Parent recommendations to support physical activity for families with young children: Results of interviews in deprived and affluent communities in South Wales (United Kingdom)

Abstract Background Physical inactivity is the fourth leading cause of mortality worldwide. Early childhood is a critical period when healthy behaviours can be instilled for a future active lifestyle. We explored community, societal and environmental factors affecting child and family physical activity and sought parent recommendations to support physical activity in families with young children. Methods We interviewed 61 parents expecting a child or with a baby ≤12 months (35 mother and father paired interviews and 26 interviews with mothers only). We purposively sampled families for neighbourhood deprivation status (Townsend Index; 26 affluent; 35 deprived). We conducted thematic analysis of interview transcripts using Bronfenbrenner's socio‐ecological framework to guide interpretation. Results We identified four themes: work family‐life balance; spaces for activity; beliefs and attitudes; and physical activity facilitators. We found that parents from deprived neighbourhoods were more likely to be underactive because of a complex web of community, social and personal factors which reduced motivation and hindered opportunity for physical activity. To increase knowledge and opportunity, respondents suggested ‘help not tell’ messages covering ‘why’, ‘how’ and ‘where’ information about physical activity, and using physical activity to support community engagement and social interaction. Conclusions Recommendations from parents highlight effective communication about the importance of early child and family physical activity and improved community access to safe facilities and opportunities. Both parents need to be engaged in designing interventions to support greater physical activity and healthy behaviours which are relevant and achievable in individuals’ lives.


| BACKG ROU N D
Physical inactivity is the fourth leading cause of mortality worldwide. 1 Greater urbanization and development of modern technology contribute to inactivity and its associated physical and psychological health problems. 2,3 Inactivity and sedentary behaviour start from an early age and are linked to weight gain through childhood. 4,5 Studies have identified early metabolic markers for high cholesterol, blood pressure and abnormal glucose metabolism in overweight children as young as five 6 and early vascular lesions in overweight children as young as three. 7 Many pre-schoolers (2-6 years) fail to complete the minimum daily 60 minutes of moderate-to-vigorous physical activity recommended for young children. 8 Targeting interventions at early years (0-8 years) could support formation and maintenance of future health behaviours 1 when physical activity can be nurtured to sustain active lifestyles. 9,10 The Department of Health in the United Kingdom has produced physical activity guidelines for children from birth to 5 years of age to encourage activity in early childhood. 11 Increasing whole-family physical activity could potentially reduce childhood obesity and improve their overall health. Poor motor development in infancy and early childhood is associated with reduced physical activity in older children but can be improved through parent-facilitated physical activity at an early age. 12,13 Children's activity rates increase if one parent is active and are highest if both parents are active. 14 Children with active parents are more likely to maintain activity levels through childhood. [15][16][17][18] However, adults with dependent young children appear to be less physically active compared to those without children. 19,20 Moreover, parents 21 and adults in general 22,23 undervalue the importance of physical activity as a means to encouraging energy expenditure and reduce weight gain. 24,25 Interventions which focus on individuals and their families can overlook social, economic and environmental barriers to undertaking physical activity; which individuals often have little control over and can prevent them maintaining positive health behaviours. [26][27][28][29][30][31] Young mothers from disadvantaged backgrounds are less likely to participate in regular physical activity. 32 Children from families in lower socio-economic groups have lower levels of physical activity 33,34 and higher body mass index. 35 Studies looking at parent-reported barriers report lack of time, 36,37 the cost of being active, 36,37 facilities within the home 36,38 and outside. 39 One way to understand how context influences parent and child health behaviours is through Bronfenbrenner's socio-ecological model. 31,40,41 This identifies four levels of influence on a child's development: the individual's immediate physical and social environment such as home and family (microsystem); the wider environment such as school (mesosystems); broader social, political and economic conditions (exosystem). These all interact to affect the beliefs and attitudes of wider society (macrosystems; see Figure 1).
This study explores the attitudes of parents across different socio-economic groups towards physical activity and the opportunities available to them. We use Bronfenbrenner's socio-ecological framework to address the reported evidence gap in ways to enable more physically active children. 42,43 In this paper, we report the views of expectant parents and families with a baby ≤12 months concerning community, societal and environmental factors affecting family physical inactivity, as this can be a time when family activity levels can decline. We also present parents' recommendations for interventions to support physical activity in families with young children.

