Correlates of physical activity enjoyment in children and adolescents for a new perspective on the treatment of overweight: A systematic literature review

The purpose of this systematic literature review was to systematically compile the state of knowledge on correlates of physical activity enjoyment in children and adolescents to influence the perspective of future physical activity promotion approaches especially for children and adolescents affected by overweight or obesity. The electronic database search was executed in the five databases PubMed, PsychINFO, SPORTDiscus, Web of Science, and BISp‐SURF, from inception to December 6, 2021. A semi‐quantitative method was used for summarizing the resulted correlates. For final analysis, 85 studies comprising 48,144 children and adolescents were included. Fifty‐seven variables could be coded for their relationship with physical activity enjoyment. Of these, 12 psychological variables, for example, the basic psychological needs, task orientation, or self‐efficacy; six interpersonal variables, for example, peer/group acceptance, parental support, and autonomy support; and one behavioral variable, the higher self‐reported physical activity, are consistent positively associated to physical activity enjoyment. A scientifically based overview could be extracted for the promotion of physical activity enjoyment in children and adolescents. There is a gap in literature focusing the perception of physical activity enjoyment in the subgroup of children and adolescents affected by overweight or obesity. Therefore, recommendations were made to enable the development of further innovative research approaches in this population.

"Ending Childhood Obesity" is a major goal of the World Health Organization (WHO) 1 and an important contributor to the achievements of the Sustainable Development Goals Agenda of the United Nations (UN), as obesity is a risk factor for many noncommunicable diseases. 1It has a detrimental effect on the physical and psychological well-being of children and is associated with secondary diseases and impairments for life. 1 The prevalence rates of overweight in children and adolescents aged 2-19 years range from 13% in developing countries to 23% in developed countries. 2Worldwide, more than 50 million girls and 74 million boys are affected by obesity. 3st of the recommendations for therapy and prevention of overweight and obesity include physical activity behavior as one important component of weight management in children and adolescents. 1,4,57][8] Nooijen and colleagues 9 conducted a meta-analysis about the effectiveness of interventions on physical activity in children affected by overweight or obesity and concluded that existing interventions are not useful to increase physical activity behavior and called for "other treatment strategies." 9 One outcome of these insights is the suggestion that it may be more effective to focus physical activity-related interventions on reducing inactivity and sedentary behavior rather than on promoting physical activity, which includes moderate to vigorous physical activity, as well as exercise and sports. 4,9The assumption is, concentrating on reducing the risk factor inactivity through activities of everyday living is easier to implement and better for a successful weight management than promoting all types of physical activity, including sports, as there are many more barriers to maintaining them over a long period of time. 4In contrast, physical activity, especially higher-intensity activities like exercise, organized sports, or moderate to vigorous recreational activity, may have a positive impact on children's and adolescent's mental health, executive functioning, or cardiometabolic health [10][11][12] and should not be disregarded.Therefore, other options are needed to promote the individual physical activity behavior in children and adolescents.
One so far underestimated viewpoint on promoting physical activity is to facilitate children's and adolescent's access to movement experiences through positive emotions.Concentrating on positive emotional states during physical activity, like the perception of physical activity enjoyment, could give innovative interventions a new perspective beyond the goal of weight loss and possibly increase their effectiveness.Physical activity enjoyment is defined as "a positively valanced emotion directed towards physical activity associated with feelings such as pleasure, joy, and fun." 13 According to the selfdetermination theory, 14 an intrinsic motivational state is important to fulfill the perception of self-determination and thus sustain a certain behavior.In contrast to the motive of weight loss, which tends to characterize an extrinsic motivational state, enjoyment is an attribute of intrinsic motivation and is intended to help maintain physically active behavior over the long term. 14,15Deforche and colleagues 16 and Ekkekakis et al. 17 also presumed that physical activity enjoyment, pleasure, or fun is particularly important for individuals affected by overweight or obesity and could increase their participation in all types of physical activity 16 by minimizing negative exercise experiences and counteracting the development of avoidance. 17wever, the research of physical activity enjoyment is wide and heterogeneous, and it is currently very hard to give a clear, systematic answer to the question of which factors affect the perception of fun or pleasure in children and adolescents during physical activity and exercise.Therefore, the purpose of the following systematic review is to present the state of knowledge on correlates of physical activity enjoyment in children and adolescents in order to influence the perspective of future research approaches especially for the physical activity promotion in children and adolescents affected by overweight or obesity.

