to Leeds Beckett Repository

ABSTRACT


INTRODUCTION
Very little is currently known about children in care's participation in physical activity (PA), or the barriers and facilitators to their engagement.Understanding of PA access and the value children in care place on PA is important since they are among the most vulnerable and disadvantaged members of society. 1 Often exposed to violence, poverty and parental substance abuse, 2,3 children in care typically suffer poor physical and mental health, have difficulties with their social and emotional wellbeing, and lack stable relationships resulting in problems with attachment and lack of resilience. 4 Those subjected to severe neglect, violence or abuse tend to find interacting with wider networks and communities outside the care system particularly difficult. 2,3It is thought that taking part in regular, structured PA offers children in care an opportunity to reintegrate into mainstream society and develop new social networks beyond the care system. 5land and the United States of America (USA), the number of children in care is increasing, with the majority cared for in a foster placement, while some are placed in residential children's homes (RCH). 6,7A significant gap exists between the quality of lives for children in care compared to their peers not in care.In fact, many children in care do not achieve the same standards in education as those in the general community, with 68% of children in care in England having a Special Educational Need (SEN). 8These figures pose significant detriment on the future life and career prospects for these young people.However, Public Health England 9 have highlighted that a positive relationship exists between levels of PA engagement and academic attainment.Research suggests that aerobically fit children achieve greater academic success 10 and have better brain function and memory task performance than less fit children. 11PA therefore can be seen to offer potential to begin to reduce the disparity in educational attainment between those in care and their peers.It is also believed that engagement in regularly scheduled, extracurricular PA provides a sense of structure to the lives of children in care that is otherwise missing due to substantial, continual changes in care placements. 12 educational attainment, PA may offer further developmental potential for these disadvantaged young people.For instance, with children in care experiencing a "higher prevalence of both psychosocial adversity and psychiatric disorder than the most socioeconomically disadvantaged children", 13(p323) the link between PA and mental health is especially important.Furthermore, engagement with competitive PA (e.g.p37) Hence, regular engagement in PA could help foster 'normal' development, strengthen assertiveness, build self-esteem and confidence, and help develop skills required to deal with future life experiences. 5,14 provide opportunities to enhance wellbeing and overall quality of life for those children in care who experience poorer educational and health outcomes compared to their peers.In a recent review of literature, it was identified that overweight and obesity prevalence in children in care is greater than in the general population, 15 increasing their risk of developing serious health conditions in later life, such as diabetes and cardiovascular disease. 16Regular PA participation could therefore help reduce the prevalence of overweight or obese young people, and thus reduce future health complications for children in care.
Finally, as alluded to above, the label 'in care', is often associated with a range of negative outcomes (albeit when compared to normative measures).This label may ultimately shape children in care's own expectations and aspirations.There is also a risk that those facing adversity and labelled 'in care' may develop a stigmatised and, ultimately, all-embracing 'master' identity such as 'child in care'. 17However, engagement in PA may allow children in care to generate a positive sense of identity, since participation in PA may enable them to diversify their repertoire of socially valued roles, re-shaping their identity. 17e the asserted benefits of PA for children in care, evidence from the USA suggests that when compared to young people living in two parent, lone parent and step family households, those in foster care and RCHs were less likely to achieve the recommended levels of PA. 18 One reading of this is that children in care appear to engage in less PA than those in some of the more diverse family structures.Thus, there is a need to better understand why children in care might be less likely to achieve recommended levels of PA since increasing PA may mitigate the negative health and educational outcomes discussed earlier.Ultimately, since there is a lack of knowledge around how active children in care are or how they engage with PA, it is first important to determine what the existing literature details in order to be able to focus future research directions and help shape policy.Due to the broad, multidisciplinary nature of PA research in general populations, coupled with a lack of children in care specific research in PA, the primary aim of this paper was therefore to conduct a scoping review of the international literature to gain insight into the barriers and facilitators to PA engagement for children in care.
To date, no study has attempted this.
As a result of the overarching aim of this study and the limited research in PA that engages with children in care, two subsequent sub-aims included: Given these specific aims, a scoping review was an appropriate method of investigation in this instance.Using this approach allowed for an open and general search of all areas that the topic area infringed upon to ensure important topical evidence was not unintentionally excluded. 19

