Preventing excess weight gain

The Committee's main aim for the update of section 1.1.1 of NICE's guideline on obesity prevention was to provide more up‑to‑date, nuanced information on factors that might help people maintain a healthy weight or prevent excess weight gain, as requested by stakeholders during the review of the guideline in 2011.

The majority of evidence in the reviews considered by the Committee tended to define a healthy weight as a BMI of 18.5 kg/m² to 24.9 kg/m², in line with existing NICE guidelines (see NICE's guideline on obesity: identification, assessment and management). However, the Committee recognised that, given that mean BMI is around 27 kg/m² for adults, the aim for many people may be to prevent further weight gain. The Committee also recognised that what is considered a 'healthy weight' may differ by ethnic group and age because of differences in body composition and fat distribution. See NICE's guideline on BMI: preventing ill health and premature death in black, Asian and other minority ethnic groups and Prentice and Jebb's 2001 paper on beyond body mass. Despite this, the Committee was able to make recommendations that apply to all population groups.

The Committee did not consider maintenance of weight among people who had previously lost weight. The Committee recognised that although the recommendations may help this group prevent regaining weight, additional action may be needed. People who are finding it difficult to maintain a healthy weight through the types of habits outlined in this guideline may benefit from effective lifestyle weight management programmes (see NICE's guideline on lifestyle weight management services for adults).

The Committee recognised that weight is gained when energy intake ('calories in') exceeds energy expenditure ('calories out'). This concept of energy balance provided the basis for developing recommendations in areas for which the evidence was lacking.

Activities that are known to increase energy expenditure or decrease energy intake were assumed to contribute to maintaining a healthy weight. The energy balance concept also implies that addressing a range of both dietary habits and physical activity behaviours is more likely to be effective than changing a single factor in isolation. In addition, evidence considered when developing related NICE guidelines – for example NICE's guidelines on preventing obesity, lifestyle weight management services for adults and lifestyle weight management services for children and young people – suggests that a multicomponent approach, addressing both dietary habits and physical activity, is likely to be effective.

The evidence base on energy balance was often unclear, because many studies adjusted for energy intake, obscuring any association and tending to bias towards no association. The Committee was of the view that adjusting for energy intake when associations between dietary habits and weight outcomes are being considered is unhelpful.

In relation to energy intake, the Committee considered that energy density was an important concept. However, the Committee recognised that there was not a consistent definition of high, medium or low energy density. Although limited evidence is available on energy density and weight outcomes, the concept of reducing energy density was particularly helpful in identifying practical dietary change that may help support a healthy weight. Similarly, limited review‑level evidence was available on portion size, but the Committee considered that practical information on appropriate portion sizes of foods and drinks may be important to reduce the risk of over consumption.

In relation to energy expenditure, the Committee recognised that any physical activity would contribute to energy expenditure. The total volume of activity is the critical issue in relation to energy balance (rather than any specific combination of intensity and duration of activity). The Committee was mindful that existing population recommendations include all forms of physical activity, from short bursts of incidental activities such as walking to periods of moderate‑to‑vigorous activity.

The Committee supported a pragmatic approach to reviewing the evidence for this work, given the span of the topic and the need to develop best practice recommendations. Members recognised that focusing only on high‑quality systematic reviews may mean that evidence for specific factors, particularly newer areas of research, may have been overlooked. However, the Committee agreed that the focused approach made the task manageable in the time available.

The recommendations in this guideline are based on the best available evidence considered by the Committee. However, members recognised that practical examples may help people using the guideline better understand the sorts of changes that might be necessary. The practical examples given are based either on available evidence or are extrapolated from the available evidence. For example, evidence considered by the Committee identified that consumption of sugar‑sweetened drinks is associated with excess weight gain. The practical examples of alternatives to sugar sweetened drinks were identified by the Committee members from extrapolation of the evidence and from their own knowledge and expertise.

