Physical activity for preschool children how much and how? 1

(1)

REVIEW / SYNTHE`SE

Physical activity for preschool children — how

much and how?

1

Brian W. Timmons, Patti-Jean Naylor, and Karin A. Pfeiffer

Abstract:Alarming trends in childhood obesity even among preschool children have re-focused attention on the impor-tance of physical activity in this age group. With this increased attention comes the need to identify the amount and type of physical activity appropriate for optimal development of preschool children. The purpose of this paper is to provide the scientific evidence to support a link between physical activity and biological and psychosocial development during early childhood (ages 2–5 years). To do so, we summarize pertinent literature informing the nature of the physical activity re-quired to promote healthy physical, cognitive, emotional, and social development during these early years. A particular fo-cus is on the interaction between physical activity and motor skill acquisition. Special emphasis is also placed on the nature of physical activity that promotes healthy weight gain during this period of childhood. The paper also discusses the strongest determinants of physical activity in preschool-age children, including the role of the child’s environment (e.g., family, child-care, and socio-economic status). We provide recommendations for physical activity based on the best avail-able evidence, and identify future research needs.

Key words:exercise, health, childhood, growth and development.

Re´sume´ :La tendance alarmante a` l’obe´site´ infantile meˆme chez les enfants d’aˆge pre´scolaire souligne a` nouveau l’impor-tance de l’activite´ physique chez ce groupe d’aˆge. Du fait de cette preóccupation accrue, il importe de de´terminer le type d’activite´ physique et la quantite´ a` pratiquer pour un de´veloppement optimal de l’enfant a` la pe´riode pre´scolaire, soit de 2 a` 5 ans. A` cette fin, nous dressons un bilan de la litte´rature scientifique afin d’e´tablir la nature de l’activite´ physique a` pra-tiquer en vue d’un de´veloppement global en sante´ (physique, cognitif, affectif et social) au cours de ces jeunes anneés. Nous portons une attention particulie`re a` l’interaction de la pratique de l’activite´ physique et de l’acquisition d’habilete´s motrices. De plus, nous analysons de pre`s la nature de l’activite´ physique propice a` la prise de poids normale au cours de l’enfance. Nous analysons aussi les principaux de´terminants de la pratique de l’activite´ physique a` cet aˆge et le roˆle de l’environnement de l’enfant : la famille, les soins de l’enfant et le statut socio-ećonomique. Nous donnons des recomman-dations en matie`re d’activite´ physique a` la lumie`re des faits solidement e´taye´s et nous proposons des pistes de recherche.

Mots-cle´s :exercice physique, sante´, enfance, croissance et de´veloppement. [Traduit par la Re´daction]

Introduction

In the context of a global childhood obesity epidemic, the importance of physical activity (PA) for this age group is re-ceiving considerable attention. Regular PA during childhood is not only important in maintaining a healthy body mass, but also brings a plethora of other physiological and psycho-social benefits (see Strong et al. 2005 for recent review). Consequently, experts in Canada and the United States have developed PA guidelines for youth, and governments and as-sociations have delivered these recommendations for several years, in an attempt to maximize the benefits derived from an active lifestyle (e.g., National Association for Sport and Physical Education 2002; Public Health Agency of Canada 2007). These guidelines have focused on school-aged chil-dren and adolescents (6 to 18 years of age); until recently (National Association for Sport and Physical Education 2002), little attention has been paid to the needs of younger children (<5 years of age). As will become evident in this paper, one reason for the absence of PA recommendations Received 3 April 2007. Accepted 17 June 2007. Published on

the NRC Research Press Web site at apnm.nrc.ca on 14 November 2007.

B.W. Timmons.2_{Children’s Exercise & Nutrition Centre,} McMaster University, Evel Building, 4th Floor, Sanatorium Road, Hamilton, ON L8N 3Z5, Canada.

P.-J. Naylor.School of Physical Education, University of Victoria, BC, Canada.

K.A. Pfeiffer.Room 3 IM Sports Circle, Department of Kinesiology, Michigan State University, East Lansing, MI 48824, USA.

1_{This article is part of a supplement entitled}_{Advancing physical} activity measurement and guidelines in Canada: a scientific review and evidence-based foundation for the future of Canadian physical activity guidelinesco-published byApplied Physiology, Nutrition, and Metabolismand theCanadian Journal of Public Health. It may be cited as Appl. Physiol. Nutr. Metab. 32(Suppl. 2E) or as Can. J. Public Health 98(Suppl. 2).

2_{Corresponding author (e-mail: timmonbw@mcmaster.ca).}

Appl. Physiol. Nutr. Metab. Downloaded from cdnsciencepub.com by 134.122.89.123 on 01/15/21

(2)

has been the lack of scientific research documenting the health effects of PA in preschool children. One likely reason limiting attention to children this young has been the as-sumption that they are sufficiently physically active. How-ever, growing evidence suggests that this is not necessarily the case.

Overweight and obesity are becoming not uncommon among preschool children. Two independent reports repre-senting separate regions in Canada estimated the prevalence of obesity among 2- to 5-year-old children at ~11% (He and Sutton 2004) and ~8% (Canning et al. 2004), respectively. Although these studies used age- and sex-specific body mass index (BMI) as a surrogate of obesity, slightly differ-ent criteria were used to establish prevalence: >95th percen-tile based on CDC growth charts (He and Sutton 2004) and values corresponding to accepted cut-off values for adults derived by the International Obesity Task Force (Canning et al. 2004). Using the International Obesity Task Force crite-ria, Shields recently reported that ~8% of 2- to 5-year-old children were obese, as defined by their directly measured heights and weights, determined as part of the Canadian Community Health Survey (Shields 2006). In light of the observed prevalence of obesity among preschool-age chil-dren and its inverse relationship to PA (Trost et al. 2003), it is appropriate to examine what we know about PA during this critical period of growth. How much PA do preschool children need? And how should opportunities for this PA be provided?

A previous attempt at formulating PA guidelines for chil-dren under the age of 5 years is available from the United States (National Association for Sport and Physical Educa-tion 2002). These American recommendaEduca-tions were intended to ‘‘provide guidance to parents, caregivers, and teachers of infants and young children about physical activity capabil-ities and needs of infants, toddlers, and preschoolers.’’ Ta-ble 1 provides an overview of the five proposed guidelines. Although these guidelines seem inherently reasonable and offer a common sense approach for those charged with the health of preschool children, what is the scientific evidence supporting these recommendations?

One purpose of this paper is to review evidence support-ing a link between PA and biological and psychosocial de-velopment during the preschool years (ages 2–5 years). We will summarize pertinent literature informing the nature of the PA required during these early years to promote healthy

physical, cognitive, emotional, and social development. A particular focus will be on the interaction between PA and motor skill acquisition. Malina (1991) has proposed that in preschool-age children general movement activities develop specific movement patterns and skills; these in turn provide the basis for acquisition of future complex skills where greater emphasis can be placed on the health, fitness, and behavioral components of physical activities. Special atten-tion will be directed to the nature of PA that promotes healthy weight gain during childhood. The paper will also discuss correlates of PA in preschool-age children, including the influence of the individual child’s environment (e.g., family, child-care, socio-economic status).

Methodology

This is not a systematic review. Given that the literature was already sparse, we decided that an overly conservative approach to identifying relevant publications would have se-verely limited the number of studies included in this paper. To optimize the retrieval of relevant publications and allow a comprehensive examination of the available literature, slightly different tactics were used for each section of this review.

For the Physical Health section, the following methods were employed:

. A literature search of PubMed used the following key-words alone and in combination: ‘‘physical activity’’ AND ‘‘preschool’’; ‘‘physical activity’’ AND ‘‘nursery’’. The term ‘‘intervention’’ was also used in combination with the above terms;

. The reference lists of relevant publications derived from the above electronic search were scanned;

. _{All issues of the journal Pediatric Exercise Science were} scanned for relevant articles.

For the Psycho-social Health section, the following meth-ods were employed:

. Relevant English-language publications were identified using Academic Search Elite, searching the following da-tabases using several terms in combination (see Table 2): PubMed, Psych Articles, Psych Critiques, Social Sciences Index, Sport Discus, ERIC, and CINAHL;

. _{Manual searching identified those articles or book} chap-ters that provided relevant information on the frequency, intensity, type or duration of PA necessary to change psycho-social outcomes.