METHODS
Scoping reviews follow many similar methodological steps as systematic reviews, 19,20 though typically they involve the mapping of a range of evidence to convey the breadth and depth of a field. 21However, a key difference is that authors do not typically assess the quality of included studies in scoping reviews and the resulting synthesis is usually more qualitative. 20ing reviews are also driven by a much broader research question.Hence, this review sought to explore: What barriers and facilitators affect PA participation for children in care?
In line with current frameworks for conducting a scoping review, 21 after identifying the research question the next phases of this study included: (1) identifying relevant studies, (2)     study selection, (3) charting the data and, (4) collating, summarizing and reporting the results.

Identifying relevant studies
A comprehensive search was performed using six academic journal databases (Sports Discus, Physical Education Index: ProQuest, Social Care Online, PsycINFO, MEDLINE, ERIC).This gave access to a range of health, sport and education journals.These databases were searched for articles published between 1989 (when the term children in care was first muted in the 1989 Children Act in England) and 2014.The term 'physical activity' is used here to encapsulate sport, curriculum PE, schools' extra-curricular provision and physical activities undertaken away from school.A list of keywords and alternative keywords were created, combined using Boolean operators ('AND', 'OR', 'NOT'), and included in the aforementioned search databases.The search strings included: (1) 'Physical activity' OR sport OR exercise OR 'physical education' OR leisure OR recreation AND (2) 'Children in care' OR 'looked-after children' OR 'foster care' OR 'residential care' An asterisk (*) was used to indicate that the key words were explored in singular, plural and other related forms.In addition, several specialist journals (British Journal of Social Work; Child and Family Social Work; Journal of Social Work Practice) were searched to allow for the most comprehensive examination of existing literature.The search also explored additional 'grey literature' such as reports and PhD theses that were identified through databases to reduce the risk of publication bias, 22 though non met the inclusion criteria to be included here.

Study selection
Following the initial search, titles and abstracts were screened against the following inclusion criteria: (1) published in English language; (2) published, peer-reviewed journals or peerreviewed reports; (3) published between 1989 and 30 th April 2014 (end date of the search); (4)     relate to children and young people living in or leaving care (in foster homes or residential centres); and (5) broadly explore participation in PA.Both authors met at the beginning, midpoint and final stages of the abstract review process to discuss any uncertainties about study selection. 21This ensured all abstracts selected were relevant for full article review.Searches conducted on the databases returned 576 papers.Following the first stage of screening (titles and abstracts only), some papers were excluded because they were not published in English (n=4) and were outside the search date parameters (n=15).However, the majority were excluded because of relevance to this study; for instance, they didn't relate to children and young people living in or leaving care (n=189) and didn't explore participation in PA (n=362).
As such, six individual papers were left.After this, all relevant references from the final six papers were followed up and included where they too met the inclusion criteria.This resulted in a further study being included, resulting in a final total of seven studies.Both reviewers then independently reviewed the full articles for inclusion in the study. 21

Charting the data
The final selection of articles subjected to full review were collated and presented in Table I.
Using a standard coding template to chart the data, the aim, study design, sample characteristics and key findings of each article were recorded (Table I).As with the previous stage, each author independently extracted data from each article and entered them into an Excel database before reviewing and finalising the recorded data. 21

Collating, summarising and reporting the results
The final sources were thematically analysed to construct a scoping review narrative. 21This allowed for an interpretation of the literature against the primary research question and study aims.Given the complex nature of the lives of children in care, a guiding framework for understanding multidimensional PA influences was needed to support the scoping review.One such approach was the use of the social ecological model.As well as exploring intrapersonal (individual) and interpersonal factors, social ecological models consider broader influential factors such as the community, institutions, and policy on positive health behaviours. 23,24Since the intent was to summarize key findings presented across the articles, an adaptation of the social ecological model by McLeroy et al. 25 was employed as a guiding framework to cast light onto the aspects that shape children in care's engagement in PA.