The Committee was mindful that lifestyle choices are not static. For example, in recent years, 'coffee culture', the increased consumption of high protein 'sports' drinks and use of handheld screens may have undermined people's efforts to maintain a healthy weight. Therefore, it will be important for practitioners to tailor recommendations in this guideline to address changes in lifestyle choices.

The majority of studies included in the systematic reviews were observational cohort studies. Although this type of evidence is appropriate for identifying associations between individually modifiable factors and weight outcomes, it cannot determine causality. In particular, the Committee was not able to determine why there is such a strong association between inadequate sleep and increased risk of weight gain in children. The Committee recognised that sleep may merely be a marker for other aspects of children's lives, such as increased time spent watching TV. However, members of the Committee were also aware of some evidence that sleep may be influenced by children's physical activity level or that sleep may influence appetite. Whatever the reason, the Committee was of the view that a recommendation on adequate sleep was justified, given the strength of the evidence considered and the wider health benefits.

The Committee assumed that the observed associations between meals eaten out of the home (particularly 'fast' and 'takeaway' foods and drinks) and weight outcomes was because of the high energy density of these foods and drinks. The Committee also assumed that observed associations between TV viewing and weight outcomes may be caused by several factors. These include increased time being sedentary and a reduction in physical activity, snacking while watching TV, being prompted to eat by TV programmes or adverts, and eating more while being distracted by TV.

The Committee was aware that some factors may cluster. For example, people who are healthier may follow a range of dietary or physical activity behaviours that will help them maintain their weight. This makes it difficult to identify associations with individual factors. The nature of the evidence prevented detailed consideration of the context within which people may be more or less likely to adopt a particular behaviour or identify any interactions between behaviours (for example, appetite and physical activity).

The Committee was aware that eating breakfast has been part of population advice on healthy eating for many years. Observational evidence supports this advice, suggesting that people who eat breakfast tend to have lower weight or less weight gain. However, evidence from interventions to encourage people to eat breakfast is not always consistent. The Committee recognised that advice should be clear that breakfast may support a healthy weight if it is eaten as an alternative to energy dense snacks and overall daily energy intake is not increased.

The randomised controlled trials included in the systematic reviews tended to be short term. The reviews provided limited information on the strength of associations or wider applicability. Very little evidence was identified on particular population groups, such as those from different social or ethnic groups, and it was unclear whether any inequalities had been investigated or identified. Therefore, although the recommendations in this guideline apply to all population groups, the Committee considered this an important gap in the evidence and has made recommendations for further research. The Committee also recognised that recommendations on tailoring advice and support for specific population groups in related NICE guidance will be important in this light. The Committee was mindful that some of the practical examples, such as using food and drink labels to identify products that are lower in fat, using apps to monitor physical activity level or reducing consumption of foods prepared outside of the home may be harder to implement for some groups than others. However, the Committee recognised that related NICE guidelines (such as NICE's guideline on obesity: working with local communities) stress the importance of identifying and managing the needs of different population groups to address any inequalities in health.

In instances where the evidence was inconclusive, the 'direction of travel' of the findings was often consistent across studies. The Committee members were also able to draw on their wider knowledge of the topic to develop recommendations. For example, the evidence on active travel and weight outcomes was limited. However, the Committee was aware of wider evidence that active travel is an important contributor to incidental activity, for which weight outcomes were available. Similarly, although the evidence on sedentary behaviour was limited, the Committee recognised that sedentary activity would be reflected in the evidence considered on TV viewing or other screen time, or total leisure time activity. (Total leisure time activity is any physical activity outside of school or work, including play but excluding active travel.)

The Committee was of the view that some unexpected associations may have been the result of 'reverse causality'. This is likely to have been the reason for the observed association between higher consumption of non-nutritive sweeteners and higher weight or weight gain. The Committee discussed that people who have a higher BMI may be more likely to consume products with non-nutritive sweeteners.

The search strategy for the evidence reviews took an iterative approach. The search focused on systematic reviews with weight outcomes. An additional, limited search for primary studies was undertaken for several factors for which no systematic review evidence was identified. However, because of the limited nature of the search for primary studies, the Committee agreed that this evidence should not be considered alongside the systematic review evidence.

