Study design, setting and study population
In this cross-sectional observational study, 30 out of the total 51 municipal preschools within the Södermalm district of Stockholm Sweden, were invited to participate. Preschools were chosen to reflect a representative sample of the different environmental characteristics (outdoor operation and different size of the playground) within the Södermalm district. All children between 3-5 years of age, at the participating preschools, were invited to participate. Informed consent was obtained from all participating children’s parents and preschool teachers and the study has been approved by the Stockholm Ethical Review Board (EPN), Dnr: 2018/890-31/2. The fieldwork measurements, comprising questionnaires for preschool teachers and parents, body measures of children and 7 days of accelerometer measures of PA in children and preschool teachers, were carried out at the participating preschools from September to November 2018.
Preschool environmental characteristics, policies and time outdoors
The Environment and Policy Evaluation Self-Report (EPAO-SR) Instrument, showing good to excellent validity and reliability [19], was administered to preschool teachers to measure environmental characteristics and formalized PA polices in the participating preschools. Subscales of the EPAO-SR were used for analyses. Preschool playground size was classified into 4 categories; (1) ≤200m2, (2) around 900m2, (3) >2700m2 and (4) outdoors activity (all time at the preschool is spent outdoors). Formalized PA policy was analyzed as a dichotomous variable (Y/N), depending on if the preschool had any written policy concerning PA or not. Time in-and outdoors was aggregated from in-out report, in which preschool teachers recorded time spent “indoors” or “outdoors” in 30-minute periods for every child on all weekdays during the PA measurements. Time outdoors was thereafter converted into quartiles (Q1 <138min, Q2 138min≤ to <187.5min, Q3 187.5min≤ to <234min and Q4 ³234min), where Q1 comprises those 25% of preschool children who spent the least time outdoors and Q4 comprises those 25% who spent the most time outdoors.
Body measures
Weight and height of participating children were measured via validated scales and stadiometers, respectively (calibrated scale: VB2-200-EC, Vetek AB, Väddö, Sweden; portable stadiometer: Seca 213, Seca, Chino, CA, USA). Body mass index (BMI) was classified as normal, overweight or obese according to an international classification by Cole et al., correcting for age and sex [20].
Physical activity and sedentary time
PA and ST was measured via the triaxial Actigraph GT3X+ accelerometer, which has been tested extensively for reliability and validity and is widely used in epidemiological pediatric research [21]. Wear protocol and analyzing techniques followed best practices and used the latest recommendations to increase accuracy [21]. Children and preschool teachers were instructed to wear the accelerometer, at the right hip, all waking hours for 7 consecutive days. A sampling rate of 60 Hz was used, and data were downloaded in 10‐s epochs and thereafter reintegrated to 60‐s epochs for analyses. We analyzed vector magnitude (Vm) activity counts (Vm = √ (X2 + Y2 + Z2)). Accelerometer data were considered valid if the child wore the accelerometer for at least 3 days during all preschool hours. Non‐wear time was defined as 60 or more consecutive minutes with zero counts, allowing up to 2 min of interruptions with non‐zero counts. Steps were determined using the manufacturer’s step algorithm, using the normal filter. MVPA, light physical activity (LPA) and ST were calculated based on cut‐offs and epochs developed specifically for the GT3X+ accelerometer, using Vm activity counts, in 4‐year‐old children [22]. ST was calculated as any minute of less than 820 counts per minute (cpm), LPA as 820-3907 cpm and MVPA as ³3908 cpm. For preschool teachers, MVPA, LPA and ST were calculated based on cut-offs and epochs developed by Santos-Lozano et al. for the GT3X+ accelerometer, using Vm activity counts [23]. ST was calculated as any minute of less than 150 cpm, LPA as 150-3207 cpm and MVPA as ³3208 cpm.
Family characteristics
At baseline, parents filled out a questionnaire on demographical and descriptive variables on anthropometry (height and weight) and highest education level, categorized into elementary school, upper secondary school and university education.
Teacher PA
Teachers’ PA outcomes were aggregated at preschool level by calculating the means of the respective PA outcomes of all teachers in each preschool. Every outcome was then categorized into high and low by the median. Cut-offs for teacher PA outcomes aggregated at preschool level were defined as: MVPAlow<24.5 min, LPAlow<310.4 min, Stepslow<6656 steps, STlow≤184.5 min.
Statistical analyses
Descriptive analyses included the distribution (means and standard deviation (SD)) of various background characteristics and PA outcomes by preschool policy, playground size, time outdoors and teacher PA.
Next, we used Linear Mixed Models (LMM) to examine associations between existence of formalized PA policy, playground size, time spent outdoors and preschool teachers aggregated levels of MVPA, LPA, steps and ST with child levels of MVPA, LPA, steps and ST, with a random intercept at preschool-level to account for similarities in child levels of PA within preschools. We analyzed each association between the exposures and outcomes independently and all predictors simultaneously in both unadjusted and adjusted models. Adjustments, in all models presented, were made for age of the child, sex and BMI which have been shown to associate with both PA (outcome) and preschool characteristics (exposure) [24].