Results
The final seven articles selected for review were published between 1998 and 2013 and conducted in the USA (3), England (2) and Norway (2).These studies are reported below in relation to the different factors that influence PA engagement for children in care.As identified in the social ecological model results are reported under individual, interpersonal, institutional, community and public policy levels of influence.

Individual factors
Individual or intrapersonal factors that may influence PA of children in care relate to individual characteristics and choices. 25This includes PA knowledge, skills, self-efficacy and individual socioeconomic circumstance.In a Norwegian study exploring adolescents' use of free time, it was suggested that children living in residential homes, reported lower scores in perception of PA ability/success, which may impact on their uptake of PA due to lack of belief in their own abilities. 26This study used a self-report questionnaire that captured salient aspects of the immediate contexts to compare the experiences of 20 adolescents (mean age 16.7) living in youth protection institutions with their non-institutionalized peers (mean age 16.2). 26It was subsequently found that they spent the majority of their free time in passive-receptive activities (e.g.watching television, daydreaming or doing nothing) that do not provide the rewards of improvement or confidence associated with self-involving activities such as physical activities. 26This is perhaps not surprising since vulnerable groups of children often suffer from feelings of helplessness, low self-esteem and perception of inability. 27n addition, finance was reported to impact on children in care's ability to engage in PA.Using semi-structured interviews, an English study of 32 care leavers aged between 18 and 32 found that a lack of money was a major barrier to involvement in leisure and social activities, with the most frequent activity being sport. 28This study provides an insight into potential reasons for drop out of structured PA post-care.While finance could be an issue for those leaving care, an American study of children living in residential care reported few differences in PA participation with regard finance, regardless of race/ethnicity. 29It is important to note however that Dowda and colleagues 29 measured PA using a subjective self-recall questionnaire.This has the potential for socially desirable responses and since young people are less time conscious than adults, tending to engage in activity at sporadic times, 30 accurately recalling activity is problematic.

Interpersonal factors
Very little is known about how various interpersonal processes and primary social groups (peers, social workers, family and foster carers) that provide identity markers, support and role modelling, influence activity choices or participation for children in care.As Safvenbom and Samdahl's 31 follow up study reported, the presence of friends had a positive influence on the desire of children in care to engage in a particular activity (e.g.sport) more often.However, their earlier study indicates participants wanted to engage in more self-involving activities in a particular context, such as through informal play in parks or recreation areas, with social workers. 26They suggest that these types of activities rarely happened because the social worker lacked competence in these activities or did not value them. 26This in turn may be linked to the social worker/support team's level of self-efficacy, skills, or knowledge of activities children in care want to engage in, which filters down to the children themselves.
It is difficult for social workers and foster carers to provide appropriate support or role modelling for children in care when the type of activities they enjoy are not known.In a USA survey of 103 foster-carers, 43% of respondents were unaware of the activities (including, but not exclusively, PA) in which the children in their care were involved in prior to their current placement. 32If policy were to address this issue, it is possible that children in care's PA engagement could increase, alleviating the issues of low self-efficacy, instability and improve resilience.Improved peer and adult relationships with those involved with, and supporting the child in the activity could lead to positive role models as well as nurture a positive support network.

