Size matters: Developing portion size guidance for consumers

The various factors that influence energy consumption through mechanisms related to appetite, satiation and satiety are the subject of extensive scientific investigation (Benelam 2009). Portion size is one of those factors but the issue of offering guidance on portion size raises an interesting paradox. On the one hand, under experimental settings, manipulating the amounts of individual foods offered consistently influences the amounts consumed on that occasion (Hetherington & Blundell-Birtill 2018), and portion sizes for many foods and drinks have been increasing over the last few decades at the same time as rates of obesity have soared (BHF 2013). It seems highly likely that this portion inflation is one of the causal contributors to the passive overconsumption of energy that leads to weight gain and obesity, along with low levels of physical activity and other food-related factors such as energy density (Rolls 2017). On the other hand, however, it can be difficult to formulate useful guidance on what constitutes an ‘appropriate’ portion size for the various foods that are part of typical diets. While it is possible to identify the broad nutritional balance of diets that can be regarded as health-promoting (as proportions of energy), and to translate that into a broad balance of different foods, as many food guides have done, for example, the UK Eatwell Guide, not all of these are accompanied by guidance on what is regarded as an appropriate portion for a comprehensive range of foods. Guidance on the proportions of nutrients or foods that make up a healthy, balanced diet is simpler to derive because it depends less on individual variation: the proportions can be applied to all, scaled up or down, depending on individual energy requirements. However, translating dietary proportions into fixed portion sizes with specific energy values involves assumptions about individual needs and habits, and this increases the likelihood of consumer confusion or misinterpretation. As for most biological phenomena, there is wide variation in energy needs; therefore, what constitutes an appropriate portion of any energy-containing food will depend on individual requirements. Other factors such as the choices of different foods and how some foods fit into an overall diet may vary considerably. What is a portion of milk? A glass or a splash in a cup of coffee? How much is a portion of bread? A slice of toast at breakfast or four slices in sandwiches as a substantive meal? Giving a range might lead people to select an extreme, irrespective of their particular needs, and so could do more harm than good. Offering advice on the average is possible but how can an individual be usefully guided as to how much is appropriate for them? These are the issues that the British Nutrition Foundation (BNF) has grappled with in trying to develop portion size guidance that is scientifically based, and at the same time helpful to the food industry and consumers alike. For the food industry, information on portion size is likely to be useful in product development, for developing internal nutrition guidelines and in consumer communications. While providing information on portion size may be challenging, it is arguably an important aspect of eating to tackle, particularly as it is estimated that UK adults are consuming 200–300 kcal more than they need per day (PHE 2018a). It appears that consumers would value information on portion size – research conducted when developing portion size guidance in Ireland found that the majority of participants thought that information on portion size was important and would like to learn more (Safefood 2014). Food-based dietary guidance (FBDG), from countries such as Ireland, the US and Australia, includes comprehensive advice on portion sizes for each of the main food groups. However, while the UK Eatwell Guide (PHE 2018b) includes portion size information for fruit and vegetables, fruit juice, fish and some information on the amounts of red and processed meat to consume, it does not include information on portion or serving sizes for foods from all the groups included in the guidance, and this is a potential ‘missing link’ in translating the proportions of the food groups shown in the guide into practice. Our aim was to provide information on suggested portion sizes for foods from the main food groups, to help people eat a diet that is consistent with the Eatwell Guide. We wanted the resource to be practical and accessible for as many people in the population as possible, not only to those already highly engaged with nutrition and health. The work was funded by members of the food industry, who formed a Working Group to provide information on portion size issues in their sector. Valuable information was gathered from this group on the portioning of foods for sale; for example, whether the majority of bread purchased in the UK is medium or thick sliced and the range of actual sizes available in supermarkets for foods such as chicken breasts or fish fillets. An Expert Group comprising academics in the field of portion size and behavioural nutrition provided advice and guidance throughout the process and included observers from Public Health England and the Department of Health and Social Care. We began by reviewing guidance from other national FBDG, specifically Australia and New Zealand, Canada, Ireland and the US. While different approaches had been taken in coming to the portion size recommendations given, in all cases this involved using information on food and drink consumption from national surveys. Each of these guidelines divided foods into food groups and gave portion size information along with guidance on the number of servings to provide from each of the groups and this often varied by age group and, in some cases, activity level. In