Table 1.Guidelines for physical activity in preschool children prepared by the National Associa-tion for Sports and Physical EducaAssocia-tion.

Guideline Criteria

1 Preschoolers should accumulate at least 60 min daily of structured physical activity

2 Preschoolers should engage in at least 60 min and up to several hours of daily, unstructured physical activity and should not be sedentary for more than 60 min at a time except when sleeping

3 Preschoolers should develop competence in movement skills that are building blocks for more complex movement tasks

4 Preschoolers should have indoor and outdoor areas that meet or exceed recom-mended safety standards for performing large muscle activities

5 Individuals responsible for the well-being of preschoolers should be aware of the importance of physical activity and facilitate the child’s movement skills

(3)

For the Determinants section, the following methods were employed:

. _{A literature search of PubMed used pertinent keywords} alone and in combination (see Table 3);

. _{The reference lists of relevant publications derived from} the aforementioned electronic search, and the National Association for Sport and Physical Education Physical Activity Guidelines for children birth to five years were scanned;

. _{All issues of Pediatric Exercise Science were scanned for} relevant articles.

Definitions

The amount and nature of PA varies considerably among the studies included in this review, with no consistent opera-tional definition. The term ‘‘physical activity’’ is defined in the introductory paper of this Supplement, but additional comments are warranted relating to the pediatric age group. Intuitively, the PA of preschool children might be described more appropriately as ‘‘play’’. Burdette and Whitaker (2005) defined play as ‘‘the spontaneous activity in which children engage to amuse and to occupy themselves’’ (p. 46). Play can also be broadly defined as ‘‘a pleasurable activity en-gaged for its own sake’’(Gabbard 2004); apparently purpose-less, or without a specific goal orientation. Thus, play can be considered as a form of PA with various levels of intensity. Physical activity as play: an overview

Although researchers have described the nature, function, and types of PA play (McCune 1998; Pellegrini and Smith 1998) and also have decried the loss of time for play (Bur-dette and Whitaker 2005; Ginsburg 2007), the amount of play necessary to achieve developmental ends has yet to be identified. Pellegrini and Smith (1998) identified three types of play: rhythmic stereotypies (in infancy), exercise play (in early childhood), and rough and tumble play (in late child-hood and early adolescence). These types of play peak at different stages of development and serve differing func-tions. Peaking in the midpoint of the first year of life, rhyth-mic stereotypies are gross motor movements without describable purpose that may account for up to 40% of an infant’s activity (Thelen 1980). Rythmic stereotypies may serve to improve control of specific motor patterns and cor-respond with maturational milestones (Field et al. 1979; Pellegrini and Smith 1998). Such movements, in combina-tion with parent-initiated physical play (e.g., bouncing on the knee, playing with objects), are probably the main

sour-ces of physical activity play in early life. Such play appears closely linked to the pre-adaptive stage of motor develop-ment described by Clark (2005).

Exercise play, gross locomotor movement in the context of play, begins somewhere near the first year of life, peaks around 4–5 years of age, and declines gradually as it is re-placed with rough and tumble play during the elementary school years. The obvious function of exercise play is for physical development: strength, endurance, and enhanced economy of skill and movement. However, evidence from older children and adults suggests that it may also bring cognitive and psychological benefits. As children age (5 years +), their play activity begins to shift to rough and tumble (Pellegrini and Smith 1998), cooperative play, for-mal games, competition, and group-oriented activities (Parten 1932). Although rough and tumble play is present in the preschool years, it represents a smaller proportion of physically active play and girls are less likely to engage in it than are boys (McCune 1998). Social development is an outcome of this type of play (Pellegrini and Smith 1998), although it is debatable whether it is as influential for fe-males as for fe-males.

Rythmic stereotypies, parent–child interaction play, and exercise play are the types of play most relevant to the 2– 5 year age group. Less stimulation from caregivers (e.g., passive movement and initiation of activities) and environ-mental constraints on natural movement (e.g., car seats and restricted play spaces) can both suppress stereotypies and exercise play (McCune 1998; Pellegrini and Smith 1998; Clark 2005). When environmental restrictions are removed, children display compensatory movement, thus demonstrat-ing that movement may fulfill a functional need (McCune 1998; Pellegrini and Smith 1998).

The previous section provides an overview of PA in the context of play. Although researchers agree that play is im-portant in motor development, the amount of play needed for adequate development of motor skills is not known. However, understanding the developmental aspects of PA in the context of play should provide an essential framework when considering how to optimize PA programmes for pre-school children.

How active are preschoolers?

An increasing number of investigations have asked how active are preschool children. Because the various studies have used different methods of analysis, it remains difficult to answer this question adequately. Nevertheless, one consis-tent theme seems the nature of PA among preschool chil-dren. When preschool children are active, they tend to Table 2.Combination of search terms used to identify relevant

publications for the Psycho-social Health section of this paper.

Term 1 + Term 2 + Term 3

Play Toddler Psychosocial factors

Physical activity Nursery Social development

Exercise Preschool Child development

Motor skill development

Early childhood Self-esteem

Obesity Young child Self-concept

Self-efficacy

Academic achievement

Table 3.Combination of search terms used to identify rele-vant publications for the Determinants section section of this paper.

Term 1 + Term 2 + Term 3

Play Preschool Correlates

Physical activity Pre-School Determinants

Exercise Early childhood Factors

Motor development Young children Parents Motor skill

(4)

engage in very brief bouts of movement, spending very little time at an intensity that could be considered vigorous. Dan-ner et al. (1991), for example, directly observed the PA hab-its of 2- to 5-year-old children over an hour. They found that ~60% of the children’s time was spent sitting quietly, swinging their arms or standing still. Moderate walking or fast running was observed during only ~11% of the hour (~7 min). Using more objective PA monitoring (accelerom-etry), Pate et al. (2004) reported that preschool children spent only ~2 min/h at a vigorous intensity of PA; over 50% of the time they were sedentary or engaged in light ac-tivity. Likewise, Benham-Deal (2005) reported that 3- to 5-year-old children spent ~20% of a typical day with a heart rate above 130 beats/min (considered to reflect moderate to vigorous PA) and the majority of the children’s PA was ac-quired in bouts of 5–10 min duration. Although it is clear that young children prefer intermittent type activity, most studies also note large inter-individual variability in PA lev-els; some children are extremely active, but others only achieve low levels of PA. Given this variability, an impor-tant consideration then becomes how much PA is required to optimize healthy growth and development?

Evidence for effects of physical activity on physical health

Theoretically, establishing appropriate PA habits in early childhood should translate into positive health consequences, and evidence is accumulating that increased PA in preschool-age children is associated with improved physical health status (Binkley and Specker 2004; DuRant et al. 1993; Saakslahti et al. 2004). However, in most cases, dose–response relationships between PA and specific health outcomes in preschool children do not exist. To examine evidence for PA effects on the physical health of preschool children, we considered studies where a clear intervention was introduced and outcomes compared with a control group. Because very few such trials exist, additional evi-dence was drawn from studies that correlated PA levels with specific physical health outcomes.

Measures of adiposity

Given the ever increasing concern regarding obesity in young children, the few published interventions have fo-cused on adiposity-related outcomes (Table 4). The results have been inconsistent, with some indication that the re-sponse to a given PA intervention may differ between girls and boys.

Mo-suwan et al. (1998) conducted a 30 week study inves-tigating the effects of an exercise intervention on markers of body fatness in children (baseline age 4.5 ± 0.4 y). The in-tervention comprised a 15 min walk before morning class and a 20 min aerobic dance session following the afternoon nap, both performed 3 times per week; the control group participated in normal kindergarten activities, without the added aerobic dance. The likelihood of an increasing BMI slope (i.e., rate of BMI increase over time) was 68% less likely for girls in the intervention group than for the con-trols. In contrast, the intervention had no effect on the boys; indeed, the sum of triceps skinfolds was even slightly higher in the exercise than in the control group.