The consideration of interventions in specific settings (such as schools or workplaces) was outside the remit of this guideline but is covered by related NICE guidelines. The Committee recognised that these settings are important in facilitating improvements in physical activity and dietary habits. For children and young people, the support and encouragement of parents, carers and staff working in schools and clubs is vital in helping them to make good choices.

The approach taken to the evidence – focusing on systematic reviews – means that the Committee was unable to make recommendations for research on specific behaviours if the evidence appeared to be lacking. Furthermore, the Committee was of the view that recommendations for research on individual behaviours would be warranted only if it was likely that the factor would have an important effect on weight and was easily modifiable by individuals. The Committee considered that the most helpful new research to enhance the maintenance of a healthy weight would be on the impact and feasibility of making a 'package' of changes, in line with the broad range of recommendations in the guideline.

The Committee was aware that many practitioners would like specific quantitative information on behaviours associated with weight gain – for example, an upper daily limit for TV viewing or amount of sugar-sweetened drinks consumed. The Committee recognised that providing upper or lower thresholds for behaviour (such as limiting TV viewing to no more than 2 hours) can be helpful for some people. However, the Committee noted that in most instances, the outcome (such as weight, weight gain or prevalence of obesity) tended to increase or decrease (as appropriate) in line with the amount of the factor. This means that, for example, any reduction in TV viewing or increase in physical activity is likely to be helpful. Commonly reported thresholds may be driven by data analysis decisions rather than being true biological thresholds. The Committee was also concerned that people who most need to change their behaviour could be put off by goals that seem out of their reach. Therefore, any thresholds included in the recommendations are practical examples rather than absolute thresholds.

Evidence of a positive association between TV viewing and weight outcomes in children is strong. The Committee recognised that many of the studies included in systematic reviews had been undertaken before other forms of screen – such as tablets or smart phones – were in common use. However, these other forms of screen viewing are also likely to reduce overall physical activity level. There is little systematic review‑level evidence available on the effectiveness of 'active viewing' games consoles on weight. The Committee was concerned that parents may purchase these items under the premise that they can increase physical activity levels to maintain a healthy weight, when in fact their effect is unclear. The Committee's focused on leisure screen time rather than total screen time, because they recognised that screen time at school or work is unlikely to be within a person's control.

The draft report of the Scientific Advisory Committee on Nutrition (SACN) on carbohydrates and health was published during the development of this guideline. The PHAC considered the draft findings of the report, noting that the SACN review was the result of a robust process and that the draft recommendations were largely in line with this guidance. The PHAC noted in particular SACN's decision to use the term free sugars, which includes the sugars within fruit juice. SACN's identification of a relationship between consumption of free sugars and energy intake may have implications for including fruit juice in population recommendations for '5 a day'. The PHAC noted that although evidence of an association between fruit juice consumption and weight gain was not identified, many people, particularly some children, may consume large quantities of juice in the belief that is a healthier option, undermining their efforts to maintain a healthy weight. As a result, fruit juice is not listed as an alternative to sugar‑sweetened drinks in recommendation 3.

The Committee was aware that patterns of behaviours have been an area of interest for researchers in recent years. Members had hoped that they might be able to make stronger recommendations about, for example, time spent standing up, eating speed, meal planning or patterns in eating and physical activity across a week. However, no systematic reviews were available on these topics.

No systematic reviews were identified that considered the effectiveness of self-monitoring of weight among a general population. The Committee was aware that self-monitoring has previously been identified as an important tool for adults maintaining weight after weight loss. Because of the lack of systematic review‑level evidence, the Committee drew on existing NICE recommendations on proven behaviour change techniques (see recommendation 7 in NICE's guideline on behaviour change: individual approaches). However, the Committee was not able to indicate the optimal frequency or method of monitoring. The Committee was aware that a range of free or low‑cost apps are now available that may support monitoring. However, the effectiveness of these products is unclear. The Committee has therefore made recommendations for research.