| Setting
Parents in this qualitative study were already taking part in an existing birth cohort study 'Growing Up in Wales: Environments for Healthy Living' 44 which examined the impact of environmental influences during gestation and post-birth on health outcomes using data linkage of routine, anonymized medical records. 45 Participants were recruited for the cohort study when they attended maternity appointments in hospitals and clinics. Exclusion criteria were pregnancies under the age of 16, incomplete pregnancies and mothers with serious health problems such as cancer. A detailed description of study participants who were eligible to take part in the birth cohort study has been provided elsewhere. 44

| Qualitative study: recruitment and data collection
We purposively sampled parents from the birth cohort sample of parents according to neighbourhood deprivation using the Townsend Deprivation Index. 46 AK contacted expectant mothers or mothers F I G U R E 1 Bronfenbrenner's socio-ecological model. Source: Bronfenbrenner 40 with infants aged ≤12 months, face to face at antenatal clinics or by telephone. AK made arrangements to interview those who agreed to take part. Interviews took place in participants' homes except for two conducted at participants' workplace (personal preference). A semi-structured interview schedule was used for interviews (Table 1).
Questions concerned parents' knowledge and views about living a healthy lifestyle; how this influenced their current choices for familybased physical activity; barriers and facilitators to physical activity; and recommendations to facilitate family physical activity. Informed consent was obtained before undertaking and recording the interviews, which were all carried out by AK. Interviews were conducted in English, Bengali and Urdu depending on which language was preferred by respondents. Parents who spoke Urdu and Bengali were included in the sample as the researcher AK was fluent in both these languages.
The interviews ranged from 30 to 60 minutes.
Recruitment and data collection happened in parallel. Initially, the intention was to interview mothers only. However, fathers were present in two early interviews and it was recognized that this broadened the discourse and enabled data capture of the whole-family perspective, and useful detail of the dynamics between parents.
Subsequently, AK sought to recruit fathers also. During the paired interviews, AK encouraged both participants to equally contribute to the discussion by rephrasing and directing questions to both participants.

| Data analysis
An inductive thematic data analysis approach was used 47 where theoretical perspectives are informed by the interpretation of raw data. AK, RL and EM independently reviewed anonymized interview transcripts, working with the first 20 completed interviews to systematically code the data to draw out themes and categories to illustrate emerging concepts. Through discussion, these themes and categories were further refined and clustered into codes and sub-codes concerning individual/family, community, environmental and societal level, using the socio-ecological model as a framework 40 (see Figure 1). These formed a codebook which was used by AK to code the remaining transcripts. Findings were discussed in four paired analysis sessions, to ensure robustness and internal validity. 47,48 A senior qualitative researcher (FR) was also available throughout to challenge and critique analytic outputs and themes as they emerged.

| Reporting
We report results according to themes identified in the data. We selected quotations to be representative of respondents' comments unless otherwise stated. Quotations are identified by respondent family identification number and whether they are a mother or father. Further information such as deprivation status of where they live and age and number of children can be found in Table 2. Respondent characteristics are shown in Table 3. Four interviews were conducted with mothers only in the participant's mother language (Bengali [3] and Urdu [1]) and were translated and transcribed by the researcher (AK). The remaining transcripts were transcribed by an independent transcriber. All transcripts were cross-checked with notes taken during the interviews to ensure data integrity. We identified four themes: work family-life balance; spaces for activity; beliefs and attitudes; and physical activity facilitators.

| Work family-life balance
Respondents reported that busy work schedules, irregular work patterns and lack of time reduced opportunities for family physical activity. Many parents expressed a keen desire to be active as a family unit and often sought support from each other to facilitate family activities, recognizing value for whole-family health and well-being and also the opportunity to maintain and strengthen family bonds. However, work patterns resulted in extended periods of time when one parent was alone with one or more young child. Often, one parent had to stay home to wait for their partner to return. This was especially evident for mothers living in deprived areas. Repeated and prolonged periods alone caring for their children also appeared to be a bigger burden for mothers.

