendations (12,14). This does not seem to be the case for preschool children (3–5 yr old) (6), and no evidence exists about children in the first few years of life (i.e., toddlers age 1–3 yr). It is not currently known how much physical activity young children regularly participate in, and whether or not children engage in more or less physical activity during different periods of the day. This lack of information concerning physical activity levels in very young children is likely due, in part, to the many challenges associated with assessing physical activity in this population group (20). However, the uptake of accelerometers to measure physical activity has offered a promising means by which to objectively measure physical activity in young children. These devices are desirable because they capture movement in real time and enable objective examination of physical activity levels and patterns throughout the day (8).
Identifying current physical activity levels and the proportion of children of this age meeting the physical activity recommendations is important because it will allow us to determine whether intervention strategies to promote increased physical activity are needed in this population group or if our efforts should be focused elsewhere. Examining patterns of physical activity engagement will provide us with more information on when young children’s activity is (and is not) occurring so that we are able to identify potential “critical windows” for intervention implementation. Furthermore, understanding the influence of demographic characteristics on physical activity levels and patterns is key to the design of intervention strategies. The sex of the child and parent employment status have the potential to affect physical activity because boys are consistently more active than girls are (13), and parental employment status largely dictates the environmental structure of a child’s day (whether children are in care or not). Thus, it is desirable to examine children’s activity behaviors in light of these characteristics.
The purpose of this study was to describe the current physical activity levels and patterns of toddlers in Melbourne, Australia. A secondary aim was to determine compliance with existing Australian physical activity recommendations for this age group.
Data were drawn from the Melbourne InFANT (Infant Feeding Activity and Nutrition Trial) Program. Described in detail previously (5), the Melbourne InFANT Program was a cluster randomized controlled trial that recruited first-time parents from within existing parent groups within the Melbourne area, when children were 3–4 months of age. The intervention ran for 15 months (with the first groups recruited in June 2008 and the last groups finishing the intervention in February 2010), until the children were 18–19 months of age. During six 2-h sessions, the intervention took an anticipatory guidance approach to targeting early health behaviors including diet, physical activity, and sedentary behavior and provided parents with knowledge, skills, and strategies to promote positive behaviors. Ethical approval for the study was obtained from the Deakin University Ethics Committee and from the Victorian Government’s Office for Children. Informed consent to participate in the study was obtained from 542 parents during the baseline phase of the intervention (June to December 2008). At completion of the intervention (August 2009 to February 2010), 417 (77%) wore an accelerometer to assess differences in physical activity level between the groups. Average time spent in light-, moderate-, and vigorous-intensity physical activity was compared between intervention and control groups using a linear regression analysis, controlling for average accelerometer wear time and the mother’s group attended. Because no intervention effect was observed on physical activity between intervention and control groups (225.3 ± 41.0 vs 237.4 ± 38.6 min, P > 0.19), the data from both arms of the intervention were combined for the analyses reported below.
Physical activity was measured every 15 s for seven consecutive days using ActiGraph accelerometers (Model GT1M; Pensacola, FL). Children were required to wear the accelerometers over the right side of their hip during all waking hours, except when swimming and bathing. Parents were asked to record any time that the child did not wear the monitor and the reason for removal.
Parent surveys were completed by the child’s main carer (all of whom were the child’s mother in this analysis) and describe the characteristics of that parent only. Child (e.g., age, sex) and parent (e.g., maternal education, current employment status) demographic characteristics were obtained through the parent survey distributed at the time of accelerometer allocation. Maternal education level was used as a proxy measure of socioeconomic status and was classified as low (secondary school or below), medium (trade and certificate qualifications), or high (university degree or higher). Maternal employment status was considered “employed” if the child’s mother worked full time or part time and “not employed” if the child’s mother was on maternity leave or was currently engaged in unpaid home duties.
Accelerometer data reduction.
Accelerometer data were downloaded using ActiLife Software and initially screened for compliance with the protocol. Validated cut points of 192–1672 and >1672 counts per minute (CPM) were used to determine time spent in LPA and MVPA per day (28). Twenty minutes of consecutive zero counts were considered to be indicative of nonwear time. For a day to be considered valid, children were required to have produced counts for 444 min (7.4 h), representing nonmissing counts for at least 80% of a standard measurement day, defined as the length of time that at least 70% of the sample wore the monitor (7). This wear time is similar to the frequently used wear time of 10 h·d−1 in older children (22), after taking into account nap times of these younger children (approximately 2 h·d−1). It also corresponded to approximately 70% of parent-reported awake time.