Table 4. Intervention studies in preschool children targeting adiposity-rel ated outcome s. Stud y Age (y) No. (male/female) Length Frequency Exercise Outcomes Mo-s uwan et al. (1998) 4.5 ± 0.4 82M/65F (intervention), 88M/ 57F (control) ~30 weeks 3 /week 15 min walk before morning class, 20 min aerobic dance session Girls: 68% decrease in likeli hood of having an in-creasing BMI slope; Boys: no difference in tri-ceps SF : in exercise group Fitzgibbon et al. (2005) ~4 146 (intervention), 154 (control) 14 weeks 3 /week 5 min warm-up, 10 min aerobic activity, 5 min cool-down No difference in BMI after 14 weeks; at year 1 and year 2 follow-up, : in BMI and BMI-Z score was greater in control group Fitzgibbon et al. (2006) ~4 171 (intervention), 160 (control) 14 weeks 3 /week 5 min warm-up, 10 min aerobic activity, 5 min cool-down No difference in BMI and BMI-Z score after 14 weeks, after year 1, and after year 2 follow-up Reilly et al. (2006) 4.2 ± 0.2 12 months 3 /week 30 min of PA to develop motor skills No effect on BMI SD score

(5)

Fitzgibbon and colleagues published two reports based on their Hip-Hop to Health Jr. intervention study, with contrast-ing results. In a predominantly Black group of children (Fitzgibbon et al. 2005), no differences from the control group were seen immediately after the 14 week intervention (essentially 10 min moderate to vigorous movement per-formed 3 times per week). However, 1 and 2 years later, children who had participated in the intervention displayed smaller changes in BMI and BMI-Z scores, compared with the controls. In contrast, in a population where Latino chil-dren predominated, those exposed to the same intervention did not differ from their control peers on any of the adipos-ity-related outcomes, either immediately after the pro-gramme or after 1 or 2 years of follow-up (Fitzgibbon et al. 2006).

Reilly et al. (2006) examined the impact of a 12 month programme of 30 min of exercise performed 3 times per week (no details were provided); the intent was to develop motor skill. At baseline, children were on average 4.2 years of age. The intervention had no effect on BMI SD score rel-ative to control groups. This paper raised considerable inter-est and debate, because the intervention to enhance PA levels did not lead to a significant augmentation of objec-tively measured PA relative to control groups.

An alternative approach to examining the impact of PA on adiposity-related outcomes is to follow cohorts of chil-dren over time. Klesges et al. (1995) found that baseline PA level (in comparison with others of the same age and sex) was a significant predictor of change in the BMI of 3- to 5-year-old children over the next 2 years. Likewise, Moore et al. (1995) found that relative to active children, those who were inactive at age 4.0 years were 2.6 more likely to in-crease their triceps skinfold measurement over a ~2.5 year follow-up.

Although longitudinal studies demonstrate an inverse rela-tionship between PA level and adiposity, interventions have not always confirmed these findings. Differences in length of intervention and dose of PA make comparison across studies difficult. It therefore remains impossible to deter-mine the amount of PA necessary to maintain a healthy body mass in preschool children.

Measures of bone health

Childhood is considered a critical period for accrual of bone mass. Specker and Binkley (2003) investigated the im-pact of enhanced motor activity on bone properties in chil-dren aged 4.0 years. For an average of 50 weeks, one group of children participated in 20 min of jumping, hopping, and skipping activities 5 d/week, while a second group of chil-dren did not participate in such activities. Both high and low activity groups were divided into sub-groups who re-ceived either calcium supplementation or a placebo. At fol-low-up, the increase in leg bone mineral content was more pronounced in children who had participated in gross motor activities and received calcium versus those who had partici-pated in gross motor activities but received placebo. Regard-less of calcium supplementation, however, children who had received the gross motor activity programme displayed greater periosteal and endosteal circumferences than their peers who did not receive the PA intervention. These effects persisted 12 months after the end of the intervention

(Binkley and Specker 2004). This study highlights the important interaction between PA and nutrition that is not always considered when interventions are performed with preschool children.

The findings of Binkley and Specker are supported by a cross-sectional study of preschool children (mean age 5.2 years); this found that PA levels were directly associated with measures of bone health (Janz et al. 2001). Bone min-eral density and bone minmin-eral content were both greater in those with a greater PA level, as determined by accelerome-try. A regression analysis suggested that an additional 10 min of daily vigorous activity would increase bone min-eral content at the hip and spine by 3% and 2%, respectively (Janz et al. 2001).

Measures of motor skill

When examining PA guidelines for preschool children, one important consideration is that motor skills are being developed during this period (Haywood and Getchell 2005). One approach might therefore be to assess evidence on the quantity and quality of PA required to achieve motor skill proficiency. Unfortunately, research in this area has been limited. One might argue that the development of motor skills should allow a child to become more physically ac-tive. Conversely, a child who is more physically active may better develop their motor skills. Nevertheless, the observed positive relationship between motor skills and PA (see be-low) suggests that motor skill development could be consid-ered an important health correlate.

Alpert et al. (1990) investigated effects of an 8 week aerobics programme in preschoolers. Twenty minutes of aerobic exercises (with music) were provided daily for a group of 12 children while 12 children in the control group were involved in outdoor play as part of their regular sched-ule. All children were tested before and after the interven-tion. In contrast to the control group, the aerobic exercise group showed a significant improvement in motor agility skills (e.g., walking backward five steps, walking forward on a balance beam).

Reilly et al. (2006) examined the impact of a thrice weekly 12 month programme of 30 min of exercise designed to develop motor skills (no details provided). In contrast to the limited impact on obesity (above), scores on tests of mo-tor skill improved more in the children completing the inter-vention than in those who did not receive the interinter-vention.

Unfortunately, more information regarding PA and motor skill development in healthy children is lacking. Although a particular focus of this paper was to be on the interaction between PA and motor skill acquisition, such studies proved difficult to find. Indeed, most literature in this area reports on how programmes to improve motor proficiency benefit children who already have a motor delay or disability (e.g., Apache 2005). However, correlation analyses do suggest that children with higher levels of PA have better motor skills. Saakslahti et al. (1999) found that parent-reported PA in 3- and 4-year-old children correlated positively with the children’s measured motor skills. Similarly, Butcher and Ea-ton (1989) reported that a high level of PA was related to gross motor skills and Fisher et al. (2005a) found a signifi-cant, but weak, association between habitual PA and funda-mental movement skills. The studies of Parizkova support

(6)

the positive relationship between PA and motor develop-ment (Parizkova 1996). However, correlational studies can-not establish causal relationships, and the improved motor performance in preschool children could be either a cause or consequence of PA participation.

Measures of cardiovascular disease risk factors

Signs of atherosclerosis are sometimes present even in early childhood. Thus, it is possible that an inactive lifestyle at this early age may accelerate the development of overt cardiovascular (CV) disease. Saakslahti et al. (1999) exam-ined the relationship between parent-reported PA in 3- and 4-year-old children and CV disease risk factors. The results were inconsistent: more time playing outdoors was related to higher systolic blood pressure; more very active outdoor play was associated with lower high-density lipoprotein (HDL) levels; and high levels of play were associated with lower total cholesterol levels. The association between out-door play and a high systolic blood pressure is not easily ex-plained. The authors attributed it to potential psychological influences, including the temperament of the child or situa-tional factors related to the measurement protocol. Parizkova reported the cholesterol levels of a small group of children classified by their kindergarten teachers and parents as the most active (n= 8) and the most inactive (n= 9) in terms of spontaneous PA throughout a year (Parizkova 1996). The only CV disease risk factor that differed between groups was HDL (higher in the most active group). Unfortunately, details of the amount of PA performed by the most active children were not provided.

In adult men and women (e.g., Blair et al. 1996, 1989), aerobic fitness is a strong predictor of mortality from CV disease. Aerobic fitness during childhood is related to CV disease risk in adulthood (Twisk et al. 2002). Thus, acquir-ing and maintainacquir-ing a high level of fitness through regular PA may be an excellent approach to minimizing CV disease later in life. The study by Alpert et al. (1990) that imple-mented an 8 week aerobics programme in 12 preschoolers noted that the heart rate during submaximal cycle ergometry was lower in children that received the intervention, sug-gesting improved aerobic fitness.