TA B L E 1 Interview schedule for physical activity
Can you describe your weekly levels of physical activity (individually and as a family)?
Does anything pose a barrier to your family taking part in physical activity?
Local environment and location of facilities?
During your childhood how physically active were you?
Can you identify the best way to inform or advise parents about the importance of physical activity?
What sort of service or support do you look to health professionals for?
What sort of families do you think will/should use or listen to advice on taking part in physical activity?
Do you know of anything being done in your local area to promote physical activity?
Do you have any ideas on ways to improve your local area to encourage family physical activity? The effect of employment patterns was more acute when one or both parents were working shifts or overtime for additional income, which was more common for those in deprived neighbourhoods.
Having regular work hours enabled families to schedule activity during the week and at weekends.

| Safety and accessibility
Many respondents talked about their safety in outdoor spaces, in addition to the physical risks in vandalized or dirty public parks.
Respondents who lived in deprived areas were unhappy with the condition of their local parks. Access to safe walking routes was also a concern for parents from both types of community.

| Beliefs and attitudes
Parents appeared to understand the importance of physical activity, for their children and also for their family. Although some mentioned it in a health context, activity was more often discussed as a time for being with other people. Some parents said they enjoyed cycling as a Other respondents, whose routine provided less opportunity for being active, perceived activity as an additional thing to arrange, often involving inconvenience or difficulty and they reduced their expectations accordingly.

No I don't mind, I could walk but… you know, I don't think my daughter [three years old] can walk for 20 minutes, you know, it's too much for her (1: Mother)
Parents' attitudes towards physical activity also appeared to be strongly influenced by the quality of the environment. Some respondents associated beaches and countryside with an active lifestyle and consciously chose to live in these areas so they could follow an ac-

| Physical activity facilitators
Parents asked for information to encourage physical activity, with messages applicable to a family which explained why physical activity was valuable, how to do it and where to access local opportunities.
Supportive and relevant messages, 'helping' not 'telling', were more likely to encourage people in place of a didactic approach which risked alienating those who found the advice hard to follow through. community-based activities as a way to achieve this. Some ideas concerned child-focused neighbourhood events which did not appear to involve parents in activity. Other respondents suggested ways for communities to unite in physical activity by bringing parents together.
Ideas included neighbourhood charity fundraising physical activities; community days out, bringing families together to be active; walk-toschool schemes to incorporate physical activity into daily routines and also reduce traffic volume and pollution; activities attached to community parenting classes and childcare provision; and community gardening areas to provide family physical activity and encourage healthy eating (see also