| METHODS
This systematic review follows the PRISMA 2020 statement (Preferred Reporting Items for Systematic reviews and Meta-Analyses) 18 and the PERSiST guidance (implementing Prisma in Exercise, Rehabilitation, Sport medicine and SporTs science). 19All action steps were documented using the PRISMA checklist (see Table S1).The electronic database search was executed in the five databases PubMed, PsychINFO, SPORTDiscus, Web of Science, and BISp-SURF, from inception to December 6, 2021, without restrictions.The search term was developed with specialists from the library and includes Medical Subject Headings (MeSH) and Thesaurus terms.All five searches with modified search terms for the different databases are documented in Table S2.The selection process began with the title screening by one researcher based on the inclusion and exclusion criteria described below.Then the abstracts were screened for eligibility.All remaining full texts were read completely and excluded with reasons, if necessary (see Figure 1).A priori inclusion and exclusion criteria were formulated according to the main objective of the study, which was divided into the following five categories: a. Population: studies with healthy children or adolescents, with and without overweight or obesity aged between 3 and 17 years, were included.The mean age of the target group was decisive.Other studies addressing adults or children and adolescents with disabilities or diseases were excluded.b.Dependent variable: the dependent variable of all included studies had to be physical activity enjoyment, which means, according to the definition described above, "a positively valanced emotion directed towards physical activity associated with feelings, such as pleasure, joy, and fun." 13 Due to heterogeneous methods of assessments, no explicit instrument for determining physical activity enjoyment was decisive, but studies that examined affective states or moods that were not comparable with the characteristics of an emotion 20 were excluded.c.Setting: to consider the whole range of correlates in the literature, studies conducted in any kind of physical activity settings were included, that is, organized sports, leisure time physical activity, and physical education.Studies performed in virtual settings were excluded.d.Independent variables: there were no limitations regarding the kind of relationship with physical activity enjoyment; that is, studies that examined correlates, moderators, mediators, determinants, or predictors were included to increase the likelihood that all relevant studies could be found.To evaluate the factors in the syntheses, they were then summarized at their lowest level of relationship as correlates.
e. Types of studies: observational studies were included.Studies with a qualitative, explorative, or narrative design were excluded, as well as dissertations, books, and expert opinions.All studies had to be written in English or German.
One reviewer summarized the resulting data and then extracted the measure of the dependent variable physical activity enjoyment and the bivariate, as well as the multivariate significant or insignificant relationship with all independent variables collected.To deal with this heterogeneous pooling of independent variables and to achieve greater consistency across studies, only the lowest relationships, that is, mainly univariate analyses like analyses of variance (ANOVAs), t tests, and Pearson's correlations, were further synthesized to combine all independent variables as correlates of physical activity enjoyment.To analyze the resulting correlates, a semi-quantitative method established in determinant research 21 was executed in a five-stage process.First, according to previous reviews, [21][22][23] all variables were classified into one of four categories, that is, (1) demographic and biological, (2) psychological and emotional, (3) behavioral attributes and skills, and (4) social and interpersonal factors.In the second stage, variables with similar meaning but different theoretical backgrounds were combined if the number of studies was too small (<3) to examine them individually.These summarized variables were marked with an asterisk (*).Like other systematic reviews within determinant research, 21 only variables that were examined at least three different times were presented in the table of results.
The third stage included labeling the independent variables assessed in subgroups.If the correlation with physical activity enjoyment was reported for a certain variable for males and females only or separately, the related study IDs were labeled with (m) for male or (f) for female.A similar procedure was performed for age-related subgroups.The subgroup of children with a mean age of less than 12 years was labeled with (c), and the subgroup of adolescents 12 years and older was not additionally labeled.No other subgroup analyses were conducted in the included studies.In the fourth stage, the independent variables were sorted according to their statistically significant positive, negative, or insignificant relationship to the dependent variable physical activity enjoyment.Finally, all associations for each correlate of physical activity enjoyment were combined into an association code according to the established proceedings first described by Sallis and colleagues. 21The coding rules are as follows: -0-33% of the studies supporting a similar direction of relationship with physical activity enjoyment can be summarized as no association and coded with "0"; F I G U R E 1 Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 flow diagram. 1834-59% of the studies supporting a similar direction of relationship with physical activity enjoyment can be summarized as inconsistent and coded with "?"; -60-100% of the studies supporting a similar direction of relationship with physical activity enjoyment can be summarized as having a positive or negative relationship and coded with "+" or "-." 21 Double coding, like "++," "--," "??," or "00," was used for each coding rule if four or more articles supported one similar statement of association. 21 assess the certainty in the methodological quality of the included studies, an adapted checklist was used that was based on the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) Statement, 24 the modified Downs & Black Checklist from Rhodes and Kates, 15 and the criteria of Zhou and Wang. 23The newly compiled checklist contains 11 items in the categories background and introduction, study type, population, measured variables, data analysis, and discussion (see Table S3).

