CHAPTER 1 INTRODUCTION
Childhood obesity and low physical activity both have serious implications for children's health. This report has twin objectives, exploring potentially modifiable determinants of both obesity and low physical activity in young children. Included in the investigation of obesity is an assessment of whether it may be linked to low physical activity. The report also contains a section focusing on children defined as overweight or obese, examining the extent to which mothers were aware of this overweight and were concerned by it.
1.1.1 Childhood overweight and obesity
Childhood obesity is associated with an increase in childhood cardiovascular risk factors. Childhood obesity may persist into adult life, and adults who were obese as children have a higher risk of diseases associated with obesity, including type 2 diabetes, hypertension, cardiovascular disease, osteoarthritis and cancer (Hannon et al. 2005; Burke 2006; Nathan and Moran 2008).
The United Kingdom has one of the highest levels of childhood obesity among developed countries1. In Scotland, the prevalence of obesity among 2-15 year-old children was 14% in 2010, and three in ten children (29.9%) were overweight or obese (Gray and Leyland 2011). Future projections have suggested increasing levels of childhood obesity in the UK, accompanied by considerable social, economic and health costs (Butland et al. 2007; Stamatakis et al. 2010b). In recent years there appears to have been little change in the prevalence of overweight or obesity among Scottish children. While childhood obesity rates may also be stabilising in England, there are indications of widening socio-economic disparities as obesity continues to rise among children from lower socio-economic groups (Stamatakis et al. 2010a).2
Tackling obesity is an important policy objective for the Scottish Government. The Scottish Government has established a National Performance Framework indicator on child healthy weight3 and NHS prevention and treatment services for overweight or obese children across Scotland. 40% of interventions during 2011-2014 are targeted to children living in the two most deprived SIMD quintiles.4 The Scottish Government's commitment to tackling the causes of childhood obesity is underlined by a number of additional policy initiatives:
- Healthy eating, active living action plan 2008-2011 (Scottish Government 2008)
- Good Places, Better Health for Scotland's Children (Evaluation group of Good Places Better Health 2011) including a childhood obesity evidence assessment published December 20115
- Obesity route map (2010) and action plan (2011) (Scottish Government 2010; Scottish Government 2011)6
1.1.2 Children's physical activity
Physical activity in children is important to promote bone strength and normal skeletal development, and greater activity has been linked with reduced adiposity and cardiometabolic risk factors in childhood (Riddoch et al. 2009; Janz et al. 2010; Owen et al. 2010). Children who are more physically active may continue being active into adulthood, when there are well established health benefits (Telford 2007). Establishing a physically active lifestyle in childhood may help counter the trend for adolescents (especially girls) to reduce their levels of physical activity (Van Der Horst et al. 2007).
In 2011 the UK Government published guidelines on recommended levels of physical activity in children and young people aged 5-18 years (Department of Health 2011). These were threefold:
- Children and young people should engage in moderate to vigorous activity for at least 60 minutes and up to several hours every day
- Vigorous activities, including those that strengthen muscles and bones, should be carried out on at least 3 days a week
- Extended periods of sedentary activities should be limited
The Scottish Health Survey in 2010 found that around 35% of children aged 2-15 years did not appear to be meeting the target threshold of at least 60 minutes of physical activity daily. Over the 2008-10 period, around a quarter of 5-7 year olds did not meet this threshold (Marryat 2011).
Incorporating physical activity as part of a healthy lifestyle forms a core part of policy measures designed to tackle childhood obesity. The Scottish Government has incorporated the UK guidelines as part of its National Physical Activity Strategy7. The need for children to enjoy regular physical activity is recognised by the Scottish Curriculum for Excellence8, and the Active Schools programme promotes opportunities for physical activity outwith PE lessons9.
1.2 Research into risk factors for childhood obesity: multiple causes
The rapid increase in childhood obesity in recent decades appears to result from a biological predisposition to put on weight, coupled with an increasingly obesogenic environment due to wider availability of low cost, high energy foods and reduced physical activity (Butland et al. 2007). Multiple environmental influences on weight gain have been implicated, and variants of an "ecological model" of obesity (Davison and Birch 2001) have been used to categorise predictors of childhood obesity from several different "levels of influence": the child's own characteristics, plus those of the family and wider community.
This ecological model has been used in several recent analyses of childhood overweight and obesity based on data from the UK Millennium Cohort Study (MCS) (Brophy et al. 2009; Hawkins et al. 2009c; Hawkins et al. 2009d; Griffiths et al. 2010; Conelly 2011; Jones et al. 2011b). Studies differed according to the age of child studied (3, 5 or 7 years) and the measures selected for study, so it is not unexpected to see some variation in findings. Consistent risk factors identified at ages 3, 5 and 7 were greater child birthweight, child ethnicity, and parental overweight (Brophy et al. 2009; Hawkins et al. 2009b; Conelly 2011). Of particular relevance to this report is the MCS study on 5 year olds (Brophy et al. 2009), which identified some additional risk factors related to maternal characteristics (poor health, smoking and low educational attainment) and to children's eating habits and a sedentary lifestyle.