| D ISCUSS I ON
This study provides insights into factors influencing child and familybased physical activity and includes recommendations from parents on ways to improve opportunities for family-based physical activity in the community. Their suggestions have been summarized in Table 4. We found that parents from deprived neighbourhoods are more likely to be underactive because of a complex web of community, economic, social and personal factors which block motivation and opportunity for physical activity. Respondents said both parents need to be involved in designing interventions to support greater physical activity and healthy behaviours if they are to be relevant and achievable in individuals' lives. Few physical activity-based interventions have targeted both mothers and fathers of young children, 15,49 or have focused on facilitators and barriers to parent and child physical activity in the community. 50 Engaging users to identify, design and deliver interventions is recommended to remedy current unhealthy lifestyles and lack of engagement with health messages. 51,52 We identified barriers to family-based physical activity which align with the levels described in Bronfenbrenner's socio-ecological model. 31,40 We also found that some issues cross the ecological levels: work patterns and income (exosystem) affect individual isolation and well-being (microsystem); and the quality of built environments and neighbourhood facilities (mesosystem) was associated with socio-economic characteristics (macrosystem). 41 Targeted mailings about local activities issued before school holidays Location-specific advertising of activities, focusing on community relevance Awareness raising by health professionals through pre-to post-natal networks Link with formal and informal parent and toddler networks Extending existing services Physical activity sessions at community parenting classes Route cards for local cycle paths and walks Childcare at adult activity sessions Improve lighting and cleanliness of cycle routes and walking paths More activity clubs in deprived areas Make use of school and community facilities after school hours Community activities Charity fundraising activities: cycling/walking/jogging Active family days out Walk-to-school schemes Community garden areas Youth clubs and activities for teenagers working mostly regular hours during the week and were therefore able to maintain some form of regular physical activity outside the home.
Obesity and other risks to adult health associated with variable work patterns, as well as the disruptive effect on the mesosystems such as family routines, have been documented. [57][58][59][60] Our findings reinforce evidence that shift work can limit involvement in regular extra-curricular activities and general socialization in the community for families with young children in particular when both parents were working opposite shifts. We found one parent cannot always facilitate physical activity without partner or family support, which reduced child activity. 56,61 We now need to understand how the wider socio-economic context of parents' variable employment patterns impacts on child health and physical activity at the microsystem level and devise targeted interventions at one or more socio-ecological levels to facilitate improved interactions between microsystem, mesosystem and exosystem factors.
Parents were not always aware of physical activities for families in their local area (ie mesosystem). Interventions targeted at the exosystem level could direct health professionals to provide information on available local opportunities for physical activity and the health benefits of being active for parents with infants and toddlers. Indeed, parents may be more inclined to enable family physical activity if they were aware of the importance for early motor development and future child health at the individual microsystem level. 62 However, Bronfenbrenner's ecological theory 42,43 would suggest that changes in the mesosystem or exosystem are unlikely to be successful in isolation. Nutbeam 63 states that providing information alone is unlikely to change behaviour. Acceptance of public health messages is determined by the way in which individuals interpret and internalize such messages. This can be mediated by microsystem factors such as individual autonomy and the ability to utilize this message, governed by contextual factors such as perception of lifestyle, level of education, disposable income, the physical environment and available support at the mesosystem and exosystem level. 28,64 Health-orientated messages delivered at different ecological levels need to be simple, clear and consistent to be effective 65 ; explain the importance of physical activity, convey the type of physical activity people should do, and ways they can be physically active. 66 Awareness raising messages can also fail to reach the target audience due to timing of interventions, and lack of publicity and poor practitioner support. 67 Parents in our study suggested 'help not tell' messages covering 'why', 'how' and 'where' information about physical activity. They suggested tailoring information for fathers and minority ethnic groups and using community and social networks for dissemination.
Participants in our study who were physically active outside the home often mentioned enjoyment and general well-being for the whole family as a motivation. People who perceive a purpose in physical activity are more likely to take part and enjoy it at the individual microsystem level. 68 Respondents who advocated community activities (charity bike rides, walking to school schemes) at the mesosystem level wanted to know their neighbours. This was particularly evident among respondents in deprived neighbourhoods who saw physical activity as a means of community engagement and social interaction to improve community cohesion. Thus, they advocated opportunities to engage in purposeful physical activity that accrued benefits above and beyond the individual. Research shows that community engagement can encourage physical activity and increase perceived social cohesion. [69][70][71] Social interaction is also a motivator for family physical activity 68,72,73 and may increase participation in community physical activity. 72 Our findings identified that physical activity may also foster social interaction and this message may provide the purposeful motivator for individuals who feel disengaged and isolated.
Our study highlighted problems at all levels of the socio-ecological model (ie microsystem, mesosystem and exosystem) preventing parents from deprived neighbourhoods from undertaking physical activity with their children. In common with people generally, they preferred to visit quality green spaces, the sea or leisure facilities, despite needing to travel a distance. [74][75][76] Compared to those in affluent neighbourhoods, facilities local to them were inadequate, pushing them further afield in search of safe, accessible facilities.
However, they physically struggled to leave their homes when the other parent was working, often lacked transport and were unable to meet the costs. Motivation and choice were consequently diminished. Frequency of outdoor activity reduces with distance required to travel to suitable locations, while access to personal transport usually increases physical activity in a choice of locations that offer varied experience. 77 People with access to green spaces often exhibit better health-related outcomes. 78 Subsidized access to travel and venues that facilitate physical activity and more places for organized activities could help reduce these challenges at the mesosystem level. Improvements to neighbourhood facilities such as parks, paths and cycle ways could also address the reported negative perceptions. [79][80][81] Physical activity generally increases in areas with more sports and recreational facilities, and attractive parks and cafes. [82][83][84] Free outdoor recreation that is safe and attractive makes it visible within a community, positively influencing social and cultural attitudes towards physical activity. 83 Such a change at the exosystemic level could bring about macro-level changes which normalize physical activity and improve community cohesion, which was a priority for parents in this study. Public spaces near to bus stations and shopping areas could encourage both parent and child activity. Well-designed neighbourhoods also enable more physical activity. 75 Initiatives at the exosystem and macrosystem level 40 require political commitment and resources to make societal changes impacting on health inequality. 85,86 In the short term, interventions that can promote physical activity in communities include improvements in active transport, housing location, urban design and neighbourhood safety. 26,41,87,88 Most successful multi-component interventions can be seen to echo Bronfenbrenner's socio-ecological theory where they are delivered alongside and supported by social marketing programmes that have raised awareness about the positive benefits of making healthy behaviour change. 89