Institutional factors
Only two studies, both from the USA, reported on institutional factors.These include the rules, regulations, practices, policies and structures of institutions (care homes, schools, neighbourhood etc.) that constrain or promote certain behaviours such as PA. 25 Such factors also include the physical and social environment of the foster home/care home.
One of the studies reported on activity of 263 children living in RCHs revealing that specific practices occurring within the homes were preventing access to PA. 29 For instance, the organisation and timing of other activities (scheduled counselling, meetings, therapy or doctor's appointments) occurred during times when children would normally be free to engage in PA. 29 As such, this may prove to be a key factor that hinders engagement in PA yet may be unique to children living in RCH as opposed to those in foster care.Dowda and colleagues 29 also suggest that the proximity of the children's home to facilities, and access to equipment for PA engagement, plays an influential role in the activity levels of children in care.This point is reiterated in a follow up study that explored determinants to activity for children living in residential homes.The study proposes that the rural or urban location of children's homes could be a confounding variable for access to relevant facilities for PA participation. 33g on 19 residential homes with 196 children in North and South Carolina, the second study from the USA found that the presence of a recreational director was a significant predictor of PA in children in care. 33Homes with a recreation director (a paid member of staff responsible for ensuring safe PA opportunities) reported higher amounts of activities provided by a skilled adult, and greater access to off-site opportunities and places to be active. 33ver, the authors suggested that employing a recreational director may be perceived as a substitute for RCH resources. 33They also argued that having a trained, skilled person directing PA could prove to be more important for children living in a residential home than enhancing only the built environment. 33

Community factors
McLeroy et al. 25 defined community factors as the relationships between institutions that influence health behaviour coupled with the various social networks, norms and standards of a specified area/locality.For children in care this may include the relationship between their carer (the care home) and a teacher (the school).In England for instance, all children in care must have a care plan which is drawn up and reviewed by the local authority that looks after them.This identifies intended outcomes and objectives for the child and includes a health plan and Personal Education Plan (PEP).Due to the complexity of their lives, the PEP is intended to be a living document that moves with children throughout their education.It provides a record of their academic achievement and participation in the wider activities of the school and other out-of-school learning activities (e.g.sporting activities).
PA including sport can be raised in PEPs.However, a recent study across 128 local authority councils in England reported that limited guidance was provided to social workers to consider PA within PEPs. 34The inclusion of PA within PEPs in England would arguably help coordinate professionals working with children in care (e.g.social workers, designated teachers and educational liaison officers) toward ensuring they access health promotion schemes and outof-school activities to enhance their well-being. 34At present, this does not appear to be the case.

Public Policy
The final level of the social ecological model centres on public policy: national or local policies that help support or hinder health behaviours such as PA, 25 yet only two studies, both from England, explored this level of influence. 28,34In England, arguably one of the most significant factors impacting on children in care's engagement in PA is the constraints of the care system, particularly placement moves. 28,34For instance, children in care have reported that maintaining leisure activities, including PA, was particularly problematic due to frequent placement moves whereby they ended up living long distances away from friends and places they usually engaged in activity. 28It is therefore important that policy addresses this and considers these factors when placing children in different homes.
A recent study exploring national and local policy influences on sport for children in care found that, despite an expectation that local authorities in England would offer free sport and leisure provision for children in care, 48% of councils did not provide this. 34Drawing from Freedom of Information requests to 152 Directors of Children's Services, this study found that only 41% of councils provided free access though this was usually restricted to certain activities (i.e.swimming) and often at restricted times 34 making regular engagement in PA difficult.In addition, to compensate for those children who lived further away from the subsidised (often) city-based leisure centres, only a few councils provided free bus passes, noting that transport for these children was a major barrier to participation. 34Murray's 34 study reported that councils tended to rely on the provision of PA for children in care within a school context.This is partly a facet of the funding structures in place in England but also relies on an assumption that schooling is universally accessed, which for many, it is not.

