Health Reports
Parent-Child association in physical activity and sedentary behaviour
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
by Didier Garriguet, Rachel Colley and Tracey Bushnik
Insufficient physical activityNote 1Note 2 and excessive sedentary time, particularly screen time,Note 3Note 4 are negatively associated with a variety of physical and mental health indicators in children. Fewer than 10% of Canadian children meet the current guideline of 60 minutes of moderate-to-vigorous physical activity (MVPA) per day.Note 5
Parents can influence their children’s physical activity through role modeling (being active themselves), material support (financial, logistic, co-participation), and encouragement.Note 6Note 7 The degree to which the low level of physical activity in Canadian children is attributable to the family environment is unknown. However, only 15% of Canadian adults meet the current guideline of 150 minutes of MVPA per week,Note 8 and two studies have reported that Canadian adults with children are less active than those without children.Note 9Note 10
A review of papers that examined links between parents’ and children’s physical activity yielded mixed results,Note 6Note 11Note 12 partly because of differences in the age groups studied and whether the physical activity was reported or measured. Few studies used accelerometers to investigate relationships between parents’ and school-aged (6 to 13) children’s physical activity; those that did focused on children in the older end of this age range. Three studies reported a positive association between parents’ and children’s MVPA,Note 13Note 14Note 15 while others observed a weak or no association.Note 16Note 17 Two analyses of pedometer data reported a positive relationship in steps per day for parents and children.Note 18Note 19
Associations between parents’ and children’s accelerometer-measured sedentary time also vary,Note 13Note 14Note 16 but parents’ and children’s reported screen-based behaviours (playing video games, watching television, using a computer) are more consistently associated.Note 13Note 16Note 20Note 21
While evidence of the importance of parental role modeling in physical activity is equivocal, evidence of the importance of parental support and encouragement is stronger.Note 6Note 11Note 12Note 22 Enrolling children in organized sports and activities,Note 23 paying for and/or providing transportation to activities,Note 15Note 24 and encouragementNote 15Note 25 are examples of parental factors that have been associated with greater physical activity in children.
This analysis examines associations between measured and reported parent and child physical activity and sedentary behaviour in a sample of 1,328 biological parent-child pairs from the Canadian Health Measures Survey (2007 through 2013). Parental role modeling and support for physical activity, adjusted for household lifestyle habits and socio-demographic characteristics, are considered.
Methods
Data source
The Canadian Health Measures Survey (CHMS) is an ongoing Statistics Canada survey that collects reported and measured health data from the household population aged 3 to 79. Respondents complete a questionnaire in their home and visit a mobile examination centre (MEC) for a series of physical measurements. Residents of Indian Reserves, institutions and certain remote regions, and full-time members of the Canadian Forces are excluded. More than 96% of the population is represented. Ethics approval for the CHMS was obtained from Health Canada’s Research Ethics Board.Note 26 Details are available in previous publications.Note 27Note 28Note 29Note 30
Households are randomly selected at specific data collection sites across Canada. When a household is identified, a member is randomly chosen to participate in the CHMS. If that person is younger than 12, an older member of the same household is also asked to participate. This strategy simplifies survey logistics, as it ensures that young children participating in the CHMS are accompanied to the MEC. As well, it means that data are collected for two members of the same household, in most cases, the same family.
This analysis is based on data from the first three CHMS cycles: 2007 to 2009, 2009 to 2011 and 2012 to 2013. A total of 4,152 children were sampled with another member of their household. Of these, 2,794 were biological parent-child pairs, 1,379 of whom had valid accelerometer data for both the parent and the child. Pregnant women were excluded from the study. Preliminary analyses of the accelerometer data revealed 48 outliers (more than 3 standard deviations from the mean MVPA for parent and child sex). The study examines a sample of 1,328 parent-child pairs in which the children were aged 6 to 11.