common with approaches in other countries, we began by looking at UK data from the National Diet and Nutrition Survey (NDNS) (PHE 2016) on portion sizes consumed for a number of foods, using those pictured on the Eatwell Guide as a starting point. We then compared this with information on portion sizes from major retailers – either the actual portion size of foods sold in a defined portion, such as bagels, sausages or individual pots of yogurt, or the suggested serving size on prepackaged foods such as pasta or mince. Having looked at these data on portion sizes, we then developed a series of 7-day test diets to investigate how portion sizes of different foods could fit into a diet, as well as the frequency with which foods from different food groups would be eaten across a day. These diets were developed to fit within current nutrient- and food-based recommendations (e.g. providing less than 5% of energy from free sugars, at least 30 g of fibre, and at least five portions of fruit and vegetables per day). Having settled on portion sizes for each of the foods included and a suggested range of frequencies for each of the food groups (e.g. 2–3 portions of dairy or alternatives per day), we then modelled the effects of following this advice in different ways on energy intakes. That is, we compared the estimated daily energy intake if selecting the smallest portion sizes within the range recommended and the lowest frequency of consumption for each food group with the estimated energy intake if consuming portion sizes at the high end of the recommended range with the highest frequency for each food group. The energy values calculated at both extremes were within estimated average requirements for adults. Looking at data on portion size in the context of dietary recommendations throws up an inherent paradox. Given that very few people in the UK meet all dietary recommendations, any diet developed that fits all of these will be unrepresentative of how the majority of people eat and so potentially unrealistic or impractical. It also raises the question of whether portion size guidance should be based on what people actually eat or on what the government suggests they should eat, or a mix of the two. In practice, we have taken a pragmatic approach and looked at portion sizes for different foods on a case-by-case basis. We were clear that we wanted any suggested portion sizes to be practical and so recommendations for foods sold pre-portioned, such as slices of bread or fish fingers, are based on what is actually available for people to buy. While serving size suggestions provided on packs of flaked cereal are typically 30 g or 35 g, NDNS data indicate that the most commonly consumed portion size is 40 g and so this is the portion we selected, given that its energy content of about 150 kcal is not excessive. Conversely, the average retail portion size for nuts is equivalent to over 250 kcal and so a smaller portion size was selected, as this was considered more appropriate given that nuts are typically eaten as a snack. We were also conscious of the issue of food waste and so avoided suggesting portion sizes, for example half a bagel, that might lead to food being thrown away. The next stage of the work was to take the portion size information that we had calculated and develop this into a consumer-friendly resource. To do this, we conducted a series of focus groups, the first of which looked at views on portion size, how people think about food and drink portions in practice, and the kind of information that would be welcomed to provide guidance on portion size. Participants represented a range of life stages and backgrounds but consistently talked about portion size selection as something intuitive, based on their habits and appetite. It was clear that portion size was not considered in isolation but in the context of the meal or snack being consumed and the diet overall. The outcomes of these groups were used to develop a range of potential designs and methods of measuring portion size, which was then reviewed in the second stage of consumer research. The focus group reviewed the draft designs and discussed whether the routes presented would help them choose appropriate portion sizes. There was a clear preference for a very simple resource, outlining the top-line messages on portion size that was visually appealing and could be displayed in the home. However, the focus group was also keen to have access to more detailed information on portion sizes for a range of different foods, either in a longer hard copy resource or on a website or ‘app’ that could be used on their smartphone. In each of the focus groups, participants highlighted the issue of individual variation and the need for any guidance to be flexible. This has been a consideration in developing the guidance, including advice on adapting it for different people and acknowledging that ‘one size does not fit all’. We hope that the final version, based on the outcomes of the focus groups, which is now in the public domain and can be assessed via the British Nutrition Foundation website (www.nutrition.org.uk), will be a valuable resource to help people choose appropriate portion sizes within a healthy, balanced diet. The British Nutrition Foundation is grateful to the following companies who provided financial support for the project: Alpro, Asda, Associated British Foods, Mars, Marks & Spencer, Sainsbury's, Tesco, Tetrapak and Waitrose, and to the Expert Group comprising Prof Martin Wiseman, Prof Peter Rogers, Dr Laura Thompson, Prof Marion Hetherington, Dr Charlotte Evans and Prof Mike Rayner with observers from Public Health England and the Department of Health and Social Care. The portion size project is an independent piece of work by the Foundation, developed with guidance from the expert panel.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.