Summary

Based on the literature reviewed, intervention studies that have examined physical health effects in preschool-age chil-dren have increased existing PA levels by 30 min/d at most; however, most studies have employed sessions of 20 min/d, 3 times/week. This approach appears to have little impact on adiposity, but may improve bone properties, motor skills, and aerobic fitness. However, we caution that these conclu-sions rest on very few studies with relatively small numbers of participants, making it difficult to identify true effects. Evidence for effects of PA on psycho-social health

The purpose of this section is to provide an overview of evidence for the psycho-social and cognitive benefits of PA in early childhood (2–5 years). The literature identified by the search strategy and subsequent manual searching is

or-ganized into three broad areas of outcome: play and child development, psychological, and cognitive outcomes of PA.

By its very nature, PA in early childhood occurs within the guise of play. A substantial body of literature affirms the developmental importance of play and its contribution to the cognitive, physical, social, and emotional well-being of children and adolescents (Burdette and Whitaker 2005; Ginsburg 2007). It has long been believed that play has both immediate and long term functional benefits for the child (Pellegrini and Smith 1998; Piaget 1952). The sug-gested mechanisms through which play contributes include: the formation of neural structures necessary for future activ-ities (synapses and connections); practice of skills such as language, motor, and social-negotiation skills (mental and emotional mastery, cooperation, problem solving, and lead-ership skills); and serving as an important medium for child-hood expression of emotion (Eaton et al. 2001). Many of the opportunities for development that exist in play in general are also available specifically through physically active play (McCune 1998; Pellegrini and Smith 1998). Since the work of Piaget (1952), the role of PA in affording exploration and providing concrete experiences in real and pretend play is well accepted (Gibson 1988). Physically active play matters psychologically and socially (Garcia and Garcia 2002; Pelle-grini and Smith 1998).

Psychological outcomes

There has long been agreement that regular, vigorous PA confers psychological benefits such as reducing symptoms of depression, creating more positive mood states, lowering levels of anxiety, and enhancing self-esteem in both adults (Spence et al. 2005; Biddle 2000; Morgan 1985; Raglin 1990) and adolescents (Calfas and Taylor 1994); however, evidence for such effects in younger children is not as plentiful. Self-perceptions (e.g., self-esteem, self-concept, self-competence), important to performance, behaviour, and health, have received some attention in literature on children and youth (Horn 2004). Ekeland et al. (2005) conducted a systematic review and meta-analysis to determine if PA had an impact on self-esteem in children and youth. They re-viewed studies including children from 3 to 19 years of age, identifying 23 randomized studies that measured the impact of gross motor PA on self-esteem. Thirteen studies used exercise-only interventions and 12 studies adopted comprehensive interventions where exercise was a compo-nent. Their review found that a small but significant treatment effect, equivalent to a 10% difference in self-esteem between exercise and control conditions with both types of intervention (Ekeland et al. 2005). This finding supports an earlier meta-analysis (Gruber 1986). The average length of the exercise-only interventions was 130 min/week (range 60–270 min), distributed over an average of 3 d/week (range 1–5 d); that is, the equivalent of 30–45 min/d, depending on frequency (3 or 4 d/week). Ten of 13 studies tested some form of locomotor and (or) aerobic fitness exercise. The findings of the meta-analysis did not change with the exclusion of studies having shorter interventions (under 10 weeks) or a strength-training com-ponent. However, there was a lack of research on pre-school children and the conclusions cannot be applied automatically to this age group. Two of twenty-three

(7)

studies targeted preschoolers and only one study (by Alpert et al. 1990) qualified as strong evidence with a low risk of bias. Alpert’s 8 week study, on 3–5-year-olds, overviewed earlier, used aerobic classroom activities with music for 20 min, 5 times/week; it demonstrated a positive impact on self-esteem relative to controls who participated in ‘‘typical outdoor play’’ (Alpert et al. 1990). Self-esteem was assessed using an abridged Thomas Self-Concept Values Test, an assessment where the participant is asked whether the person in a polaroid photo of the child is happy, smart, etc. The other preschool study identified in the meta-analysis of Ekeland et al. (2005) was that of Platzer (1976), who investigated the 10 week impact on gross motor skills and self-concept of 30 min daily percep-tual-motor training. Self-concept, but not gross motor skills, improved in the experimental group, but not in a control group who received no specialized intervention. One of the limitations with this study was that all partici-pants entered the study with deficits in gross motor skills and self-concept. Thus, it is unclear whether similar changes would be observed in normally developing chil-dren.

Global esteem or concept has been linked to self-perceptions across a set of sub-domains (e.g., academic, social, physical), each of which also has related perceptions of competence (Horn 2004). Individuals with high global self-worth are able to discount the importance of those sub-domains in which they have lower self-perceptions (Harter 1990, 1999). This is one potential mechanism through which PA experiences in the 2–5 year age range have long-term influence on behavioural choices. Although an interesting possibility, there is as yet no evidence of a significant association between self-perceptions and PA in preschool children. During the preschool years, children’s self-evaluations tend to be very positive and relate to concrete skills and abilities (Horn 2004; Harter 1999). Preschool children use peer comparison to obtain information that helps them achieve mastery of a task (Ruble and Frey 1991). Thus, Horn (2004) emphasized the importance of providing mastery experiences with positive adult feedback. These allow children to construct high perceptions of com-petence across multiple domains, and such mastery experi-ences are also connected to learning and play (McCune 1998). The importance of mastery experiences for preschool children is exemplified by a study where developmentally delayed kindergarten children were exposed to a high mas-tery climate; they achieved and maintained higher levels of perceived physical competence when compared with chil-dren in a low autonomy intervention (Valentini 2004). Cognitive functioning

Regulation of arousal, play as a break from cognitive tasks, play as a means of distributing cognitive learning practice, development of a sense of mastery, enhanced social cognitions (negotiation, hierarchy, and emotional aware-ness), and gains in spatial cognition are all potential mecha-nisms through which physical play (exercise and rough and tumble) may influence cognitive outcomes (Pellegrini and Smith 1998). Research into the relationship between PA and cognitive outcomes in early childhood has focused largely on the interaction between motor and cognitive skill

development (Son and Meisel 2006). The early theoretical work of Piaget suggested that motor skills contributed to cognitive development by facilitating exploration of the en-vironment (Piaget 1952). Researchers have since found con-nections between locomotion and the organization of spatial information (Bai and Bertenthal 1992; Campos et al. 2000) and significant relationships between fine motor and hand-eye coordination in kindergarten and early school achieve-ment in mathematics and language (Son and Meisel 2006). Conversely, gross motor skills, although statistically signifi-cant, explained trivial amounts of variance in achievement in Son and Meisel’s study (2006). PA has been associated with short-term gains in concentration (Taras 2005) and cre-ativity in older children and adolescents. The research to date has generally utilized cross-sectional and longitudinal descriptive designs rather than experimental interventions and currently it provides little information about the dose– response relationship. What it does offer, however, is some indication that during the preschool years facilitating motor skill development within the broader framework of play and PA is potentially important if cognitive outcomes are sought.

One measure of cognitive function, academic achieve-ment, has been explored in relation to PA primarily in the school-aged population. Although there is considerable liter-ature in this area, it is unknown whether the findings can be applied to preschool children who do not receive a struc-tured curriculum per se. Therefore, we will not discuss the issue of academic achievement, although recent reviews on this topic are available (e.g., Taras 2005).

Summary

More research is needed to evaluate the effects of PA and physical play on psycho-social variables in the early years of life. In particular, there is a need for more evidence about the nature and amount of PA that benefits psychological and cognitive outcomes. However, the scant evidence cur-rently available suggests that as little as an additional 20 min of aerobics-type activity may improve aspects of self-esteem. There is a substantial amount of literature from the child- and motor-development fields about the type of activities that are most likely to promote these outcomes, and how they are best achieved.