Measures
The household interview includes questions about sedentary behaviour. Parents are asked about their child ́s average time per day: 1) watching TV, videos or playing video games, and 2) on a computer (working, playing games, e-mailing, chatting, surfing the internet). Parents are also asked about their own screen-based activity: average time in a typical week during the past three months 1) on a computer, 2) playing video games, and 3) watching television, DVDs or videos. For this analysis, "screen time" variables were created for children and parents by summing the answers to these questions. For children, the mid-point of the response category selected was used in the summation of screen time (for example, 1.5 hours per day was used if the 1- to 2-hour interval was selected). The response categories for the screen-time questions were changed in cycle 3 from “1 to 2 hours” to “1 to less than 3 hours.” The same midpoint (1.5 hours per day) was used for cycle 3.
Body mass index (BMI) was calculated as measured weight in kilograms divided by measured height in metres squared (kg/m2). Height was measured to the nearest 0.1 centimetre using a ProScale M150 digital stadiometer (Accurate Technology Inc., Fletcher, USA), and weight, to the nearest 0.1 kilogram with a Mettler Toledo VLC with Panther Plus terminal scale (Mettler Toledo Canada, Mississauga, Canada). Parents’ BMI was classified as underweight/normal, overweight, or obese based on an international standard using thresholds of 25 (overweight) and 30 (obese) kg/m2.Note 31 Children’s BMI was classified as thin/normal, overweight or obese using BMI z-scores and thresholds defined by the World Health Organization.Note 32
Upon completion of the MEC visit, ambulatory respondents were asked to wear an Actical accelerometer (Phillips – Respironics, Oregon, USA) over their right hip on an elasticized belt during their waking hours for 7 consecutive days. All data were blind to respondents while they wore the device. The Actical measures and records time-stamped acceleration in all directions, providing an index of physical activity intensity via a count and step value for each minute. A valid day was defined as 10 or more hours of wear time. A valid respondent was defined as a minimum of 4 valid days. Wear time was determined by subtracting nonwear time from 24 hours. Nonwear time was defined as at least 60 consecutive minutes of zero counts, with allowance for 1 to 2 minutes of counts between 0 and 100. Published movement intensity thresholds were applied to the data to derive sedentary time and MVPA.Note 33Note 34Note 35 A complete description of the accelerometer data reduction procedures is available elsewhere.Note 5Note 8Note 28Note 29Note 30Note 36
Covariates
Covariates were divided into four groups: parental role modeling, parental support of physical activity, household lifestyle habits, and socio-demographic characteristics.
Parents’ physical activity can be regarded as an example for their children (role modeling). Parental role modeling variables were MVPA, step counts, sedentary time, and reported screen time. Associations with these movement variables overall, on weekends, and on weekdays after 3 p.m. were investigated.
The child’s participation in lessons or league or team sports was used as a proxy for parental support.Note 23 It was determined with the question, “About how many hours a week does s/he usually take part in physical activity (that makes him/her out of breath or warmer than usual): ... outside of school while participating in lessons or league or team sports.” Response categories were: never, less than 2 hours, 2 to 3 hours, 4 to 6 hours, and 7 or more hours.
Given that lifestyle practices tend to cluster within individuals,Note 37Note 38Note 39 adjustments for several factors were made in the regression analyses: obesity (child and parent BMI category), child and parent fruit and vegetable consumption (sum of the frequency of daily consumption of 100% fruit juices, fruit, tomatoes or tomato sauce, lettuce or green leafy salad, potatoes, and spinach, mustard greens or collards), smoking status of the parent (daily or occasional smoker versus non-smoker), and child exposure to second-hand smoke.
Socio-demographic characteristics were sex (child and parent), education of parent,Note 40 lone-parent household, and age of parent.
Statistical analysis
The analyses were weighted with the CHMS combined survey weights generated by Statistics Canada for cycles 1, 2, and 3.Note 41 Pearson correlations were examined in measured and reported physical activity and sedentary behaviour in parent-son and parent-daughter pairs for specific times: overall, weekends, and weekdays after 3 p.m. Linear regressions were examined, initially adjusting for the sex of the parent and the child, and interaction between the parent’s sex and the physical activity variables. Interaction terms were removed because they were not significant in any model. Models were stratified by the child’s sex. The final models adjusted for parental role modeling (movement) variables, the parental support variable, household lifestyle and socio-demographic variables. The data were analyzed using SAS 9.3 (SAS Institute, Cary, NC) and SUDAAN with 35 degrees of freedom. To account for survey design effects, 95% confidence intervals were estimated with the bootstrap technique.Note 28Note 29Note 30
Results
The sample of children (n = 1,328) is representative of the Canadian population aged 6 to 11. Their average age was 8, and they accumulated an average of 60 minutes of MVPA a day (Table 1). The parents tended to be mothers (59%), were, on average, 39 years old, and accumulated 21 minutes of daily MVPA (Table 1). The physical activity levels of parents and children in this sample did not differ from those of the complete CHMS sample (data not shown).