Discussion
To our knowledge this is the first review to explore the PA influences of children in care and reveals they may struggle to access PA equal to their non-looked after peers, though clearly further research is required here.Despite using a social ecological model to review the literature, it is important to consider that only seven papers were found that contributed to this knowledge base.Although several of the barriers to PA experienced by children in care may be similar to those experienced by the most vulnerable families in society (including finance and transport opportunities), 35 they also face further challenges unique to their social situation.
One of the primary factors reported here is that children in care may have a low perception of their own ability (self-efficacy).Self-efficacy is considered a central influence on healthy behaviours, 36 particularly with regard to children in care's perceived ability to execute specific skills required for some physical activities.Thus at an intrapersonal level, low self-efficacy may inhibit engagement in PA.This could be addressed through introducing, encouraging and modelling more self-involving activities in children's lives. 26However, little is known about Physical activity and children in care whom the key social agents are that could influence children in care's engagement with PA through modelling or providing support and encouragement, though friends, social workers and foster carers may play a particularly important role. 26,31,32In England the influence of friends is particularly problematic since placement moves make it extremely difficult to maintain friendships over prolonged periods of time 28 and yet, ironically, continued access to PA may enable children in care to develop friendships beyond their immediate environment. 5,14,28Moreover, some social workers and carers may lack the necessary competence, 26 and knowledge of previous experiences, 31,33 to be able to successfully support them.It would appear that in England, there is a need to ensure social workers and carers are given guidance and advice about the role of PA for these young people.
The literature would also suggest that specific care home policies may restrict access for children in care.For instance in the USA, the scheduling of appointments has been found to occupy time that could be used for PA, 29 while the location and built environment of residential homes may not be conducive to PA. 29,33 These latter findings are consistent with previous research that highlights the proximity to places to be active and availability of sports equipment as factors that influence PA participation. 37Despite the different geographical location of the studies, these factors may well translate to a UK context where RCH may operate in similar ways, are located in areas of high deprivation and have limited space for activity.A particularly important finding, however, is that some homes in the USA appoint a recreation director to organise activities for children in care. 33Thus, the most important policy for the unique setting of a RCH may be to appoint (where possible) a recreation director to help facilitate PA.
More broadly, constraints of the care system ultimately impact on engagement with PA.In England, it has been reported that inconsistent provisions of free sport and leisure access and sporadic provision of a free bus service to access activities hinders children in care's engagement in PA.However, placement moves unquestionably have the biggest impact.For instance, one study reported that a girl entered care at 16 and had 17 different care placements by the time she left care aged 21. 28 School attendance for children in care is therefore problematic since many experience frequent changes in care placements.Children in care are also more likely to be permanently excluded. 38As such, they often miss out on school based sport activities (and are more dependent on out-of-school activities than other children), which is particularly problematic if councils in England are relying on schools to provide access to PA.
A particular strength of this study was the use of the social ecological model that allowed the authors to explore the multiple influences on PA at different levels.In keeping with the interrelated nature of the social ecological model, it would appear that local and national policies have a major impact on various factors that influence children in care's engagement in PA.However, before considering these policy implications, it is worth reiterating that these findings are based on only seven studies from three very different countries where child welfare approaches and systems differ.For instance, unlike England and Norway, the USA has a relatively decentralised welfare system with federal law shaping minimum standards, with each state having relative autonomy in how they operate. 39These differences within and across countries may ultimately shape PA factors differently.In addition, this search was limited in terms of what the databases returned and how accessible they were, and therefore could have excluded smaller case study work that would have informed this study.
That said, it was reported that children in care suffer from low self-efficacy which inhibits their ability to participate in physical activities, a finding that may transcend geographical location.
Stable friendship groups appear important facilitators in improving self-efficacy although with frequent placement moves these relationships prove difficult for children in care to maintain.
Continuous engagement with PA provides a channel to foster stable friendship groups and thus improve self-efficacy through role modelling and peer support.Yet the frequent placement moves impact on the continuity of activities and mean it is extremely difficult for children in care to attend regular, structured physical activities to develop such relationships.It is therefore imperative that future research and policy address these issues if children in care are to lead a fulfilling childhood comparable to those not in care.