Correlates of physical activity in preschool children

The preceding sections were concerned with how much PA is related to specific health outcomes in preschool chil-dren. Unfortunately, intervention studies in this regard are few. In contrast, numerous studies have investigated the cor-relates of PA in preschool children, using cross-sectional studies. In the following sections, we discuss what is known about the factors related to PA in preschool children. Biological and demographic

Biological and demographic factors such as age, sex, body mass or body mass index, and family risk of obesity are related to PA. Research has shown a progressive decline of PA with age in children and adolescents (e.g., Sallis et al. 2000). In the preschool population, some investigations

(8)

show a positive association of PA with age, but others show no effect of age. Jackson et al. (2003) showed a cross-sectional association between accelerometer measurements of PA and age in 3-year-old boys (but not in girls); a 1 year follow-up on a sub-sample of participants showed that PA increased in both sexes over the course of a year. The investigation continued further until participants were 5 years old, and the increase in activity with age persisted (Reilly et al. 2004). On the other hand, others have shown no age-related differences in accelerometer measurements of PA (Finn et al. 2002; Pate et al. 2004). Two direct ob-servation studies showed a decline in home-based PA with age during the preschool years (Jago et al. 2005; Nader et al. 1995), whereas simultaneously school-based activity in-creased with age (Nader et al. 1995). Reported PA patterns may thus differ, based upon environmental circumstances (e.g., home versus preschool) and method of assessment. Jackson et al. (2003) noted that their results should be in-terpreted with caution, since methods of interpreting accel-erometer data are still being honed.

Tracking is not a traditional longitudinal analysis. It re-flects individual stability in a characteristic, and represents how well a given individual maintains a rank or position rel-ative to others over time (Malina et al. 2004). In some sam-ples of preschool children, tracking of PA is poor over two years (Sallis et al. 1995), but in others it is fairly stable over three years (Pate et al. 1996). One of these investigations used direct observation of PA (Sallis et al. 1995), and the other used a heart rate index score (Pate et al. 1996). Fur-ther, one focused on both home and school activity, while the other included home environment only (Pate et al. 1996). Based on the coefficients from the heart rate study (r = 0.53–0.63) and the direct observation study (r = 0.36 for 4 d at home versus r= 0.09 for 4 d at preschool), it is possible that PA tracks better in the home than in the school environment. However, it is difficult to compare tracking coefficients from studies using different measures of PA.

Most studies report that boys are more active than girls, from the preschool age through to adolescence. Several stud-ies report this as being the case for preschoolers, regardless of method of activity assessment or study design (Bara-nowski et al. 1993; Finn et al. 2002; Jackson et al. 2003; Kelly et al. 2006; McKenzie et al. 1992; Pate et al. 2004; Reilly et al. 2004). It is not clear whether the sex difference in activity is biologically based or if it is environmentally determined; it likely reflects a combination of genetics and environment.

Body composition of the child is related to PA in pre-schoolers. Two studies indicate that children with a higher BMI are less active than those with lower BMI, but the re-sults are dependent on the child’s sex. Saakslahti et al. (1999) found that girls who played indoors for a higher pro-portion of time had a higher BMI than those who played outside, but boys who played indoors for a higher proportion of time had a lower BMI than boys who played outside. Trost et al. (2003) found that overweight boys were physi-cally less active than non-overweight boys, but there were no differences in activity between overweight and normal weight girls. Other investigators have examined the relation-ship between percent fat and PA, finding that a higher per-cent fat was associated with lower PA (Davies et al. 1995;

Moore et al. 1995). In terms of a child’s relative weight, one study by Klesges et al. (1986) found that higher inten-sity of PA was negatively associated with a child’s weight. On the other hand, in a second sample a higher relative weight was associated with higher PA (Klesges et al. 1990). The authors suggested that their unexpected finding may re-flect reactivity to direct observation, or variables related to diet. Although many PA studies have not taken dietary vari-ables into consideration, the majority of investigations have shown an inverse correlation between a child’s body compo-sition variables and PA, suggesting that activity itself is re-lated to body composition.

The BMI of the parent is also related to PA in pre-schoolers. Two studies showed a father’s BMI was a signifi-cant predictor of a child’s PA; as the father’s BMI increased, the childs PA decreased (Finn et al. 2002; Sallis et al. 1988). Similarly, Klesges et al. (1990) showed that the family risk of obesity was associated with PA; as the number of obese parents increased, the child’s PA de-creased. It is not clear if parental BMI is related to the PA of the child through a biological or environmental mecha-nism; as with sex differences in PA, a combination of genet-ics and environment is likely responsible.

A few investigations have examined associations between race or ethnicity and PA in preschool children. Pate et al. (2004) found that Black children were more physically ac-tive than White children. However, Hispanic children were said to be less active than White children (McKenzie et al. 1992; Sallis et al. 1993). Another investigation that included Black, Hispanic, and White children showed no race differ-ences in PA (Baranowski et al. 1993). The Pate et al. study assessed activity at preschool, which may have influenced the results, since the studies involving Hispanic versus White children reflected activity both at home and in pre-school. Thus, it remains unclear what relationships may ex-ist between race or ethnicity and PA.

There is only one study regarding socioeconomic status (SES) and PA for the preschool population (Kelly et al. 2006). The authors found no effect of SES on PA or seden-tary behavior. However, the participants were from Scot-land, and it is not known if such results could be generalized to other populations. Inevitably, race and ethnic-ity is tied to SES, and it is difficult to tease out the inde-pendent effects of race and SES. This may be the reason why some studies show significant effects of race and eth-nicity when others do not, and why SES may appear not to be a significant variable.

Psychosocial

Encouragement and discouragement, prompts for activity, and parental PA are psychosocial factors related to PA. However, it is difficult to examine traditional psychological variables in a preschool population, since young children are unable to complete surveys and inventories relating to these variables. Authors of one investigation compared associa-tions between PA and the teacher-reported personality char-acteristics of children (Buss et al. 1980). The results indicated that the more-active children were more energetic and restless, less inhibited, less compliant, less shy, more as-sertive, more competitive, and more manipulative than their less-active peers. No other studies have addressed

personal-Appl. Physiol. Nutr. Metab. Downloaded from cdnsciencepub.com by 134.122.89.123 on 01/15/21

(9)

ity and PA, so it is not clear how universal the results from these 129 children may be.

Much of the literature relating psychosocial variables to PA has focused on social variables, particularly the influ-ence of parents (or other adults). Encouragement and dis-couragement of PA are very similar to the concept of prompts for activity. Most of the literature addressing en-couragement–discouragement prompts comes from the Stud-ies of Child Activity and Nutrition (SCAN) project, as discussed below. Parental PA is another important variable related to preschool PA. Moore et al. (1991) determined that children who had physically active mothers were twice as likely to be active as those with inactive mothers. Chil-dren who had physically active fathers were 3.5 times more likely to be active than those with non-active fathers; how-ever, children whose parents were both physically active had an almost 6-fold increase in likelihood of being active (Moore et al. 1991). Another investigation found a signifi-cant correlation between PA and the amount of time parents spent exercising (Poest et al. 1989). Sallis and colleagues (1988) noted that parental activity was a significant predic-tor of preschool activity. The studies showing these relation-ships used three different measures of PA, (accelerometer, questionnaire, and direct observation, respectively). The demonstration of a relationship across methods of measure-ment may highlight its significance; nevertheless, some stud-ies (e.g., Trost et al. 2003) have shown no association between parental and child activity.

Environmental

The greatest amount of evidence for PA-related influences on the preschool-age population exists for environmental variables. Season, time spent outside, diet, time spent in sed-entary behaviour, preschool characteristics, and outside envi-ronment and (or) characteristics of the play space are all related to preschool PA. Studies of seasonality have shown less PA during winter (Fisher et al. 2005b; Poest et al. 1989), but there is no consistent time of year when activity is highest. Fisher et al. (2005b) suggested that PA was low-est in the spring. Kelly et al. (2006) collected data over a two-month period, finding that children were more active in September than in October. One study showed an interaction between gender and season, boys being more active outside than girls in October through December (Baranowski et al. 1993). Although available data indicate that seasonality plays a role, geographic differences are likely. Further, some investigations (e.g., Finn et al. 2002) have shown no effect of season.

