Individual and School Correlates of Adolescent Leisure Time Physical Activity in Quebec, Canada

Background: Leisure time physical activity (LTPA) correlates have been mostly studied in relation to adolescents’ home neighbourhoods, but not so much in relation to the environment of their schools’ neighbourhoods. We sought to investigate how objective environmental measures of the schools’ vicinity are related to adolescents’ self-reported LTPA. Methods: Individual data from the Quebec High School Students Health Survey (QHSSHS) were matched with schools’ socioeconomic indicators, as well as geographic information system-based indicators of their built environments. Self-reported levels of LTPA during the school year were assessed according to intensity, frequency and index of energy expenditure. Associations per gender between covariates and LTPA were estimated using ordinal multilevel regression with multiple imputations. Results: Boys (21% of which were highly active) were more active than girls (16% of which were highly active) (p ≤ 0.01). The incremental variance between schools explained by the contextual variables in the final models was higher among girls (7.8%) than boys (2.8%). The number of parks or green spaces within 750 m around their schools was positively associated with student LTPA in both genders. Conclusions: The promotion of parks around schools seems to be an avenue to be strengthened.

Q1 During the school year, did you do these activities?

Physical activity frequency
The value given for the answers to questions Q2 and Q3 (frequency per week) correspond to the choice of the respondent (example: 2 days a week = 2) 2. Physical activity duration Table S1 specifies the duration assigned according to the choice of answers to questions Q2 and Q4 (time per day). The value "0" hour is assigned for the first choice of answer (less than 10 minutes), according to the present recommendations. Also this choice refers to the total duration in a typical day and not to a particular session throughout the day. So this choice answer (less than 10 minutes) automatically classifies the individual in the sedentary level. The central value of the time interval is assigned for the other five choices.

Physical activity intensity
For the level of physical effort (intensity), a value, in METs, is attributed to each of the response options (Table S2).

C. Calculation of the IEE
The IEE is obtained in the following way: frequency x duration x intensity. Example: 5 days / week x 0.83 h / day x 4 METs (4 kcal /kg / hour*) = 16.6 kcal / kg / week (achieved IEE).

D. Characterisation of students' level of LTPA
So, for this example (intensity of 4 METs, frequency 5 days / week and IEE of 16.6 kcal / kg / week), the level physical activity would be "moderately active" (see Table S3). Finally, the sedentary level is automatically assigned if the answer choice is "Less than 10 minutes" to question Q4.
The three criteria (intensity, frequency and IDE) must be respected to be classified at a given level. More details are provided in the following references: [23][24].

Supplementary Material 2: Definition of Individual Variables a) Levels of LTPA
• 'Active': The required volume (≥30 kcal·kg −1 ·week −1 ) must be reached with a frequency of five or more practices per day/week and an intensity of more than 3 MET. This volume (≥30 kcal·kg −1 ·week −1 ) is the equivalent of the WHO's international recommendation for energy spending for 5 to 17-year-olds (e.g. ≥60 minutes/day, of moderate to high intensive activity, every day).
• 'Moderately active': A volume of ≥15 kcal·kg −1 ·week −1 and ˂ 30 kcal·kg −1 ·week −1 , reached with a frequency of three or more practices per day/week and an intensity of more than 3 MET.
• 'Mildly active' and 'barely active': A volume of ˂ 15 kcal·kg −1 ·week −1 , reached with a frequency of at least one practice per day/week with no minimal intensity requirements • 'Sedentary': Less than one practice per week or no activity whatsoever. This category also includes students reporting less than 10 minutes of PA in a typical day. Nonetheless only a small proportion of students belong to this group (1.1%). In our study, the level "Active" was renamed "Highly active". Also, the categories "barely active" and "mildly active" were combined in one "Slightly active" category since taken separately, they are clinically meaningless.

b)
Educational curriculum: Information about the type of the educational curriculum (general curriculum versus other curriculum) was also sought from the person in charge of the school, for each selected class. The other types of curriculum are mainly for adolescents with learning disabilities. c) Satisfaction with body image: A valid and usable visual tool was used to assess their current and desired physical appearances [2][3]. This tool was a seven-unit gradual scale ranging from very thin to very strong body shape. Then an index of satisfaction with their body appearance was derived by making the difference between the current and the desired appearance. The satisfaction index included five categories: (1) 'satisfied' when the student selected the same shape twice, meaning he/she desires to maintain his/her appearance; (2) 'desire for a slight weight loss' when he/she chose a slimmer figure as his/her desired appearance with a gap of 1 unit; (3) 'desire for a major weight loss' when his/her desired appearance was slimmer with a gap ≤ 2 units. (4) 'desire for a slight weight gain' when his/her desired appearance was larger with a gap of 1 unit; (5) and 'desire for a major weight gain' when his/her desired appearance was larger with a gap ≥ 2 unit. d) Family situation: Students were asked to report their type of family life: (1) two-parent family (students living with both biological or adoptive parents); (2) stepfamily (students living with their mother or father and the mother's/father's spouse); (3) single parent family (students living with their mother or father only); (4) shared custody (students whose parents are separate and equally sharing their time with them); (5) and special situations (tutoring, family or foster home; roommates, independent, etc.). e) Family employment status: The parents' employment status was chosen to be an indicator of the economic status of students' families. This indicator was defined in three ways: (1) both parents at work; (2) only one parent working; (3) and both parents unemployed. f) Clustered harmful behaviours: Several harmful behaviours to the adolescents' health such as smoking, alcohol and illicit drug consumption were also collected in the QHSSHS. The smoking status indicator was constructed according to the number of cigarettes smoked during life and the frequency of consumption reported by students during the thirty days preceding the survey. A typology of alcohol and illicit drugs also based on frequency of use reported by students was used. Finally, information on the perception of the students' health as the only psychosocial-level variable was also composited.