| Strengths and limitations
The size of this study sample, which also included mothers and fathers, is a strength of this study. Respondents came from diverse socio-economic and cultural groups and provided a rich and widespread of views. Where both parents were present, we were able to obtain varying perspectives on the topic. Parents either confirmed or contradicted a point or added detail to each other's responses. However, we acknowledge that the presence of the partner could have negatively influenced some responses where respondents felt unable to speak freely. There was a possibility of participant bias as mothers had already agreed to participate in a birth cohort study so may have been more motivated towards healthy behaviours compared to the general population.
However, views appeared generally consistent within socio-economic groups, suggesting the responses concerned issues shared by those populations.

| CON CLUS ION
Parents from deprived neighbourhoods experience a range of community, social, economic and personal barriers which interact to limit their ability to be physically active with their children. To increase knowledge and opportunity relating to physical activity, they suggested 'help not tell' messages covering 'why', 'how' and 'where' information, tailored for audiences including fathers and minority ethnic groups. They saw physical activity as a means of community engagement and social interaction and advocated local activities for children and families. Engaging parents and communities in identifying, designing and delivering interventions may help support increased physical activity which are relevant and achievable in individuals' lives.

ACK N OWLED G EM ENT
We would like to thank the parents for taking part in the 'Growing

CO N FLI C T O F I NTE R E S T
The authors declare they have no actual or potential competing financial interests.

AUTH O R S' CO NTR I B UTI O N S
RAL and SB designed the 'Growing Up in Wales' birth cohort study.
AK, RAH and KM were involved in the recruitment of participants to this birth cohort study. AK conceived the qualitative research concept and study design and sampled participants from the birth cohort study for interview. AK developed the interview schedule with advice from SB and RAH. AK conducted all the interviews.
AK, RL and EM analysed and interpreted the qualitative interview data and drew up a draft document of themes. FR offered guidance on qualitative methodology. AK drafted the manuscript with support from BAE, and all authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T S
The data that support the findings of this study are available from the corresponding author upon reasonable request.