Time spent outside is related to preschool PA. In the SCAN project, time spent outside was a significant predictor of PA at all three sites (Baranowski et al. 1993; Klesges et al. 1990; Sallis et al. 1993). The method of activity assess-ment in the SCAN project was direct observation; unfortu-nately, many surveys do not assess time spent outside adequately. However, studies involving older children sup-port the view that time spent outside is an imsup-portant corre-late of childhood PA (Sallis et al. 2000).

One study, which included 18-month-old children, ad-dressed the potential relationship between dietary variables and PA (Vara and Agras 1989). The authors observed a

neg-ative correlation between calories consumed and PA; chil-dren with a high caloric intake had low activity levels.

Time spent in sedentary behavior, including television (TV) watching, has been examined as a correlate of PA. TV watching per se was not an independent predictor of PA in one investigation (Jago et al. 2005); instead, minutes of time spent in sedentary activity were inversely related to PA (r= –0.7 andr= –0.8 in regression models run for two sep-arate years). A longitudinal investigation showed that chil-dren watched more TV as they became older, but there was a high degree of variability in TV watching (DuRant et al. 1996). Another study showed a negative correlation between TV watching and PA (McKenzie et al. 1992). Thus, it is un-clear if TV watching and PA are related in the preschool population; there is more literature regarding older children, but the relationship between TV watching and PA is unclear even in that population (Marshall et al. 2004).

Recent evidence suggests that a child’s PA level may be related to the practices and policies of their preschool. Both Finn et al. (2002) and Pate et al. (2004) studied children at-tending preschool, finding in regression analyses that the preschool attended was a significant predictor of PA. Nei-ther study could identify the exact characteristics of the pre-school responsible for the relationship. A follow-up to the Pate study indicated that preschools offering more field trips and having more university-educated teachers had more physically active children than preschools that did not (Dowda et al. 2004). Further, a low-quality preschool, as as-sessed by the Early Childhood Environment Rating Scale – Revised Edition, was associated with sedentary activity. Similarly, Boldemann et al. (2006) showed that children’s PA was related to the quality of the play environment. With improvements in methods of environmental assessment, more data will soon be available regarding the impact of en-vironmental factors on the PA of preschool children; how-ever, preliminary evidence indicates that it may be an important variable.

The Studies of Child Activity and Nutrition (SCAN) project

The Studies of Child Activity and Nutrition (SCAN) proj-ect was a multi-site, 3-year longitudinal investigation focus-ing on diet, PA, and cardiovascular disease risk factors in 3- and 4-year-old children drawn from different ethnic groups in three regions of the United States (Galveston, Tex.; Memphis, Tenn.; San Diego, Calif.). In this section of our paper, we discuss solely those publications derived from the SCAN, with the appropriate citation to support the spe-cific correlate of PA. Direct observation of PA was used at all sites. The Texas site’s sample consisted of low to middle SES African American, Mexican-American, and White chil-dren (Baranowski et al. 1993; Greaves et al. 1989). The Tennessee site included overweight and normal-weight White children, nearly half of the sample being of middle-to upper-class (Klesges et al. 1990). The Californian site in-cluded Mexican-American and White, low- to middle-in-come children (McKenzie et al. 1992).

Cross-sectional baseline results showed that sex, race and ethnicity, time spent outside, seasonality, TV watching,

(10)

child’s body mass, and family risk of obesity were all re-lated to the child’s PA, both at home and during recess at preschool (Baranowski et al. 1993; Klesges et al. 1990; McKenzie et al. 1992). Specifically, boys were more active than girls, and White children were more active than His-panic children. Time spent outside and child’s relative weight were positively associated with PA, whereas TV watching, family risk, and the winter or early spring months were negatively associated with PA. The investigators at the Californian site also examined a sub-sample of the low-in-come children to determine family-related variables that were related to PA during preschool recess. A regression model predicting moderate physical activity accounted for 39% of the variance and indicated that family risk of cardiovascular disease (negative association), parental vigo-rous PA (positive association), and father’s BMI (negative association) were significant predictors of PA (Sallis et al. 1988).

The longitudinal results from SCAN showed many simi-larities to the initial, cross-sectional investigations, but there were a few differences. Sedentary activity, not TV watching per se, predicted a low PA (Jago et al. 2005). Encourage-ments and discourageEncourage-ments (prompts) for PA were posi-tively related to PA, but were not predictors of PA in regression models, as in the baseline results (DuRant et al. 1996). Requests for PA and time spent outside were posi-tively related to PA, although time spent outside decreased with age (Nader et al. 1995). Further analysis of the prompts at home (Elder et al. 1998) and at recess (McKenzie et al. 1997) indicated that the amount of PA prompts from teach-ers and parents decreased over time. In contrast, prompts from peers increased from preschool to elementary school. Compliance with prompts increased in the home setting (Elder et al. 1998). Sex and race or ethnicity differences per-sisted over time, with White children being more active than Mexican-American children and boys being more active than girls (Sallis et al. 1993). Indoor and outdoor rules, con-venient play space, and time and frequency spent in play spaces were also significantly related to PA. Overall, the authors’ interpretation of the results was that PA is a multi-dimensional behaviour that depends on social and environ-mental factors in addition to time spent outside and prompts to be active.

Summary

Numerous factors, alone and in combination, can deter-mine the PA level of preschool children. Since this field of study is in its infancy, some findings are contradictory. However, the availability of outdoor spaces and time spent outside, parental interactions and modeling of activity, and the child’s sex seem important correlates of PA in preschool children.

Conclusions and recommendations

The purpose of this paper was to provide scientific evi-dence supporting a link between PA and biological and psy-chosocial development in preschool children (defined as ages 2–5 years). Despite the assumption that PA during early childhood is natural and of benefit, there is very little research to support the hypothesis that enhancing the

exist-ing level of PA among preschool children will significantly improve their health. As little as an additional 60 min/week of exercise may improve bone properties, aerobic fitness, and motor skills in some children, but much more research with larger sample sizes is required to confirm these find-ings. Surprisingly, very little information is available to demonstrate that enhanced PA controls adiposity among young children. In contrast, a burgeoning body of literature describes the determinants or correlates of PA among pre-school children. This information is particularly useful when deciding how to make preschoolers more active. Based on the literature reviewed herein, we make the fol-lowing recommendations:

. _{Promotion of PA for preschool children should consider} their natural activity patterns, which are typically sponta-neous and intermittent;

. _{PA for preschool children should focus on gross motor} play and locomotor activities that children find fun; . _{PA experiences for preschool children will be enhanced}

by adult facilitation (including modeling) that provides mastery experiences and contingent feedback about those experiences;

. Whenever possible, preschool children should be given access to play spaces and equipment outdoors

These recommendations represent a balance of current scientific evidence and our expert opinion, and may be used to inform tactics to enhance PA among preschool children. However, we emphasize that the amount and nature of PA required to optimize healthy growth and development during the preschool years is as yet unknown. The scientific evi-dence is too weak to determine how much PA a preschooler needs. It is also important to consider international reports in a Canadian context. Access to play spaces and equipment outdoors, for example, is intimately linked with seasonal ac-cess. Most regions of Canada experience extremes of both cold and heat, and weather conditions can be a significant barrier to PA for preschool children.

Future research

There are numerous areas requiring future study. Given the trends in obesity even at this young age, we recommend research be conducted in this area. Based on our review of the literature, the amount of PA necessary to maintain a healthy body mass may be greater than that necessary for a healthy bone mass, motor skill acquisition, aerobic fitness, or self-esteem. Consequently, there is a need for large-scale studies that assess PA objectively and provide indicators of physical and psychosocial health. From such studies, ap-proximations of dose–response relationships may be gleaned and recommendations for future PA guidelines can be devel-oped. In addition to measuring PA objectively, more infor-mation is necessary regarding the context of PA; it is necessary to determine which types of environment facilitate PA in this age group. In Canada, the impact of season can-not be overlooked, but it requires more research. Given the essential relationships between parents and preschoolers, fu-ture research should examine how PA for the family can be optimized. These approaches are sure to include community involvement.

(11)

Acknowledgements

This project was supported by the Public Health Agency of Canada. The leadership and administrative assistance were provided by the Canadian Society for Exercise Physi-ology.

