Physical activity of parents and of their children: a systematic review of Brazilian sample studies – Report Card Brazil

The aim of this study was systematically review the researches in regard the association of PA of parents and the PA of their children. A literature search was performed in five databases (Medline, Embase, Cinahl, Lilacs and SciELO) using combined terms about youth, PA and social support, with restriction to publications with Brazilian sample and to English and Portuguese language. Thirteen studies were included in this review and was observed a positive association of parental PA and the PA of their children for the majority of the studies (n=11). Only two studies did not observed significant association between PA of parents and their children and, otherwise, it was not observed negative associations in the review. It was observed different associations according to the gender of both parents and children. This findings support the importance for the practice of PA by parents as encouragement for the practice of PA by their children among Brazilians. The strategies to promote the increase in practice of physical activity of children and adolescents of both genders may consider the social support as an important factor, especially in relation to the practice of PA of their parents.


INTRODUCTION
A worldwide prevalence of 81% of school-aged population did not reach recommended levels of physical activity in the last decade 1 .Surveys have shown a decrease in prevalence of physical activity practice over the years and according to the increase of age 2,3 .Moreover, obesity affects 3 times more people aged 6-19 years during the last three decades and faced with this alarming condition, the practice of physical activity among children and adolescents is widely considered to be an important factor in the prevention of these problems in adult life, besides promoting better health conditions in short-middle term since early ages 4 .
However, youth is a period characterized by large biological and psychological developments, which can also result in social, cultural and educational implications 5 .In this sense, the social support has been investigated for decades as a predictor of health outcomes in clinical and epidemiological approaches 6 and consists of indirect support in a verbal and nonverbal way (e.g., encouragement, information, examples of life habits and health conditions), besides the direct help, such as transportation and encouragement from parents, have been consistently associated with the practice of adolescent's physical activity 7 .In addition, children and adolescents present limitations of autonomy for daily activities that may compromise physical activity, and it may be attenuated by family actions, such as spending time together, communicating with each other and strengthening family ties 8 .
Although fathers and mothers may encourage the practice of physical activity of boys and girls in different forms, it was observed that children were more active if both parents share the responsibility of support their practice of physical activity 9 .Social support was associated with adolescents who dedicated more days to physical activity in the week compared to those who did not receive this incentive 10 .
In the social support context, the parental practice of physical activity has been considered as an important determinant for physical activity of children and adolescents 11 .However, in a country with continental dimensions such as Brazil, which has a wide diversity of climate, cultures and income distribution 12 , it becomes necessary a systematic approach of findings involving physical activity of parents and their children, aiming to guide effective strategies of health promotion in country-wide spectrum.
Therefore, the objective of this study was systemically review the findings of studies investigating the relationship between parents' and youth's physical activity among Brazilians.

METHODOLOGICAL PROCEDURES
Literature searches were performed from February to April 2018, in the following electronic databases: Medline, Embase, Cinahl, Lilacs and SciELO.We combined search terms regarding youth (e.g.children and adolescents), physical activity (e.g.exercise, motor activity, sport practice), and social support (e.g.encouragement, incentive).Our searches were restricted to publications in English and Portuguese.
Two independent reviewers performed the titles and abstracts screening of the records retrieved from the search.Then, two independent reviewers evaluated the full-texts of the potentially eligible articles following the inclusion criteria.In any case of disagreement, a third reviewer would cast the deciding vote.
We included studies investigating the association between parents'and youth's physical activity among Brazilians, regardless of the study design.Therefore, cross-sectional, longitudinal, and clinical trials were included in this review.In addition, studies were considered eligible if they had included participants until 18 years in the sample, with their respective parents, or at least provide information about them, reported by parents or children's part.Finally, physical activity assessed by subjective (e.g.questionnaires and physical activity diary) and objective methods (e.g.accelerometer, pedometer) were considered eligible.
Two independent authors extracted the following information using a standardized form: region; sample characteristics (i.e.number of participants, sex, age); study design (i.e.cross-sectional, longitudinal study); physical activity assessment (i.e.objective or subjective methods); and main results.The results of the included studies were described in narrative form and displayed in tables.