Parent-child accelerometer-measured physical activity and sedentary time were correlated (Table 2). For all days, correlations between parent-child MVPA (R = 0.17 to 0.24) and steps (R = 0.19 to 0.26) were similar and exceeded those for sedentary time (R = 0.11 to 0.19). However, these overall results hide differences by the child’s sex. Parent-daughter correlations tended to be higher than parent-son correlations, except for MVPA and sedentary time on weekdays after 3 p.m. These differences were driven by the mother-son correlations (R = 0.08 and 0.10 for weekends and overall MVPA correlations, respectively). Relative to the measured data for all days, parent-child correlations were lower for reported screen time and its two components (TV and computer) (Table 2).
Parental role modeling
Overall, a significant association was apparent between parent and child MVPA on all days, on weekends, and on weekdays after 3 p.m. This amounted to an extra 5 to 10 minutes of MVPA for the child if the parent was active for 20 minutes. These associations persisted for girls regardless of the time period, but only on weekdays after 3 p.m. for boys (Table 3).
Steps per day by the parent was also associated with the child’s steps for all days, on weekends, and after school (Table 3). The association was significant for boys and girls.
A significant association emerged between parent and child sedentary time overall. As well, the relationship with the parent’s reported screen time was significant overall, but not when the child’s sex was taken into account (Table 3).
Parental support for physical activity
Reported child participation in at least 2 hours of lessons or league or team sports per week was significantly related to the child’s MVPA on all days, after school, and on weekends (Table 4). Parental support for the child’s physical activity was also associated with the parent’s MVPA overall and on weekends (Figure 1), an association driven mostly by girls (data not shown).
Parental role modeling and support
Parental support for physical activity had an additive effect beyond role modeling on child MVPA (Figure 2). The effect was strongest for the least active parents (less than 10 minutes of MVPA per day) and decreased as the parent’s activity level increased. For the most active parents (more than 30 minutes of MVPA per day), the additive effect was still positive, but no longer significant.
Household lifestyle and socio-demographic factors
A child’s BMI (especially boys’) was significantly associated with their MVPA, steps, and sedentary time. Other household lifestyle factors (parent’s BMI, parent’s and child’s fruit and vegetable consumption, second-hand smoke exposure, parent’s smoking status) were not associated with the child’s physical activity (data not shown). As well, the socio-demographic characteristics were generally not significant―the only associations were with lone-parent household status in the steps-per-day model, and with the parent’s education in the sedentary time model (data not shown).
Discussion
This analysis shows that parental role modeling and support for physical activity were independently associated with the physical activity of a representative sample of Canadian children aged 6 to 11. The effect amounted to an increase of 5 to 10 minutes in a child’s MVPA for every additional 20 minutes of the parent’s MVPA. Regardless of parents’ physical activity, supporting children through enrolment in lessons or league or team sports led to further increases children’s physical activity. Whether the parent was a mother or father did not affect the associations; however, differences were apparent between parent-daughter and parent-son pairs.
Overall, parents’ MVPA was positively correlated with the MVPA of their sons and daughters. A lack of correlation was observed between mothers and sons, a finding also reported by Fuemmeler et al.Note 14 While the significant association in parent-child MVPA in the CHMS analysis confirm previous research,Note 13Note 14Note 15 comparisons of studies are limited by differences in child age, in the availability of data for one or both parents, and in how the accelerometer data were reduced and analyzed. As well, in the studies reporting a significant association between parent and child MVPA in adjusted regression models,Note 13Note 14Note 15 including the present analysis, the effect was relatively narrow, with Beta values ranging from 0.23 to 0.44.
Two earlier studies that did not report a significant association between measured parent and child MVPANote 16Note 17 noted sample biases that may explain the lack of agreement; specifically, the data were not representative of a wide range of socioeconomic groups, and the parents were predominantly mothers (more than 80%).