| RESULTS
After reviewing all records identified in the databases and removing duplicates, 3043 titles remained for title and abstract screening.As presented in Figure 1, 191 studies were sought for retrieval, and 170 full texts were then analyzed and assessed for eligibility.The main reasons for excluding full texts were the absence of the dependent variable physical activity enjoyment and a language other than English or German.Two studies from the reference lists of the full texts also met the inclusion criteria and were included in this systematic review.
The descriptive information about all studies is summarized in Table 1.Eighty-five studies comprising 48,144 children and adolescents were included in the final analysis.Most of them, 37,262 in 65 studies, were adolescents with a mean age of 12 years and older.S4).Different outcome measures for the analysis of physical activity enjoyment were used.Thirty-one out of 85 articles refer to the Sport-Commitment-Model 26,29 and 17 articles to the Intrinsic Motivation Inventory, 27 and each applied the subscale for enjoyment.Other commonly used instruments are the Physical Activity Enjoyment Scale (PACES) 25 and the Satisfaction/Enjoyment/Boredom in Sport Scale. 28In general, the majority of studies selected specific items and did not use the full instrument set of the abovementioned scales.To assess certainty, all 85 cross-sectional and longitudinal studies were rated with 1 (item is fulfilled) or 0 (item is not fulfilled) for each item of the 11-item checklist (Table S3).None of the included studies had a value lower than 6.Twenty-nine included studies had a rating of 9 or more, and 56 studies had a rating between 6 and 8 points.
In 85 articles, 57 variables associated with physical activity enjoyment were evaluated in at least three different studies (see Table 2).
Of these, 10 variables had no relationship, 33 were positively associated, and nine were negatively associated to physical activity enjoyment.A statement about relationship was not possible for five variables because of inconsistent findings.Most of the variables found could be assigned to the psychological and emotional category.
Within this category, an important, that is, double-coded positive relationship to physical activity enjoyment was found for perceived competence, (perceived) self-efficacy, ability and incremental beliefs,    2).In total, three variables were double-coded as unrelated to physical activity enjoyment, that is, younger age, task-avoidance goals, and ego orientation.All variables, including the single-coded variables, are presented in Table 2.
As mentioned before, only nine studies included weight-related measurements in their articles, either as a correlate (BMI) of physical activity enjoyment or as a target population (children or adolescents who are overweight or obese).When considering body weight as a correlate of physical activity enjoyment, two different methods were used in the included studies.First, BMI was calculated and treated as a continuous variable with its mean mostly within the range of normal weight.For this variable, a higher BMI was not associated with physical activity enjoyment and was double-coded with "00" (see Table 2).
The second method was to calculate a categorical variable based on BMI and compare the normal weight group with the overweight or obese group.This variable was coded with "-"in the table of results (see Table 2), which indicates a negative correlation of overweight with physical activity enjoyment.Only two included studies investigated correlates of physical activity enjoyment in a group of youth affected by overweight, 31,49 what prevented a separate presentation and analysis of correlates for children and adolescents who are overweight or obese.The first study looked for associations between psychological stress, sex, ethnicity, BMI percentiles, and physical activity enjoyment. 31On the basis of a regression analysis, they found that only psychological stress had a negative impact on enjoying physical activity (β = À0.41,p < 0.01). 31The second study examined the differences between subjectively and objectively assessed physical fitness in relation to physical activity enjoyment, as well as the relationship of physical activity enjoyment to parental support. 49The authors presented significant positive correlations between physical activity enjoyment and subjectively assessed physical fitness, like perceived cardiorespiratory fitness (0.44, p < 0.01), perceived flexibility (0.26, p < 0.01), and perceived physical strength (0.31, p < 0.01). 49However, they found no correlation outcome for their objectively measured counterparts.Only objectively assessed flexibility (0.18, p < 0.05) correlated slightly with physical activity enjoyment.Perceived parental support (0.23, p < 0.01) was directly related to physical activity enjoyment, and at the same time, this correlation was mediated by the youths' perceived cardiorespiratory fitness and flexibility but not by objectively assessed physical fitness. 49 view of the different developmental stages, data for children and adolescents were presented separately (see Table 2).Only 18 variables could be evaluated for children.Most of the coded results were similar to the overall group and the group of adolescents, but the perceived ego-involving climate was double-coded as negatively associated with physical activity enjoyment for children and not associated, coded with "0," for adolescents.In addition, the male sex, introjected motivation, and the moderate to vigorous physical activity were positively associated for adolescents and not associated with physical activity enjoyment or inconsistent for children.