The use of an ecological model for childhood obesity is complicated by the strong inherited component to childhood weight (Haworth et al. 2008; Silventoinen et al. 2010). Associations between child or family characteristics and obesity may thus reflect genetic as well as environmental influences. A second complication relates to a difficulty in establishing the nature and timing of any environmental influences, especially when using data gathered over a restricted time period. It is important to bear in mind that associations found between the child's environment and obesity do not necessarily show a causal relationship, but might reflect other influences operating at an earlier date. Several studies have pointed to early risk factors for children's obesity that predate the birth of the child. These include mother's pre-pregnancy BMI (Hawkins and Law 2006), mother's smoking during pregnancy (Oken et al. 2008) and intra-uterine effects on appetite, metabolism, and activity levels (Smith et al. 2007; Oken 2009). These complications mean that while ecological "levels of influence" form a useful conceptual model, they leave many unanswered questions about mechanisms for any risk factors identified.
1.3 Parental influences on childhood obesity and constraints on healthy weight management
Parents are likely to have a dominant influence on the lives of young children. They may also act as "filters" for many other family- and community-level influences. This means that although some family and community-level factors may have a direct effect on the child, the influence of many other such factors may depend on how parents respond to pressures such as low income or a poor neighbourhood environment.
This report uses a simplified ecological model, focusing on two main categories of influence on childhood obesity:
- Parental influences
- Family-level and neighbourhood-level constraints on healthy weight management
The report also contains a section focusing on children defined as overweight or obese, investigating the extent to which their mothers are aware of this overweight and are concerned by it.
1.3.1 Parental influences
Several aspects of parenting are relevant when considering possible influences on childhood obesity. Children's health-related practices including nutrition, exercise and sleep are under the control of the parent. While nutrition and exercise have an obvious link to a child's energy balance, sleep has also been linked with children's obesity. A review of several studies found an association between short sleep duration and children's obesity (Chen et al. 2008). This review included a British prospective cohort study, which found that short sleep duration at 30 months was associated with increased obesity risk at 7 years (Reilly et al. 2005). The mechanism for the association is however unclear, although it may relate to appetite regulation.
Parental example is also likely to be important. Here we include modelling healthy behaviours, together with the well established association between parental and child overweight (Wu and Suzuki 2006). There are, however, a number of possible reasons for the association between parental and child overweight: these include a genetic component, epigenetic effects, continuation of the same neighbourhood influences that originally contributed to parental overweight, sharing of attitudes and practices relevant for the child's nutrition and physical activity, and modelling body image.
A third, less explored, parental influence relates to more general patterns of parenting. A recent review (Kitzmann et al. 2008) sought to explain why attempts to improve health-related parenting practices have often failed to reduce children's weight or improve their diet and exercise. It suggested that day-to-day parenting is critical for the sucessful institution of health-related parenting practices. Parents who are permissive or "indulgent" (lacking rules, routines and enforcement of behavioural boundaries) will not provide adequate support for children learning how to self-regulate their behaviour. Families with low support and cohesion or with high levels of conflict may find it more difficult to institute health-related parenting practices such as family meals or joint family outings to parks and playgrounds. These ideas have found support in observational studies (Sleddens et al. 2011).
1.3.2 Family and neighbourhood constraints on healthy weight management
At the level of both the family and the wider community, there may be factors that act as a set of "constraints" or limitations on the abilty of children to follow a healthy lifestyle. Many of these constraints may impact on the parental influences identified in the previous section.
Some family- and community-level factors may affect parental institution of health-related practices for their children. For example, low income or living in a poor neighbourhood could limit access to resources such as shops providing nutritious food, or gardens and parks for physical exercise (Dunton et al. 2009). Parents with longer working hours might have less time and energy to spend on preparation of healthy meals or exercise with their children (Hawkins et al. 2009a). Other constraints might operate through their impact on parental example. A recent UK study found that conditions in deprived neighbourhoods showed few direct associations with their 3-year-old's weight, but were associated with a mother's overweight (Hawkins et al. 2009e). Lastly, some constraints might impact on more general patterns of parenting. For example, a lack of time or poor parental health may create family stresses that impair effective parenting (Topham et al. 2010).
1.3.3 Parental recognition of overweight, and concerns about their child's weight
Parents who do not recognise that their child is overweight, or who express no concern about their child's overweight, are unlikely to be motivated to try to manage the child's behaviour so that the child eats a healthy diet and takes physical exercise. Other work has found that mothers generally find it difficult to recognise that a child is overweight (Parry et al. 2008; Jones et al. 2011a). Mothers' recognition of overweight appears to be more difficult when children are young (Crawford et al. 2006; Eckstein et al. 2006).
The ability of parents to recognise overweight is best studied in relation to children who are defined as overweight, using objective measures such as BMI. More generally, low levels of parental concern about a child's weight might be viewed as one of the "family constraints" on management of healthy child weight. Here, the assumption is that parents who are not concerned about their child's weight may be less likely to ensure their child adopts healthy dietary and physical activity behaviours.