RESULTS
Our searches retrieved a total of 3,033 articles, which 210 were excluded as being duplicated and 2788 were excluded after assessment by titles and abstracts.Then, a total of 35 articles were selected to full-text review, of which 22 articles were excluded for not meet the review criteria for inclusion.Finally, 13 articles remained selected for been considered eligible in proposed inclusion criteria.Figure 1 shows the processes of this review.
The Table 2 shows the main findings of selected studies.A positive association between physical activity of parents and their children was observed by eleven studies.Christofaro et al. 11 observed that parents who practice physical activity during their childhood or adolescence increase the chance of their children practice physical activity, besides the emotional and financial incentive.Fernandes et al. 19 showed a positive association of sports practice of parents and their children, with a determinant factor for the gender.This gender difference was also observed in a study by Ramos et al. 20 , where boys were more likely to be physically active if having an adult at home who practice physical activity.It was also observed that the sports practice among boys was associated with having physically active mothers 13 .Prado et al. 14 observed in both genders that the practice of physical activity was positively associated with the frequent company of family and friends.Cheng et al. 22 showed that physical activity was differently associated according to gender, being the physical activity of the father associated with physical activity of their sons and the physical activity of the mother associated with physical activity of their daughters.Loch et al. 15 observed that the association between physical activity of children and their parents remained significant in older adolescents regard the age, and among sons and fathers regard the gender.Raphaelli et al. 16 observed that of physical activity of daughters was positively associated with physical activity of their fathers.It was observed by Wanderley Júnior et al. 23 a smaller prevalence of low physical activity level among children whose parents reported to practice physical activity.Lemos et al. 21observed that the association of physical activity of parents and their children was positively significant if both fathers and mothers reported to practice physical activity.Mendonça et al. 24 observed association of physical activity in adolescents with different kinds of social support by their parents, with increase in chance of adolescents be physically active according to accumulation of sources from parental support, such as joint participation, encouragement, transportation, watching and comments.
Two studies did not observed significant association between physical activity of children with physical activity of their parents 17,18 , although one of them presented a borderline significance level for positive association 18 .

DISCUSSION
The practice of physical activity of parents and the physical activity of their children was positively associated in the majority of the selected studies.However, it was not found a large amount of evidence among Brazilian sample studies in literature.This review found studies with samples from different regions of the country, but not from all of them.In this sense, it is important to highlight that Brazil has a wide racial miscegenation and a continental dimension, with great socioeconomic disparities in an estimated population of more than 200 million inhabitants 25 .Nonetheless, the majority of Raphaelli et al. 16 Positive PA of daughters was positively associated with have father who were physically active (OR=1,4 [95%CI=1,1; 1,8].It was not observed significant association among mothers who were physically actives and the PA of their children.