The recommendations on behaviours that can help people maintain a healthy weight reflect the evidence considered by the Committee. However, members recognised that the habits recommended may be very different from many people's usual choices. The Committee therefore emphasised the importance of communicating the benefits of even gradual changes, and the fact that any improvements in dietary habits and physical activity level are likely to help.

The evidence considered by the Committee suggests that the acceptability of messages about weight differs across the population. The Committee noted that the way in which messages are framed or worded may make them less acceptable to some people, for example, people with disabilities, from different age or ethnic groups, or with different BMI or waist circumferences.

The Committee recognised that some of the messages for maintaining a healthy weight are complex. The Committee also recognised that although the range of potentially modifiable factors covered in the recommendations may be daunting for some people, emphasising a range of physical activity and dietary habits is vital.

The recommendations therefore emphasise the importance of tailoring messages according to local knowledge and the information needs of different groups.

The economic assessment for this guideline was based on existing economic modelling that NICE has undertaken for other guidelines. New modelling was not attempted because no new data were available on the sustainability of weight maintenance or loss over long periods of time. Furthermore, evidence of cost effectiveness for weight loss can be applied to the prevention of weight gain.

For adults, the Committee considered analyses based on previous cost effectiveness modelling for NICE's guidelines on lifestyle weight management (see the economic modelling report for NICE's guideline on lifestyle weight management services for adults) and preventing type 2 diabetes: population and community interventions (see the report on cost-effectiveness evidence and methods for economic modelling for NICE's guideline on population and community interventions for preventing type 2 diabetes). Previous modelling shows that at least a 1 kg per head weight loss among overweight or obese adults, if maintained for life, is likely to be cost effective, provided that the cost per person of intervening is less than £100. This suggests that preventing at least a 1 kg weight gain for the same cost will also be cost effective. The Committee noted that the effect size for many of the observed associations in the evidence reviews was higher than 1 kg per person, or is likely to be so if a collection of factors is undertaken as part of a 'multicomponent approach'. Previous modelling has also shown that very low‑cost public health interventions (costing less than £10 per head) are likely to be cost effective for an average weight loss of less than 1 kg per head. Based on this previous work, the Committee concluded that the types of approaches suggested in this guideline are likely to be cost effective, particularly because the recommendations may replace ineffective or incorrect advice.

The Committee noted that the concept of 'maintaining a healthy weight' used for adults needs to be modified for children and young people because of growth in height and because a healthy BMI increases from age 6 to age 18. The concept for most children is instead about 'maintaining a healthy BMI for one's age'. (For more information see NICE's guideline on lifestyle weight management services for children and young people.)

For children and young people, information was obtained from modelling for the cost effectiveness of lifestyle weight management in children. The evaluation is more complicated, because to achieve a healthy weight, overweight and obese children do not necessarily have to lose weight but may be able to maintain weight while growing taller ('growing into a healthy weight'). In some cases, they may put on weight at a slow rate and still achieve a healthy weight in future by growing sufficiently quickly in compensation. However, if they are overweight or obese when they have finished growing in height they will need to lose weight to achieve a healthy weight.

Previous economic modelling estimated that interventions for overweight children costing £100 per head would usually be cost effective from a public sector perspective. This would be the case if a group of overweight children moved to a lower average weight trajectory and this was maintained for life. (This is true for a weight loss of as little as 0.5%). Therefore, by inference, interventions that prevent a child moving onto a 0.5% higher average weight trajectory (or greater) for life would also be cost effective.

Brief advice interventions were inferred to be cost effective when their low cost was considered against the cost of weight management interventions. The Committee noted that activities and interventions to help people maintain a healthy weight and prevent overweight and obesity are already ongoing in many local areas. The recommendations in this guideline are likely to make interventions more effective and there is virtually no cost attached to improved, more specific advice.

Based on this previous work, the Committee concluded that the types of approaches suggested in this guideline are likely to be cost effective.