References

Alpert, B., Field, T., Goldstein, S., and Perry, S. 1990. Aerobics enhances cardiovascular fitness and agility in preschoolers. Health Psychol.9: 48–56. doi:10.1037/0278-6133.9.1.48. PMID: 2323328.

Apache, R.R.G. 2005. Activity-based intervention in motor skill development. Percept. Mot. Skills, 100: 1011–1020. PMID: 16158688.

Bai, D.L., and Bertenthal, B.I. 1992. Locomotor status and the de-velopment of spatial search skills. Child Dev. 63: 215–226. doi:10.2307/1130914. PMID:1551326.

Baranowski, T., Thompson, W.O., DuRant, R.H., Baranowski, J., and Puhl, J. 1993. Observations on physical activity in physical locations: age, gender, ethnicity, and month effects. Res. Q. Ex-erc. Sport,64: 127–133. PMID:8341835.

Benham-Deal, T. 2005. Preschool children’s accumulated and sus-tained physical activity. Percept. Mot. Skills, 100: 443–450. doi:10.2466/PMS.100.2.443-450. PMID:15974355.

Biddle, S.J.H. 2000. Emotion, mood and physical activity.In Phy-sical activity and psychological well-being. Edited by S.J.H. Biddle, S.H. Boutcher, and K.R. Fox. Routledge, London, UK. Binkley, T., and Specker, B. 2004. Increased periosteal

circumfer-ence remains present 12 months after an exercise intervention in preschool children. Bone, 35: 1383–1388. doi:10.1016/j.bone. 2004.08.012. PMID:15589220.

Blair, S.N., Kohl, H.W., Paffenbarger, R.S., Clark, D.G., Cooper, K.H., and Gibbons, L.W. 1989. Physical fitness and all-cause mortality. A prospective study of healthy men and women. JAMA, 262: 2395–2401. doi:10.1001/jama.262.17.2395. PMID:2795824.

Blair, S.N., Kampert, J.B., Kohl, H.W., Barlow, C.E., Macera, C.A., Paffenbarger, R.S., and Gibbons, L.W. 1996. Influences of cardiorespiratory fitness and other precursors on cardiovascu-lar disease and all-cause mortality in men and women. JAMA, 276: 205–210. doi:10.1001/jama.276.3.205. PMID:8667564. Boldemann, C., Blennow, M., Dal, H., Martensson, F., Raustorp,

A., Yuen, K., and Wester, U. 2006. Impact of preschool envir-onment upon children’s physical activity and sun exposure. Prev. Med. 42: 301–308. doi:10.1016/j.ypmed.2005.12.006. PMID:16448688.

Burdette, H.L., and Whitaker, R.C. 2005. Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Arch. Pediatr. Adolesc. Med.159: 46–50. doi:10.1001/archpedi.159.1.46. PMID:15630057.

Buss, D.M., Block, J.H., and Block, J. 1980. Preschool activity le-vel: Personality correlates and developmental implications. Child Dev.51: 401–408. doi:10.2307/1129273.

Butcher, E., and Eaton, W.O. 1989. Gross and fine motor profi-ciency in preschoolers: Relationships with free play behaviour and activity level. J. Hum. Mov. Stud.16: 27–36.

Calfas, K.J., and Taylor, W.C. 1994. Effects of physical activity on psychological variables in adolescents. Pediatr. Exerc. Sci. 6: 406–423.

Campos, J.J., Anderson, D.I., Barbu-Roth, M.A., Hubbard, E.M., Hertenstein, M.J., and Witherington, D. 2000. Travel broad-ens the mind. Infancy, 1: 149–219. doi:10.1207/ S15327078IN0102_1.

Canning, P.M., Courage, M.L., and Frizzell, L.M. 2004. Prevalence of overweight and obesity in a provincial population of Cana-dian preschool children. CMAJ,171: 240–242. PMID:15289421. Clark, J.E. 2005. From the beginning: a developmental perspective

on movement and mobility. Quest,57: 37–45.

Danner, F., Noland, M., McFadden, M., DeWalt, K., and Kotchen, J.M. 1991. Description of the physical activity of young hcildren using movement sensor and observation methods. Pediatr. Ex-erc. Sci.3: 11–20.

Davies, P.S.W., Gregory, J., and White, A. 1995. Physical activity and body fatness in pre-school children. Int. J. Obes.19: 6–10. Dowda, M., Pate, R.R., Trost, S.G., Almeida, M.J.C.A., and Sirard,

J.R. 2004. Influence of preschool policies and practices on chil-dren’s physical activity. J. Commun. Health, 29: 183–196. doi:10.1023/B:JOHE.0000022025.77294.af. PMID:15141894. DuRant, R.H., Baranowski, T., Rhodes, T., Gutin, B., Thompson,

W.O., Carroll, R., Puhl, J., and Greaves, K.A. 1993. Association among serum lipid and lipoprotein concentrations and physical activity, physical fitness, and body composition in young chil-dren. J. Pediatr. 123: 185–192. doi:10.1016/S0022-3476(05) 81687-7. PMID:8345412.

DuRant, R.H., Thompson, W.O., Johnson, M., and Baranowski, T. 1996. The relationship among television watching, physical ac-tivity, and body composition of 5- or 6-year-old children. Pe-diatr. Exerc. Sci.8: 15–26.

Eaton, W.O., McKeen, N.A., and Campbell, D.W. 2001. The wax-ing and wanwax-ing of movement: implications for psychological de-velopment. Dev. Rev.21: 205–223. doi:10.1006/drev.2000.0519. Ekeland, E., Heian, F., and Hagen, K.B. 2005. Can exercise im-prove self esteem in children and young people? A systematic review of randomised controlled trials. . .including commentary by Coren E. Br. J. Sports Med. 39: 792–798. doi:10.1136/bjsm. 2004.017707. PMID:16244186.

Elder, J.P., Broyles, S.L., McKenzie, T.L., Sallis, J.F., Berry, C.C., Davis, T.B., Hoy, P.L., and Nader, P.R. 1998. Direct home ob-servations of the prompting of physical activity in sedentary and active Mexican- and Anglo-American children. J. Dev. Behav. Pediatr. 19: 26–30. doi:10.1097/00004703-199802000-00004. PMID:9524302.

Field, T., Ting, G., and Shuman, H.H. 1979. The onset of rhythmic activities in normal and high-risk infants. Dev. Psychobiol. 12: 97–100. doi:10.1002/dev.420120203. PMID:456754.

Finn, K., Johannsen, N., and Specker, B. 2002. Factors associated with physical activity in preschool children. J. Pediatr.140: 81– 85. doi:10.1067/mpd.2002.120693. PMID:11815768.

Fisher, A., Reilly, J.J., Kelly, L.A., Montgomery, C., Williamson, A., Paton, J.Y., and Grant, S. 2005a. Fundamental movement skills and habitual physical activity in young children. Med. Sci. Sports Exerc. 37: 684–688. doi:10.1249/01.MSS. 0000159138.48107.7D. PMID:15809570.

Fisher, A., Reilly, J.J., Montgomery, C., Kelly, L.A., Williamson, A., Jackson, D.M., Paton, J.Y., and Grant, S. 2005b. Seasonality in physical activity and sedentary behavior in young children. Pediatr. Exerc. Sci.17: 31–40.

Fitzgibbon, M.L., Stolley, M.R., Schiffer, L., Van, H.L., Kaufer-Christoffel, K., and Dyer, A. 2005. Two-year follow-up results for Hip-Hop to Health Jr.: a randomized controlled trial for overweight prevention in preschool minority children. J. Pediatr. 146: 618–625. doi:10.1016/j.jpeds.2004.12.019. PMID: 15870664.

Fitzgibbon, M.L., Stolley, M.R., Schiffer, L., Van, H.L., Kaufer-Christoffel, K., and Dyer, A. 2006. Hip-Hop to Health Jr. for Latino preschool children. Obesity, 14: 1616–1625. PMID:17030973.

(12)

Gabbard, C.P. 2004. Lifelong Motor Development. 4th ed. Pearson Benjamin Cummings, San Francisco, Calif.