The CHMS analysis found that for every 1,000 steps that a parent accumulated per day, the child accumulated 200 to 350 additional steps. This is similar to a Canadian study that used pedometers and found that 1,000 steps per day by a parent equated to 195 to 439 steps per day by their child.Note 18
According to the CHMS results, parents’ and children’s measured sedentary time were related, but the magnitude was low―each additional hour of a parent’s sedentary time was associated with an 8- to 15-minute increase in the child’s sedentary time. Other research, too, has found either no association or weak associations between parents and children in measured sedentary time.Note 13Note 14Note 16
Reported screen time tends to be more consistently associated between parents and children,Note 13Note 16Note 20Note 21 but in the present study, an association emerged only for parent-daughter pairs. Differences in the presence and strength of associations between measured and reported sedentary behaviour have been noted in relation to health outcomesNote 42 and may result from the low inter-individual variability in measured sedentary time.Note 43
Parents’ support and encouragement are associated with children’s physical activity.Note 6Note 11Note 12Note 22Note 23Note 24Note 25 However, the only relevant variable in the CHMS is the number of hours a week a child participates in lessons or league or team sports, for which it is assumed that parents arrange registration and funding, and provide transportation.Note 23 The present study found that child participation in lessons or league or team sports was associated with increased physical activity not only in the child, but also in the parent. It is not known if this indicates parental involvement in the activity itself (coach, referee, volunteer), but the likelihood of co-participation in physical activity has been shown to increase when children are involved in organized sports and activities.Note 44 Although the CHMS analysis did not examine parent and child MVPA levels at the same time, the data pertained to periods when children were more likely to be with their parents (after school, evenings, weekends).
Tate et al. reported that inactive parents could overcome their lack of role modeling through effective encouragement, but for active parents, encouragement did not further augment their children’s already-high levels of physical activity.Note 15 Although the analytical approach and definition of support/encouragement in the CHMS analysis differed from that of Tate et al., the present study demonstrated that support has an additive and positive effect, regardless of the parents’ level of MVPA. The findings of Tate et al. suggest that parental role modeling and support/encouragement may be interchangeable, whereas in the present study, parental role modeling and support had independent effects on children’s physical activity.
The sex of the parent was not significant in the CHMS analysis, which is consistent with some previous research.Note 15Note 17Note 18Note 45 Others have reported a stronger effect of mothers’ physical activity on that of their children.Note 19Note 22Note 23 The explanation relates to mothers’ spending more time with their children compared with fathers and having more responsibility for providing logistic support for participation in activities.
Strengths and limitations
Among the strengths of this analysis is the large sample of biological parent-child pairs for whom accelerometer-measured physical activity data were collected. As well, the study is representative of Canadian children aged 6 to 11, uses validated cut-points for MVPA and sedentary time, and focuses on specific times when parents and children are likely to be together. Models were adjusted for several potential confounders, including household lifestyle and standard socioeconomic variables.
Nonetheless, caution is warranted in the interpretation of results. Accelerometers are limited in the measurement of water-based activities, cycling and load-carrying, and so may underestimate MVPA for both parents and children. The overall survey response rate was low, and the study sample size was further reduced by including only biological parent-child pairs. Although previous research has shown that children are more active when both parents are active,Note 14 this could not be investigated because data were available for only one parent. Household lifestyle characteristics cluster in children,Note 38Note 39 but the variables selected (for example, fruit and vegetable consumption, exposure to second-hand smoke) were not significant in any of the models. These variables were meant to represent the general value a household places on healthy living, but they may be too far removed from physical activity patterns. Participation in lessons or league or team sports is an approximation of parental support that was meant to represent facilitation (transportation, paying for equipment and materials). Although this measure has been used before,Note 23 interpretation is limited by the nature of the question.
Conclusion
In a representative sample of Canadian children, parental role modeling and support for physical activity were independently associated with children’s level of physical activity. Examining key parental influences on children’s physical activity may support the development of family-based health interventions and guidelines.
Acknowledgements
The authors are grateful to Statistics Canada employee, Andrew Quigley, for his help in preparing the child-parent paired dataset.
- Date modified: