Factors Associated with Physical Activity in Kindergarten Children

Purpose:  

Reversing obesity is an urgent public health need for children: worldwide, in 2011 more than 40 million children under the age of five were overweight ¹. Increasing intensity and duration of physical activity (PA) is a strategy to aid in preventing and reducing obesity.  According to limited available research, PA in kindergarten children is decreasing below recommended guidelines and sedentary behavior is increasing ²;the potential to change the obesity trajectory, however, is greater in early childhood when healthy behaviors are “imprinted” and entrenched ³.  More than 11,207,000 children in the United States attend preschool/kindergarten programs that primarily provide sedentary activity, a level highly associated with childhood obesity ^4-7.

A lack of research to describe physical activity precisely and to identify factors associated with PA regarding obesity of children fostered this study of a small population of children to help determine and clarify such patterns and factors. The purpose of this study was two-fold, first to describe the patterns and duration of daily PA of kindergarten children at different levels of intensity (sedentary, light, moderate-to-vigorous, and vigorous) during a normal day in public kindergarten.  Second, to identify the most predictive enabling factors (body mass index (BMI), motor skills, playtime on play equipment) and supporting factors (parental self- efficacy, parent modeling, teacher modeling, perceptions of competence, and enjoyment of PA) associated with duration of objectively measured moderate to vigorous and vigorous physical activity during the school day in 4 to 7 year old kindergarten children while controlling for demographic factors (age, gender, and ethnicity). 

Methods:

A cross-sectional observational study identified and described factors associated with the duration of PA at different levels of intensity.  A sample of 38 children in a full day  public kindergarten setting from five classrooms of one elementary school in southeast Texas was studied between September 2011 thru May 2012 school year.

Rather than observational measurements, an Actigraph GTX3+ accelerometer more accurately and objectively measured the duration of PA at different levels of intensity. Questionnaires and other factor-specific tools measured factors including body mass index (BMI), motor skills, duration of access to play equipment, parental support, and parental perceptions of competence and enjoyment, and parent and teacher PA behaviors.  This study used the Preschool and Kindergarten Physical Activity Promotion Model modified by this researcher for the younger child from Welk’s Youth Physical Activity Promotion Model ⁸  to conceptualize  these factors.

PA was measured and averaged from three 6-hour school days in 15-second epochs for an average daily measurement. Each epoch was categorized by intensity levels with the following cut-points per 15-second epoch: < 373 counts for sedentary (S), ≥ 373 to <420 counts for light (L), ≥ 420 to < 824 counts for moderate to vigorous(MV) and ≥ 824 counts/15 second epochs for vigorous (V). Anthropometric measures and motor skills were measured on the same school day the following week. Anthropometric measurements followed Lohman, Roche, and Martorell ⁹ procedures. Motor skills measurement followed Williams ¹⁰ protocol for the Children’s Motor Skill Tool.  The Questionnaire on Physical Activity for Parents of Kindergarten Children and the Questionnaire on Physical Activity for Teachers of Kindergarten Children, were developed by the researcher based on McMinn’s ¹¹ Southampton Women’s Survey and were used to measure other factors. 

PA was described by the average daily duration of PA at different levels of intensity by gender and classroom and a time series approach was used to describe the patterns of PA over the 6-hour school day.  Pearson Correlation was used to identify factors correlated with MV and V PA. Multiple regression was used to identify predictive factors.

Results:

The 38 children’s ages ranged from 5.14 to 7.1 years old with a mean age of 6.12 ±0.142 years. Females represented 60% of the sample. The prevalent ethnicity was African American (93%). Forty-seven percent of children lived with both mother and father.  Based on BMI using U.S. norms (normal BMI 14-17 m/kg²for 4-7 year-olds, age and gender specific);16% were overweight and 16% were obese. Mothers’ mean BMI was 30 (SD 5.03) and teachers’ mean BMI was 27.79 (SD 8.6).

The PA intensity measurements of the children were sedentary/light for 91% of the school day.  The overall mean duration of moderately vigorous (MV)/and vigorous (V) PA was 36 minutes (95% CI 31 to 40). Patterns analyses indicated an increase in MV and V from 10 a.m. to 11 a.m. (free play) and from 1 p.m. to 2:30 p.m. (classroom guided PA and dismissal preparation, including turns at toileting and free play).

Enabling factor child’s zBMI ranged from -1.77 to 2.27 with a mean of 0.59 (SD 1.06). Females (0.79) had a higher average zBMI than males (0.29).  For motor skills, males and females total mean score was 129.4 (SD 13.29) and 119.22 (SD 14.76).  The factor access to play equipment had little variability and was not used in subsequent analyses. 

Supporting factor parental self-efficacy scored 4.06 (SD 0.74) on a 5-point Likert scale. Parent modeling and teacher modeling self-report mean of 178 minutes/day and 84 minutes/day, respectively. Parent self-report data was out of range of probability and was not used for further analyses.  Supporting factors parental perceptions of their child’s competence and enjoyment of PA and were measured using a dichotomous, and a 5-point Likert scale respectively. Seventy- four percent of the children were rated more active than their peers and the mean of child’s PA enjoyment mean was 4.49 (SD 0.77). Exploratory ANOVA analysis to investigate statistical significance between classrooms and MV and V PA as the dependent variable were F (4) = 6.439 with a significance of p = 0.001 between classrooms.

Ethnicity was the only demographic factors with a p< 0.20 and was retained to control for confounding.  The remaining enabling and supporting factors were entered into a regression analysis.  Teachers PA hr/day was the only statistically significant variable retained with 14% of the variance in MV and V PA can be explained by teacher PA while controlling for ethnicity..

Conclusion:

This is one of the first studies to describe the patterns and duration of daily PA of kindergarten children at different levels of intensity during a normal day in public kindergarten using accelerometers. The two main findings were the patterns and duration of physical activity associated with brief periods of scheduled playtime;and the statistical significance of teacher PA predicting MV and V PA.   While primarily sedentary during the school day, short durations of increased PA intensity spiked during free and guided play and dismissal preparation highlighting the need to incorporate increases of higher intensity level PA at other times during the school day.  The significant variation in minutes of MV and V PA among classrooms, and the association of the teacher’s self-reported PA on the dependent variable point to the potential for increasing children’s PA by intervening with teachers and the need to take into account the clustering effect within classrooms in future studies.  Limitations of the study include the small sample size and lack of racial/ethnic diversity.   Strengths of the study include precision of PA measurement and use of an age-appropriate conceptual model.  Further research is needed to replicate this study in a larger, more diverse sample.  Questions remain regarding predictive factors to increase and influence PA levels among children not meeting recommended intensity levels of PA in kindergarten settings.