To date, no studies have reported the number of days needed to reliably determine physical activity levels in young children. As such, reliability was estimated for LPA, MVPA, and CPM using the Spearman–Brown prophecy formula. An intraclass correlation (ICC) value was computed using random valid days from each participant to eliminate possible biases associated with selecting particular days from which to run the reliability analysis. For each participant with two or more valid days of data, a random number was assigned to each valid day. Five ICC values were calculated for LPA, MVPA, and CPM by comparing random valid days with a) at least 2 d of data, b) at least 3 d of data, c) at least 4 d of data, d) at least 5 d of data, and e) at least 6 d of valid data. As the ICC values changed (albeit slightly) when different combinations of random days were chosen, the reliability analyses were run 10 times each, and only the average ICC value (for 2, 3, 4, 5, and 6 d in each intensity) was recorded. These five ICC values were then averaged to give a final ICC value, which was used in the Spearman–Brown prophecy formula to determine the number of days of monitoring needed to achieve an acceptable reliability of >0.70 (19). It is important to note that, although the average values for 2, 3, 4, 5, and 6 d were taken, the values that were averaged differed very little and thus did not substantially influence the number of days of monitoring needed to reliably estimate physical activity in this sample.
Table 1 shows the number of days of monitoring needed to reliably estimate toddlers’ physical activity. Four days of monitoring provided acceptable (>0.70) reliability for LPA and CPM and high (>0.80) reliability for MVPA while maintaining a reasonable sample size (n = 295). Four valid days of data was therefore set as the inclusion criterion for the physical activity levels and pattern analysis in this study. Because this study focused on the time spent in LPA and MVPA rather than the CPM, the number of days of monitoring needed to reliably estimate toddlers’ physical activity using CPM was calculated to inform future studies interested in using CPM as a main outcome measure.
Complete data were obtained from 295 children (158 boys and 137 girls). The main reasons for exclusion from the analyses were no data available (n = 69) or insufficient valid data available (n = 53). Time spent in LPA and MVPA per day was determined for each child as described previously and averaged across the number of valid days. In addition, accelerometer data were analyzed using customized Excel macros to examine the data by hourly periods. This allowed for the determination of the time spent in LPA and MVPA for each hour between the hours of 6:00 a.m. to 8:00 p.m. The time frame of 6:00 a.m. to 8:00 p.m. was chosen because it represented the hours during which a minimum of 20% of the sample would be included in the analyses (as determined from the examination of a random subsample of children) and represented the logical times during the day for interventions to be implemented. To be included in the pattern analyses, children were required to have worn the accelerometer for 50% of each hour (i.e., 30 min) (24) and have valid accelerometer data for the whole day (i.e., a wear time of >444 min). A 30-min wear time criterion was established for the pattern analyses to attempt to reduce the potential underestimation of toddlers’ physical activity associated with napping, a characteristic of this young population. The time that each child spent in LPA and MVPA during the same valid hourly period (e.g., 9:00 to 10:00 a.m.) was averaged over all valid days; the average value was used in the final analysis.
Descriptive statistics were initially calculated for all measured variables. To assess differences in LPA, MVPA, child age, child sex, and maternal education, exploratory one-way ANOVA (LPA, MVPA, and child age) and χ2 analysis (sex and maternal education) were conducted to assess differences between children included and excluded from the analysis. The main analysis consisted of a one-way ANOVA, where differences in daily and hourly LPA and MVPA were examined, with sex and maternal employment status as the independent variables. Data were analyzed using SPSS version 17.0 (SPSS, Inc., Chicago, IL). Statistical significance was set at P < 0.05.
Table 2 reports the demographic and physical activity characteristics of the study population. There were no significant differences in demographic variables and MVPA between included and excluded children, although included children engaged in approximately 40 min more of LPA a day than excluded children did and were, on average, 1 month younger than their excluded counterparts (19.1 vs 20.1 months).
Physical activity level.
Table 3 outlines the minutes spent in LPA and MVPA for boys and girls. There were no significant differences in the time spent being physically active between boys and girls and between children whose mothers were employed or not employed, for either intensity. In total, 90.5% of the children met the physical activity guidelines.
Physical activity patterns.
Figure 1 depicts the time spent in LPA and MVPA for this sample between 6:00 a.m. and 8 p.m. The periods during the day with the highest physical activity were between 8:00 and 10:00 a.m. and between 4:00 and 5:00 p.m. The most active hour was 4:00 to 5:00 p.m., with children spending 20.3 ± 4.5 min in LPA and 5.4 ± 2.6 min in MVPA. Children were least active in the early morning (6:00 to 8:00 a.m.), around midday, and after 5:00 p.m.
Figure 2 illustrates the time spent in LPA and MVPA per hour for boys and girls separately. Boys engaged in significantly more LPA than girls did between 6:00 and 7:00 a.m. (13.4 vs 9.4 min), between 7:00 and 8:00 a.m. (15.4 vs 13.3 min), between 8:00 and 9:00 a.m. (18.5 vs 17.2 min), and between 9:00 and 10:00 a.m. (18.8 vs 17.5 min) and in more MVPA than girls did between 6:00 and 7:00 a.m. (3.3 vs 2.2 min), between 7:00 and 8:00 a.m. (3.8 vs 2.9 min), and between 8:00 and 9:00 a.m. (4.5 vs 3.8 min), respectively.
Figure 3 shows the time spent in LPA and MVPA by maternal employment status. Significantly higher LPA was demonstrated by children with a mother who works full or part time during 8:00 to 9:00 a.m. (18.8 vs 16.2 min), 9:00 to 10:00 a.m. (18.9 vs 16.8 min), 10:00 to 11:00 a.m. (19.3 vs 17.9 min), 3:00 to 4:00 p.m. (19.3 vs 17.6 min), and 4:00 to 5:00 p.m. (20.8 vs 19.4 min), whereas the opposite occurred during 6:00 to 7:00 p.m. (18.1 vs 16.8 min) and 7:00 to 8:00 p.m. (18.0 and 15.2 min). Significantly higher MVPA was found in children with a mother who was employed during 9:00 to 10:00 a.m. (4.6 vs 4.0 min) and 10:00 to 11:00 a.m. (4.9 vs 4.1 min) and in children with a mother at home during 7:00 to 8:00 p.m. (5.5 vs 4.4 min).
Wear time between 8:00 a.m. and 6:00 p.m. varied very little (53.7–57.7 min·h−1), with the exception of the first two periods of the day (6:00 to 8:00 a.m.), during which mean wear times of 45.8 and 49.4 min·h−1 were observed.
To the best of our knowledge, this study was the first to examine the physical activity levels and patterns of toddlers using objective measures and to determine compliance with age-appropriate physical activity recommendations. Most children met the current physical activity recommendations, although this was mainly achieved through light-intensity physical activity. This finding suggests that previous studies reporting parental perceptions that young children are highly active (12,14) is likely a perception of the time spent in LPA, rather than MVPA, given that children only spent, on average, 47 min in MVPA per day.
Although there were no significant differences in total daily LPA or MVPA between boys and girls in this study, boys’ MVPA tended to be higher than girls’ (P < 0.06). Previous reviews have shown that boys are more active than the girls are in the preschool (13) and primary school (27) years, and it seems this phenomenon is emerging as early as 19 months of age. Furthermore, when comparing the number of minutes spent in MVPA per day in this study to the meta-analysis of Bornstein et al. (3), examining physical activity levels in preschool children, it seems that daily MVPA increases by about 5 min for boys between the ages of 19 months and 3–5 yr but remains the same for girls. Although it is difficult to determine whether this phenomenon is biological, sociological, or both, this finding has important implications for future intervention strategies because MVPA may need to be targeted in girls from a very early age. Future studies warrant investigation into potential reasons why approximately 10% of children did not meet the current physical activity recommendations.
When looking at patterns of LPA and MVPA throughout the day, boys’ and girls’ activity displayed a remarkably similar pattern characterized by increasing activity in the morning, which drops down around midday, and then increases in the afternoon until about 4:00 p.m. Compared with other studies, this pattern of activity is quite unique, as it typically occurs in reverse in primary school populations (at least on weekends when they are not limited by the school day), where activity levels increase up until midday and then decrease thereafter (23). One consistent finding when compared with previous physical activity pattern studies in primary school children (22,23) was that boys were significantly more active than girls were in the morning hours. It is difficult to determine whether the periods of high activity for toddlers are those in which children tend to be most active naturally, if these are the periods that children have the most opportunity to be active, or both. However, this pattern analysis does highlight times of the day when physical activity levels could potentially be increased, particularly for MVPA, because a maximum of only 5.5 min was spent in MVPA in any given hour. Future research should investigate whether it is most advantageous to encourage physical activity during the periods when children are already the most active and may be more responsive to physical activity promotion efforts or during those times when decreased physical activity levels occur.
This study also examined the daily physical activity patterns of children based on maternal employment status because this would largely dictate whether children spend the majority of their day in childcare or at home. Children with mothers who were currently employed outside the home exhibited significantly higher LPA and MVPA levels mid morning and mid afternoon, whereas children whose mothers did not work outside the home exhibited the highest physical activity in the early evening. Thus, although different patterns of physical activity engagement were observed across the day, children engaged in similar levels of daily physical activity regardless of maternal employment status. This suggests that physical activity patterns may differ according to environmental opportunities and that, although different periods of the day do not contribute equally to daily physical activity engagement, similar levels are observed when considering daily physical activity. One possible explanation for this finding is that children have an innate set point for physical activity (termed the activitystat), suggesting that, if physical activity engagement during one part of the day is low, physical activity engagement is increased in other parts of the day to maintain the innate set point (25). The present data suggest that compensation of physical activity across the day may be occurring; however, to date, this hypothesis has not been investigated in children of such a young age, and it has not been consistently supported when examined in research with older children (2,11,30). Longitudinal or experimental research is needed to examine this hypothesis further.
In addition, this study examined the reliability of physical activity data in a young age group. We found that toddlers needed 4 d of valid data to provide acceptable (>0.70) reliability for all physical activity levels, similar to what Penpraze et al. (21) reported for a comparable wear time criteria in preschool children using CPM. Our study, however, examined reliability of LPA and MVPA and found that MVPA was less variable in this age group than LPA. This was likely influenced by the fact that little MVPA was undertaken each day compared with LPA, although it may also suggest that interventions aimed at increasing MVPA may carry over into daily activity patterns more consistently than light activity.
Strengths of this study include a large sample size and an objective measure of physical activity in a population whose physical activity has not previously been investigated. Furthermore, despite the potential logistical challenges of using accelerometry to measure physical activity in young children, 70% of our sample had valid, reliable data to use. However, the number of children included in each period analysis differed, because not all children met the 50% wear time criteria for each period (largely because of the variability in nap times). This was unavoidable, however, because children of this age are still napping at different times during the day. In addition, reduced mean wear time during the first two periods of the day may have influenced the lower physical activity levels observed from 6:00 to 8:00 a.m.
In summary, this study contributed to our understanding of physical activity levels and patterns in young children who are just beginning to establish physical activity behaviors. In this study, most children met the current Australian physical activity recommendations for toddlers; however, 10% did not, which could be of some concern at this young age. Furthermore, the daily physical activity level and pattern of toddlers were consistent between boys and girls, although boys did accumulate more LPA and MVPA during the morning hours. Because low levels of MVPA were observed in children across the day, interventions are warranted in this population. On the basis of the hourly data presented, interventions delivered at certain times of the day may hold more promise than others will.
Jill Hnatiuk was supported by a Deakin University International Postgraduate Research Scholarship. Nicola Ridgers was supported by an Alfred Deakin Postdoctoral Research Fellowship. Jo Salmon was supported by a National Heart Foundation of Australia and Sanofi-Aventis Career Development Award. Karen Campbell was supported by the Victorian Health Promotion Foundation. Kylie Hesketh was supported by a National Heart Foundation of Australia Career Development Award. The authors also thank Trina Hinkley and Eoin O’Connell for assistance with the reliability analyses for this article. The authors would like to thank the National Health and Medical Research Council, Heart Foundation Victoria, and Deakin University for funding support and the InFANT staff members, the Maternal and Child Health Nurses in each participating community, and the parents and children who graciously participated in the InFANT program. No conflicts of interest are declared by the authors. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Australian Department of Health and Ageing. Discussion Paper for the Development of Physical Activity Recommendations for Children Under Five Years
. Canberra (ACT): Australian Department of Health and Ageing; 2008. pp. 1–191.
2. Baggett C, Stevens J, Catellier DJ, et al.. Compensation of displacement of physical activity in middle-school girls: the Trial of Activity for Adolescent Girls. Int J Obes (Lond)
. 2010; 34: 1193–9.
3. Bornstein DB, Beets MW, Byun W, McIver K. Accelerometer
-derived physical activity levels of preschoolers: a meta-analysis. J Sci Med Sport
. 2011; 14 (6): 504–11.
4. Burdette HL, Whitaker RC. Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Arch Pediatr Adolesc Med
. 2005; 159 (1): 46–50.
5. Campbell K, Hesketh K, Crawford D, Salmon J, Ball K, McCallum Z. The Infant Feeding Activity and Nutrition Trial (InFANT) an early intervention to prevent childhood obesity: cluster-randomised controlled trial. BMC Public Health
. 2008; 8 (1): 103.
6. Cardon GM, De Bourdeaudhuij IMM. Are preschool children active enough? Objectively measured physical activity levels. Res Q Exerc Sport
. 2008; 79 (3): 326–32.
7. Catellier DJ, Hannan PJ, Murray DM, et al.. Imputation of missing data when measuring physical activity by accelerometry. Med Sci Sports Exerc
. 2005; 37 (Suppl 11): S555–62.