1.4 Research into factors associated with children's physical activity
Like research on risk factors for childhood obesity, research on children's physical activity has typically categorised possible determinants of physical activity into a number of different groups (Sallis et al. 2000a; Van Der Horst et al. 2007; Hinkley et al. 2008). These may include consideration of the child's growth and physical fitness, together with psychological factors such as the child's motivation and attitudes to physical activity, although the group of psychological factors is difficult to study in young children. Three other groups of influence commonly studied are:
- Behavioural attributes and skills. Children's physical activity may depend on the early development of active behaviour patterns and formation of important skills
- Physical environment (for example, access to facilities such as play parks)
- Social and cultural influences. This includes factors such as parental modelling and encouragement of physical activity, attitudes of the child's peers, and school-level policies favouring active behaviour
Reviews of the determinants of physical activity in children (Sallis et al. 2000a; Van Der Horst et al. 2007; Hinkley et al. 2008) have noted many inconsistent findings between studies. In part, these may relate to differences in the populations studied. In addition, different aspects of physical activity (such as children's school-based activity or out of school activity, or organised activity versus informal active play) may have different determinants. Determinants of physical activity found studies of children aged 4-12 years were male gender, self-efficacy (the child feeling that he/she is capable of active behaviour), parental physical activity and parent support (Van Der Horst et al. 2007). In pre-school children, male gender and parental physical activity were also shown to be important, together with spending time outdoors (Hinkley et al. 2008).
Less is known about the determinants of children's sedentary behaviour. It may have different determinants from physical activity, and may not substitute for physical activity in a simple fashion. However, a recent large UK cross-sectional study using objective measurements of 7-year-olds' physical activity suggested that there were some common factors underlying both higher physical activity and reduced sedentary behaviour, including the child's lower weight status and parental modelling behaviour (King et al. 2011).
1.5 Using the Growing Up in Scotland (GUS) study to investigate children's overweight/obesity and activity
The Growing Up in Scotland study (GUS) is well placed to investigate the role of parents on children's overweight or obesity and their physical activity. GUS has measures of both health-related parenting practices (e.g. taking child to the park or playground) and general patterns of parenting (rules, routines, supervision, involvement with the child, warmth and conflict). GUS can also be used to examine effects of parental overweight and modelling of active and sedentary behaviours. There is a wide range of potential family and neighbourhood constraints on healthy child weight management, ranging from parental attitudes and knowledge to social and physical aspects of the child's neighbourhood.
1.6 Research questions
Two main research areas were examined in this study: identification of potentially modifiable factors associated with children's overweight and obesity, and factors associated with low activity levels in children. These areas are followed by a third topic exploring mothers' recognition of children's overweight.
1.6.1 What potentially modifiable factors are associated with children's overweight and obesity?
Potentially modifiable factors were categorised as parental factors or influences, and family or neighbourhood constraints on healthy weight management.
- Parental factors. Three areas were distinguished:
1. Health-related practices related to the child's diet and activity levels. For example, what is the impact of snacking on unhealthy foods, or having a TV in the child's bedroom?
2. Parental example. What is the influence of mother's overweight, and of mother's physical activity and sedentary behaviour?
3. General patterns of parenting. What is the role of a strong mother-child relationship and effective behaviour management in preventing children's overweight or obesity?
- Family constraints on healthy weight management. Can children's overweight and obesity be attributed to a mother's low level of education or low family income? What is the role of a mother's concerns for her child's weight and her views on desirable levels of children's physical activity?
- Neighbourhood constraints on healthy weight management. Can children's overweight or obesity be linked to living in an urbanised rather than a rural environment, or to living in a more deprived area? Can overweight and obesity be linked to access and quality of local facilities such as children's play areas, or to local crime and antisocial behaviour?
1.6.2 What potentially modifiable factors are associated with children's low activity levels?
The areas examined were identical to the previous section for overweight and obesity (removing health-related practices from the set of parental factors).
- Parental factors. Two areas were distinguished:
1. Parental example. What is the influence of mother's overweight, or of mother's physical activity and sedentary behaviour?
2. General patterns of parenting. What is the role of a strong mother-child relationship and effective behaviour management in supporting children to be more active?
- Family constraints on active behaviour. Does low family income or long parental working hours limit children's opportunities for active behaviour? What is the role of a mother's concerns for her child's weight and her views on desirable levels of children's physical activity?
- Neighbourhood constraints on active behaviour. Does living in an urbanised or a more deprived area limit children's scope for active behaviour, compared to more rural or more affluent areas? Do accessible, high quality children's play, exercise and sports facilities promote physical activity? Do safety concerns and high antisocial behaviour reduce opportunities for activity?
1.6.3 Mother's recognition of overweight in their child
This part of the investigation focused on the subset of children who were overweight (including obese). It explored levels of maternal recognition of overweight and concerns for overweight, and what factors were associated with greater or lower recognition. It then looked at whether a mother's ability to recognise overweight in their child at age 4 may have led to the child losing weight by age 6.