| DISCUSSION
The main purpose of this study was to systematically compile the state of knowledge on correlates of physical activity enjoyment in children and adolescents.It was possible to present a detailed overview and assign all correlates to different categories with consistent, that is, double-coded findings in three areas, summarized in Figure 2.
Most of the variables could be evaluated for the second category on psychological and emotional factors.Therein, consistent with previous findings, the different types of motivation and achievement-related factors are important correlates of physical activity enjoyment, 115,116 which demonstrate a certain connection of physical activity enjoyment and the three underlying theories self-determination theory, 14 mastery motivation theory, 117 and achievement motivation theory. 118The results underpin Deci and Ryan's 14 attribute of bivariate relationship between modes of intrinsic motivation and physical activity enjoyment and support the hypothesis that basic psychological needs are not only important for intrinsic motivation but also for perceptions of physical activity enjoyment.In addition, task orientation and task-approached goals correlate with physical activity enjoyment, whereas ego orientation or ego-approached goals were unrelated to physical activity enjoyment.According to our analysis, one of the most important variables related to physical activity enjoyment is perceived competence, which is consistent with its prominent role within the theory of mastery motivation. 117Due to the additional consistent positive correlation of physical activity enjoyment and the higher self-reported physical activity (see Table 2), it can be assumed that enjoyment plays a role in explaining behavior, in this case physical activity behavior.Nevertheless, further research is needed to understand the correlation between the behavioral theories and physical activity enjoyment or to interpret a causal relationship.Furthermore, also the social network, that is, family, friends, teachers, and their support, are relevant for perceiving enjoyment of physical activity.It seems important to consider them when developing effective strategies to promote the physical activity enjoyment in children and adolescents.
The additional purpose of this study was to influence the perspective of future research approaches towards physical activity enjoyment, especially for the physical activity promotion in children and adolescents affected by overweight or obesity.Demographic and biological factors play a minor role in this systematic review, with only the overweight/obesity variable negatively correlated with physical activity enjoyment.This finding is consistent with previous research of psychosocial differences between adolescents with and without overweight, 16,119 which suggests that it could be more difficult to promote physical activity enjoyment in the group of youth who are overweight or obese.However, there is a lack of knowledge regarding the enjoyment of sport and physical activity in children and adolescents affected by overweight or obesity.Only two included studies have examined correlations with physical activity enjoyment in this specific population, and it was not possible to close the research gap.The first study found only psychological stress negatively related to physical activity enjoyment, 31 and remarkably, the second study underlines a certain influence of subjectively assessed physical fitness parameters in contrast to their objectively measured counterparts. 49These findings support previous explanations from Ekkekakis and colleagues 17 that psychological and psychosocial barriers seem to be even more important in the group of individuals who are overweight or obese, through decreasing physical activity enjoyment, diminishing the intrinsic motivation, and subsequently leading to a higher avoidance of physical activity. 17The results of this systematic review also found worry, anxiety, parental pressure, entity beliefs, and for children the perceived ego-involving climate to be negatively correlated with physical activity enjoyment.In this context, it can be assumed that these psychosocial barriers also play an even greater role in children and adolescents affected by overweight or obesity.