Positive
The PA level of parents influenced the PA level of their children (Parents who were physically actives [45%] vs parents who were physically inactive [22%] among physically actives adolescents, p<0,05).It was not observed significant results if only one of parents was physically active (p>0,05).
Silva et al. 13 Positive The practice in individual sports of boys was associated with have mothers who were physically actives (OR=1,28 [95%CI=1,04; 1,58]).This association was not observed in girls and either in regard collective sports in both genders.
Silva et al. 18 Null It was not observed a significant association between the PA practice of children and the PA level of parents.Although, the p-value has been borderline in general analysis (p=0,053).
Note.PA=Physical activity; OR=Odds ratio; PR=Prevalence ratio; 95%CI= 95% Confidence Interval included studies from different regions observed similar results about the practice of physical activity of parents and their children, where children and adolescents whose parents practice physical activity are more likely to also practice.These findings are convergent with observations of other studies across the world 26 .Some studies of this review observed differences between gender in regard to the association of physical activity of children and the physical activity of their parents 13,15,16,19,20 .This observation is similar to findings involving other adolescent sample, where the physical activity of girls was more strongly related with the practice of physical activity of the mothers, and in boys was related with physical activity of their fathers 9 .One possible hypothesis is that boys present higher practice of physical activity than girls, besides the fact that girls are less favored in socioecological aspects, such as individuality, family and scholar environment for practice of physical activity 28 .Another possibility is that males adult present higher practice of physical activity in leisure time than women 29 and this can reflect a more strongly association of physical activity of children and adolescents with their fathers than their mothers.
The assessment of physical activity by questionnaire for majority of the selected studies corresponds to a methodological limitation of the evidences.A previous systematic review observed that self-reported measures of physical activity can present results below or above from the objective measurements, that can importantly compromises the confiability of information 30 .Therefore, although the evaluation of physical activity by accelerometry promote slightly consistent results, the questionnaires provide complementary information such as the domains of physical activity andthe use of both instruments can provide more complete informations 31 .
Only two studies observed no significant association between physical activity of parents and their children in this review 17,18 .One possible hypothesis is that the study of Silva et al. 18 does not have a design performed specifically to assess this outcome, since it was analyzed the data from 972 participants from 20 to 69 years, of which 384 adults reported to have at least a children with age between 6 to 18 years (~40% of the sample).The other study 17 , that report no significant association, has a sample with age lower than all the others included studies, with the exception of only one 23 .In this sense, there is limited evidence in literature that parental physicalactivity is a predictor of physical activity of children with 6 years or less 32 and this relation is still dependent on the type of physical activity and on the age group 33 although the participation of parents has been the preferredcondition reported by children of 3-6 years for the practice of physical activity 34 .Another possible hypothesis for the divergent results between the two studies was the use of different methods of physical activity assessmentand were conducted in different regions of the country, where one study from South region used accelerometer and assessed only the mothers 17 and the other study from Northeast region used a questionnaire for physical activity assessment and evaluated both fathers and mothers 23 .
Only one included study evaluated a sample residing in rural area 16 .Previous study observed that the level of physical activity of children and adolescents from rural areas was significantly lower than those who residing in urban areas 35 .Raphaelli et al. 16 observed significant association between the practice of physical activity of parents and the physical activity of their children.These findings are convergent with observed in the study of Larsen et al. 36 , that evaluated a sample between 8 and 11 years from a rural community in Midwest region of United States.One possible hypothesis is that even the practice of physical activity is lower in rural areas, the family environment may exert similar influence when analyzed the parental physical activity.
In accordance to the findings of Mendonça et al. 24 , Pyper et al. 37 showed that besides the parental practice of physical activity, the social support of parents are predictors significantly related with sufficient levels of physical activity of their children, such as taking the children to places where they can be physically actives.Thus, the parents have a fundamental role in the consolidation of physically active lifestyle in early ages, which contributes to the increase of chance to become physically active adults and consequently have a lower incidence of morbidities due to sedentary lifestyle, such as obesity 38 .Despite this, it was observed that the level of physical activity of parents is lower than their nonparents counterparts and the mothers presented larger inadequate eating habits and higher body mass index when compared to women with no children 39 .
An important limitation of this review is the cut point used to identify sufficient engagement in physical activity, that is widely problematic in the literature, mainly because the studies that have composed this systematic review use a large variety of cut points and different domains (e.g., sports practice, time reported in moderate-to-vigorous physical activity, physical activity level by quartiles of dimensionless score) for both parents and children, such as the lack of specific social support definition, limiting comparisons among them.Otherwise, as practical application, this review highlight the important role of familiar environment for promote healthy habits in early ages, with the practice of physical activity by parents as encouragement of the practice of physical activity by their children.
In this sense, the education of parents contributes to higher probability of adequate growth of their children, in part for the increase of the possibility for getting more resources, been highlighted that educate parents residing in low and middle-income countries it is an investment in health 40 .

CONCLUSION
The practice of physical activity of parents was associated with physical activity engagement of their children in the most of studies.Two studies did not observed significant relationship, but otherwise, no study observed negative associations.Strategies to encourage the practice of physical activity aimed at children and adolescents might to consider the important role of family environment, with emphasis in practice of physical activity of parents, that may also differently associated with physical activity of boys and girls.

Figure 1 .
Figure 1.Flowchart of the articles selection process.

Table 1 .
Methodological characteristics of included studies.
Cross-sectional Questionnaire Note.PA Assessment=Physical activity assessment.

Table 2 .
Main findings of included studies in regard the practice of physical activity of parents and their children.