Garcia, L., and Garcia, C. 2002. Dynamical development of self-es-teem in young children. J. Sport Exerc. Psy. 24(Suppl.): S59. [Abstract.]

Gibson, E.E. 1988. Exploratory behavior in the development of perceiving, acting and the acquiring of knowledge.InAnnual re-view of psychology. Vol. 39.Edited byM.R.R.L.W. Porter. An-nual Reviews, Palo Alto, Calif.

Ginsburg, K.R. 2007. The importance of play in promoting healthy child development and maintaining strong parent-child bonds. Pediatrics, 119: 182–191. doi:10.1542/peds.2006-2697. PMID:17200287.

Greaves, K.A., Puhl, J., Baranowski, T., Gruben, D., and Seale, D. 1989. Ethnic differences in anthropometric characteristics of young children and their parents. Hum. Biol. 61: 459–477. PMID:2807266.

Gruber, J. 1986. Physical activity and self-esteem development in children: a meta-analysis.InEffects of physical activity on chil-dren. Edited by G.S.a.H. Eckern. Human Kinetics, Champaign, Ill.

Harter, S. 1990. Causes, correlates and the functional role of global self-worth; a lifespan perspective. In Competence considered.

Edited byJ.R. Sternberg and J. Koligian. Yale University Press, New Haven, Conn.

Harter, S. 1999. The construction of the self: a developmental per-spective. Guilford Press, New York, N.Y.

Haywood, K.M., and Getchell, N. 2005. Life span motor develop-ment. Human Kinetics, Champaign, Ill.

He, M., and Sutton, J. 2004. Using routine growth monitoring data in tracking overweight prevalence in young children. Can. J. Public Health,95: 419–423. PMID:15622789.

Horn, T.S. 2004. Developmental perspectives on self-perceptions in children and adolescents. In Developmental sport and exercise psychology: a lifespan perspective.Edited byM.R. Weiss. Fitness Information Technology, Inc., Morgantown, W.V. pp. 101–144. Jackson, D.M., Reilly, J.J., Kelly, L.A., Montgomery, C., Grant, S.,

and Paton, J.Y. 2003. Objectively measured physical activity in a representative sample of 3- to 4-year-old children. Obes. Res. 11: 420–425. PMID:12634440.

Jago, R., Baranowski, T., Thompson, D., Baranowski, J., and Greaves, K.A. 2005. Sedentary behavior, not TV viewing, pre-dicts physical activity among 3- to 7-year-old children. Pediatr. Exerc. Sci.17: 364–376.

Janz, K.F., Burns, T.L., Torner, J.C., Levy, S.M., Paulos, R., Will-ing, M.C., and Warren, J.J. 2001. Physical activity and bone measures in young children: the Iowa bone development study. Pediatrics, 107: 1387–1393. doi:10.1542/peds.107.6.1387. PMID:11389262.

Kelly, L.A., Reilly, J.J., Fisher, A., Montgomery, C., Williamson, A., McColl, J.H., Paton, J.Y., and Grant, S. 2006. Effect of so-cioeconomic status on objectively measured physical activity. Arch. Dis. Child. 91: 35–38. doi:10.1136/adc.2005.080275. PMID:16239246.

Klesges, R.C., Malott, J.M., Boschee, P.F., and Weber, J.M. 1986. The effects of parental influences on children’s food intake, physical activity, and relative weight. Int. J. Eat. Disord. 5: 335–346. doi:10.1002/1098-108X(198602)5:2<335::AID-EAT2260050212>3.0.CO;2-T.

Klesges, R.C., Eck, L.H., Hanson, C.L., Haddock, C.K., and Klesges, L.M. 1990. Effects of obesity, social interactions, and physical environment on physical activity in preschoolers. Health Psychol. 9: 435–449. doi:10.1037/0278-6133.9.4.435. PMID:2373068.

Klesges, R.C., Klesges, L.M., Eck, L.H., and Shelton, M.L. 1995. A longitudinal analysis of accelerated weight gain in preschool children. Pediatrics,95: 126–130. PMID:7770289.

Malina, R.M. 1991. Fitness and performance: adult health and the culture of youth, new paradigms? In New possibilities, new paradigms? (American Academy of Physical Education Papers No. 24).Edited byR.J. Park and M.H. Eckert. Human Kinetics, Champaign, Ill. pp. 30–38.

Malina, R.M., Bouchard, C., and Bar-Or, O. 2004. Growth, maturation, and physical activity. 2nd ed. Human Kinetics, Champaign, Ill. Marshall, S.J., Biddle, S.J.H., Gorely, T., Cameron, N., and

Mur-dey, I. 2004. Relationships between media use, body fatness and physical activity in children and youth: a meta-analysis. Int. J. Obes.28: 1238–1246. doi:10.1038/sj.ijo.0802706.

McCune, L. 1998. Immediate and ultimate functions of physical activity play. Child Dev.69: 601–603. doi:10.2307/1132189. McKenzie, T.L., Sallis, J.F., Nader, P.R., Broyles, S.L., and

Nel-son, J.A. 1992. Anglo- and Mexican-American preschoolers at home and at recess: activity patterns and environmental influ-ences. Dev. Behav. Pediatr.13: 173–180.

McKenzie, T.L., Sallis, J.F., Elder, J.P., Berry, C.C., Hoy, P.L., Na-der, P.R., Zive, M.M., and Broyles, S.L. 1997. Physical activity levels and prompts in young children at recess: a two-year study of a bi-ethnic sample. Res. Q. Exerc. Sport, 68: 195–202. PMID:9294873.

Moore, L.L., Lombardi, D.A., White, M.J., Campbell, J.L., Oli-veria, S.A., and Ellison, R.C. 1991. Influence of parents’ physi-cal activity levels on activity levels of young children. J. Pediatr. 118: 215–219. doi:10.1016/S0022-3476(05)80485-8. PMID:1993947.

Moore, L.L., Nguyen, U.D.T., Rothman, K.J., Cupples, L.A., and Ellison, R.C. 1995. Preschool physical activity level and change in body fatness in young children: the Framingham children’s study. Am. J. Epidemiol.142: 982–988. PMID:7572980. Morgan, W.P. 1985. Affective benificence of vigorous physical

ac-tivity. Med. Sci. Sports Exerc.17: 94–100. PMID:3157040. Mo-suwan, L., Pongprapai, S., Junjana, C., and Puetpaiboon, A.

1998. Effects of a controlled trial of a school-based exercise program on the obesity indexes of preschool children. Am. J. Clin. Nutr.68: 1006–1011. PMID:9808215.

Nader, P.R., Sallis, J.F., Broyles, S.L., McKenzie, T.L., Berry, C.C., Davis, T.B., et al. 1995. Ethnic and gender trends for car-diovascular risk behaviors in Anglo and Mexican-Amercan chil-dren, ages four to seven. J. Health Educ.26: S27–S35.

National Association for Sport and Physical Education. 2002. Ac-tive Start: a statement of physical activity guidelines for children birth to five years. AAHPERD Publications, Oxon Hill, Md. Parizkova, J. 1996. Nutrition, physical activity, and health in early

life. CRC Press, Bocan Rotan, Fla.

Parten, M. 1932. Social play among preschool children. J. Abnorm. Soc. Psychol.27: 243–269. doi:10.1037/h0074524.

Pate, R.R., Baranowski, T., Dowda, M., and Trost, S.G. 1996. Tracking of physical activity in young children. Med. Sci. Sports Exerc. 28: 92–96. doi:10.1097/00005768-199601000-00019. PMID:8775360.

Pate, R.R., Pfeiffer, K.A., Trost, S.G., Ziegler, P., and Dowda, M. 2004. Physical activity among children attending preschools. Pe-diatrics, 114: 1258–1263. doi:10.1542/peds.2003-1088-L. PMID:15520105.

Pellegrini, A.D., and Smith, P.K. 1998. Physical activity play: the nature and function of a neglected aspect of playing. Child Dev. 69: 577–598. doi:10.2307/1132187. PMID:9680672.

Piaget, J. 1952. The origins of intelligence in children. International Universities Press, New York, N.Y.