| Limitations
There are several limitations to consider when referring to the results of our review.In particular, the small number of studies investigating correlates of physical activity enjoyment in children and adolescents who are overweight or obese limit the transferability of the resulted correlates for this target group, and we can only assume implications for further research.Furthermore, due to various variables, sample sizes, analysis strategies, and measurement approaches, it was not possible to summarize the items with statistical effect sizes.All included studies were observational in nature, and the analyzed variables are correlates, giving a first overview of all possible associations.
Due to these bivariate correlations, the causal relationship cannot be proven.Some consistent findings were presented, and implications can be made, but they must be verified through randomized controlled trials or other interventional studies.Additionally, due to the observational study design, only a certainty analysis for methodological quality was conducted and not an original risk of bias analysis as recommended in the PERSiST guidance. 19Finally, it was possible to separate the target group adolescents from the target group children, but because of significantly fewer studies with children, the expressive power from this subgroup analysis is limited.

| Practical implications
On the basis of the results of this systematic review and the reflections made in the context of the target group, we tried to summarize a few important recommendations for further research.There is a gap in literature focusing on the feelings and emotions during physical activity in children and adolescents who are overweight or obese.
Future studies should therefore be conducted to gain more insights into this topic, allowing to design interventional approaches where the children and adolescents find intrinsic enjoyment in the physical activity behavior itself.In this context, it would be interesting to explore the exact differences between children and adolescents who are overweight and those who are not, in order to make appropriate recommendations for the promotion of physical activity enjoyment.
One possibility is that especially the psychosocial barriers found in this systematic review like anxiety, pressure, boredom, and amotivation need to be considered differently in the two subgroups.However, another possibility is that there are no differences between them, and all children need to be addressed exactly in the same way to experience enjoyment.
Altogether, this systematic literature review should be used as a starting point for future research and intervention approaches to positively influence the physical activity enjoyment and subsequently, if possible, the physical activity behavior.Therefore, one type of consistent correlates could be chosen to design and execute an individual research approach and evaluate the relationship with physical activity enjoyment scientifically.For example, satisfying the basic psychological needs of autonomy, competence, and relatedness through supportive language; making decisional balance; providing choices; providing group sessions; helping to set realistic goals; and offering tailored activities 120 are promising opportunities.Another one is concentrating on the social and interpersonal environment, that is, the social network of the children and adolescents.Teachers, coaches, therapists, and family members can try to provide a task-based climate in which realistic tasks are goals that can be achieved and are crucial to demonstrate competence, in contrast to the ego-related goal of being better than others.Evaluating their relationships with physical activity enjoyment could be a way to deal with the knowledge summarized in the presented systematic literature review.

| CONCLUSION
This is the first review that systematically summarizes possible correlates of physical activity enjoyment for children and adolescents.A user-friendly overview was created using a clear coding system, which has been successfully proven in previous studies. 21,23 Not a single study considered children younger than 6 years, but two studies observed children and adolescents in a subgroup analysis.Only a small group of studies included weight-related measurements in their syntheses.Seven articles examined body mass index (BMI) scores as independent variables of physical activity enjoyment, and two articles observed their results completely in a sample of adolescents affected by overweight.No others included weight-related measurements.Most of the studies were conducted in organized sports and in physical education settings.All articles were published in the last 30 years and 61% in the years between 2011 and 2021 (see Table T A B L E 1 Descriptive statistics for the included studies (n = 85).
Abbreviations: BMI, body mass index; na, not applicable; PA = physical activity.

F I G U R E 2
Consistent correlates to promote physical activity enjoyment in children and adolescents.Abbreviation: BPN, basic psychological needs.
Creating a pleasant environment through task-involving climate and autonomy support, satisfying the basic psychological needs and support the task orientation and perceived competence in children and adolescents are aspects that are related to the enjoyment of physical activity.However a causal relationship must be verified in future research.We have identified a gap in literature on physical activity enjoyment in children and adolescents affected by overweight or obesity.Therefore, recommendations were made to close the research gap and to enable the development of innovative research approaches to promote lifelong physical activity by shifting the perspective of the children and adolescents "from 'I should because it's good for me' to 'I want to because it's fun'." 121