8. Chen KY, Bassett DRJ. The technology of accelerometry-based activity monitors: current and future. Med Sci Sports Exerc
. 2005; 37 (Suppl 11): S490–500.
9. Cliff DP, Okely AD, Smith LM, Kim M. Relationships between fundamental movement
skills and objectively measured physical activity in preschool children. Pediatr Exerc Sci
. 2009; 21 (4): 436–49.
10. Colwell M, Lindsey E. Preschool children’s pretend and physical play and sex of play partner: connections to peer competence. Sex Roles
. 2005; 52 (7/8): 497–509.
11. Dale D, Corbin CB, Dale KS. Restricting opportunities to be active during school time: do children compensate by increasing physical activity levels after school? Res Q Exerc Sport
. 2000; 71 (3): 240–8.
12. Dwyer GM, Higgs J, Hardy LL, Baur LA. What do parents and preschool staff tell us about young children’s physical activity: a qualitative study. Int J Behav Nutr Phys Act
. 2008; 5: 66.
13. Hinkley T, Crawford D, Salmon J, Okely AD, Hesketh K. Preschool children and physical activity: a review of correlates. Am J Prev Med
. 2008; 34 (5): 435–41.
14. Hinkley TBA, Salmon JP, Okely ADE, Crawford DP, Hesketh KMD. Influences on preschool children physical activity: exploration through focus groups. Fam Commun Health
. 2011; 34 (1): 39–50.
15. Janz KF, Burns TL, Levy SM. Tracking of activity and sedentary behaviors in childhood: the Iowa Bone Development Study. Am J Prev Med
. 2005; 29 (3): 171–8.
16. Janz KF, Burns TL, Torner JC, et al.. Physical activity and bone measures in young children: the Iowa Bone Development Study. Pediatrics
. 2001; 107 (6): 1387–93.
17. Kristensen PL, Møller NC, Korsholm L, Wedderkopp N, Andersen LB, Froberg K. Tracking of objectively measured physical activity from childhood to adolescence: the European Youth Heart Study. Scand J Med Sci Sports
. 2008; 18 (2): 171–8.
18. Lindsey EW, Colwell MJ. Preschoolers’ emotional competence: links to pretend and physical play. Child Study J
. 2003; 33 (1): 39–52.
19. Mattocks C, Ness A, Leary S, et al.. Use of accelerometers in a large field-based study of children: protocols, design issues and effects on precision. J Phys Act Health
. 2008; 5 (1 suppl): S98–S111.
20. Oliver M, Schofield GM, Kolt GS. Physical activity in preschoolers: understanding prevalence and measurement
issues. Sports Med
. 2007; 37 (12): 1045–70.
21. Penpraze V, Reilly JJ, MacLean CM, et al.. Monitoring of physical activity in young children: how much is enough? Pediatr Exerc Sci
. 2006; 18 (4): 483–91.
22. Riddoch C, Mattocks C, Deere K, et al.. Objective measurement
of levels and patterns of physical activity. Arch Dis Child
. 2007; 92: 963–9.
23. Ridgers ND, Graves LEF, Foweather L, Stratton G. Examining influences on boys’ and girls’ physical activity patterns: the A-CLASS project. Pediatr Exerc Sci
. 2010; 22: 638–50.
24. Ridgers ND, Timperio A, Crawford D, Salmon J. Five-year changes in school recess and lunchtime and the contribution to children’s daily physical activity. Br J Sports Med
. 2011. doi:10.1136/bjsm.2011.084921.
25. Rowland TW. The biological basis of physical activity. Med Sci Sports Exerc
. 1998; 30 (3): 392–9.
26. Sääkslahti A, Numminen P, Varstala V, et al.. Physical activity as a preventive measure for coronary heart disease risk factors in early childhood
. Scand J Med Sci Sports
. 2004; 14 (3): 143–9.
27. Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc
. 2000; 32 (5): 963–75.
28. Trost SG, Fees BS, Haar SJ, Murray AD, Crowe LK. Identification and validity of accelerometer
cut-points for toddlers. Obesity (Silver Spring)
. 2011. doi:10.1038/oby.2011.364.
29. UK Department of Health. Making the Case for UK Physical Activity Guidelines for Early Years
. London (UK): Department of Health; 2011. pp. 22–4.
30. Wilkin TJ, Mallam KM, Metcalf BS, Jeffery AN, Voss LD. Variation in physical activity lies with the child, not his environment: evidence for an ‘activitystat’ in young children (EarlyBird 16). Int J Obes (Lond)
. 2006; 30: 1050–5.