Methodological considerations
As evidenced by the scarce number of studies (specifically concerning PA) that directly engage with young people (n=4), research with children in care can be quite complex, especially with regard to gaining access and seeking consent. 40This is in part due to the large number of gatekeepers involved (social workers, service managers, birth parents, carers etc.) and part due to the ethical considerations that are integral to working with vulnerable young people.Gaining access and recruitment is therefore a slow and time consuming process and must be built into future study timeframes.Gatekeeper's perceptions of children in care's vulnerability may also mean that children are denied the opportunity to participate without them ever even knowing about the research. 41While their duty to protect the best interest of the child is probably at the heart of gatekeeper's decisions to include or exclude children from research, these decisions raise questions about children's rights and ability to make their own decisions.
Moreover, listening to the voices of children in care is not unproblematic.Perhaps in an effort to mitigate difficulties with access and ethical considerations associated with talking directly to this vulnerable group, that those studies that directly engaged with children in care reported here, all used quantitative, self-report measures.However, this approach has particular limitations with regard to children's memory and their cognitive abilities to accurately recall events from the past.Only one study directly engaged with care leavers and did this through qualitative interviews.It would therefore seem reasonable to assume that with fewer restrictions on gatekeepers and less ethical implications (given that most care leavers are classed as adults), then engaging with this group could be explored further.Despite this, most studies included in this review engaged with key stakeholders instead of the young people themselves.
For future studies in this area it is important to remember that while gatekeepers are often barriers to children and young people in care, they are also the gateway to their participation. 42As such, to help improve access, Murray 42 suggests there is a need to invest considerable time in building relationships with gatekeepers at various hierarchical levels and sustaining contact with them throughout the study.Research from other fields also reveals that asking carers (foster carers or residential staff) to be involved in the study alongside young people may help increase the number that agree to their children's participation, 40 while others have offered remuneration in the form of small gift vouchers to increase participation. 42

Gaps in the literature
While there is evidence to suggest that PA participation can have a positive effect on educational attainment and the physical and psychological wellbeing of children, further research is required specifically with children in care.To maximise any effect of engagement in PA and to help them recover from any earlier disadvantaged, further priority should be given to exploring whether children in care are as active as they could be, whilst attempting to understand the factors that influence their PA engagement.There is little research on the types of PA that children in care engage in and there is also a need to consider any differences in PA that children in care engage in while living in foster care or in RCHs, as the two settings differ.
Information about the availability of equipment or places to go to be active for children living in care homes in England or abroad is likewise required.Despite the different geographical contexts of the studies reported here, it is apparent that most of the factors identified may well translate to England and other countries where children live in care.However, there currently lacks a national and international body of work that engages with children in care to explore the meaning, value and relevance of PA to them.Finally, if future research begins to address these gaps, it will better inform how access to PA can enhance the physical and mental wellbeing, and thus the quality of life, for children in care.Care leavers aged 18-24 (n=32) who were in care at age 16 were interviewed.Some participants (n=27) were interviewed a second time a year later.Interviews with carers and professionals that lookedafter children identified as being particularly supportive were also conducted (n=14).
The majority of those interviewed were engaged in a wide range of social and leisure activities with the most cited being sport.41% engaged in some kind of sporting activity.Participating in sport was important in developing friendships and widening the social network with those not in care.Sport had provided an important source of stability and consistency in their lives.Quantitative study of physical activities and sedentary pursuits that were reported over a 3-day period using the 3-Day Physical Activity Recall (3DPAR).
Participants included 263 children (52% male, 40% 11 to 14 years old, 60% 15-18 years old, 53% White, 23% African American, and 24% other race/ethnic groups) from 23 residential children's homes in North and South Carolina.The median length of stay in the homes was 6 months.
Children in residential homes participate in activities that are similar to children living with their parents.However, children in residential homes may participate in some physical activities for shorter periods of time than children living with their parents.Practices within the homes (e.g.scheduled appointments) were preventing access to physical activities.

1 .
Exploring how research literature have used different methodological approaches for engaging with children in care in relation to PA and; 2. Identifying future research directions and the policy implications based on the review outcomes.

Table I .
Background, methodological details and key findings of included studiesYouth in care reported significantly fewer activities categorized as self-involving (e.g.physical activities) and more activities categorized as passive receptive.Looked-after children spent most of their free time in solitude and seemed to withdraw from public settings.Findings suggest the introduction of activities that are self-involving as a way to enhance self-esteem.