|Home | About | Journals | Submit | Contact Us | Français|
The aim of this review is to provide an overview of (1) underlying mechanisms of the effect of portion size on energy intake, (2) external factors explaining the portion size effect and (3) interventions and measurements aimed at food portion size.
Previous studies have shown that portion sizes have increased in recent decades. Many experimental studies have been conducted to unravel the mechanisms underlying the portion-size effect on food intake (e.g. the appropriateness mechanism, the ‘unit bias’ mechanism, the ‘previous experience/expectation’ mechanism, the ‘visual cue’ mechanism and the ‘bite size’ mechanism). In addition, external factors have been found to drive food portion selection and consumption (e.g. value for money, mindless eating, levels of awareness, estimation bias. Research on several interventions (ranging from ‘providing information’ to ‘eliminating choice’) have been conducted, but remain scarce, especially intervention studies in which portion size is a key focus in weight loss. Moreover, only three new instruments with respect to portion control behavior have been developed.
There is considerable evidence for the portion-size effect on energy intake. However, the work on interventions targeting portion size and measurements for portion control behavior are limited. Moreover, from the literature it is not yet clear what type of interventions work best, for whom and in what context.
In the past few decades, the prevalence of obesity and overweight in adults has increased, more than doubling since the 1980s . Overweight and obesity are related to various health problems such as diabetes mellitus type 2, cardiovascular disease and several types of cancer . Moreover, overweight and obesity are associated with certain psychosocial problems, such as depression and anxiety disorders . In tandem with the increased prevalence of overweight and obesity, an increase in food portion sizes has also been observed . These growing food portion sizes might well be one of the factors contributing to the increased prevalence of overweight and obesity in adults.
In recent years, several studies have been conducted to shed light on the mechanisms behind the portion-size effect. Moreover, increasing attention has been paid to interventions aimed at portion size. This paper presents an overview of recent developments. Notwithstanding the fact that the portion-size effect is also present in children, this paper focuses on the adult population, as the effects are largest in adults [5•, 6•].
Very few studies have been conducted into trends in food portion sizes, with most carried out in or before 2009. Studies on developments in food portion sizes have been conducted in the USA [4, 7–10], the UK [11, 12], Denmark  and the Netherlands . All of these studies showed that portion sizes of numerous energy-dense foods have increased in the past decades, also highlighting the introduction of ‘super-sized’ portions. Examining changes in food portions between 2000 and 2009, Young and Nestle  showed that this trend has continued into the present century, noting the introduction of many new larger sized portions . With respect to home-cooked meals, Wansink and Payne  have demonstrated that portion sizes, as stated in a popular cookbook, have increased over the past 70 years , while Eidner et al.  found the same trend over the past 100 years in the case of a Danish cookbook . Given this apparent continuing and constant change in portion sizes and noticeable international differences , it is important to follow up these studies and continue the monitoring of food portion sizes globally.
Numerous studies have demonstrated that people’s energy intake increases when they are offered larger portions. In a meta-analytic review, Zlatevska et al. [6•] showed that doubling a food portion leads, on average, to an increase in energy consumption of 35% [6•]. In their Cochrane review, Hollands et al. [5•] also found a consistent effect of portion size on energy intake. They estimated that the energy intake from food and non-alcoholic beverages attributable to differences in product sizes was between 215 and 279 kcal/day [5•]. Benton  emphasized the fact that we need more studies of real-life situations to establish the true effect of portion size on energy intake in daily life . Such a study was conducted by French et al. , who demonstrated that exposure to a high-energy lunch over a period of 6 months led to significant increases in energy intake and weight gain in a real-life work setting .
Many experimental studies have been conducted over the past decade to unravel the mechanisms underlying the portion-size effect on food intake. Dual process theory may be important to the understanding of these mechanisms, distinguishing between a system of deliberate, conscious reasoning and a more automatic and fast system of associative reasoning . Some studies have attempted to unravel the deliberate processes that steer portion-size selection using a cognitive approach to explain the portion-size effect whereas other studies have attempted to unravel the non-deliberate processes using an automatic approach in explaining the portion-size effect [21, 22•]. Below, we briefly summarize the key mechanisms that are frequently highlighted when explaining the portion-size effect. A more in-depth and comprehensive explanation of the underlying mechanisms of the portion-size effect can be found in recently published papers by Herman et al.  and English et al. [22•, 23].
Additional factors that are not addressed by the portion-size effect—or the way the portion is served—may also impact on the portion size consumed. A wide range of external factors influence food consumption (and surplus portion-size intake), but a few important external factors that are frequently suggested in the literature as affecting the portion size selected and consumed are explained below.
Although we know increasingly more about the portion-size effect and its working mechanisms, interventions tackling the portion-size effect are still scarce. Nevertheless, some interventions have been developed and tested recently (see Table Table1).1). Interventions can be categorized according to the ladder of interventions of the British Nuffield Council on Bioethics , starting with simply providing information and ending with the more rigorous intervention of eliminating choice (i.e. not offering large portion sizes at all, see Fig. 1).
As Table Table11 shows, most of the interventions that have been developed and tested are on the lower rungs of the ladder. Some interventions showed promising results, such as using segmentation cues in food packaging , forming implementation intentions  and the use of other self-regulatory, portion control strategies [31, 61•].
Studies with portion size as the key focus in weight loss interventions are still very rare. A study by Poelman et al.  showed that a decrease in BMI can be achieved by focusing on different strategies, solely to control food portion sizes. The effects on BMI were mediated through portion control behaviours [61•]. This is in accord with the findings of the SHED-IT trial, where portion size was found to be one of the mediators of the intervention effect on body weight at 6 months, alongside physical activity .
More research into interventions aimed at portion size, in which all rungs of the ladder are taken into account, is required. The higher rungs of the ladder might not be reached without government regulation and thus government action will need to be combined with non-regulatory interventions to ensure use of the entire ladder of interventions aimed at tackling the portion-size effect (see e.g. Marteau et al. ). Marteau et al.  also noted that public acceptance is of crucial importance if governments and other private parties are to act ; however, public acceptance of the government intervening in the food environment is in general rather low. In their study of the public acceptance of policies to reduce the consumption of sugar-sweetened beverages, Gollust et al.  found that American consumers appeared to be most in favour of labelling and restricting sales on school property, while the least support was found for taxes and portion-size restrictions. It could well be that public acceptance decreases as interventions climb the ladder of interventions although this might be slightly different in the case of interventions targeted at children . More insight into the public acceptance of policy measures targeted at portion size will help in the development of strategies to enhance acceptance of policies.
In addition to public acceptance, it is also of importance to include intermediaries, such as restaurant owners in the development of strategies to control portion size. For example, Gase et al.  studied a voluntary programme in which restaurants offered reduced sized portions. Although they concluded that the programme was feasible, many barriers to offering reduced sized portions were identified by the restaurant owners, such as logistical barriers and concerns about revenue loss. Moreover, healthy eating was mainly seen as the responsibility of the consumer .
With the increasing attention being paid to interventions, the need for measurements of portion control behaviour has emerged. In recent years, three new instruments have been developed (see Table Table2).2). Fast et al.  developed a portion control self-efficacy scale, which showed good validity and reliability, with scores on the scales associated with dieting success [70, 31]. derived 32 portion control strategies from the literature, with the overall Cronbach’s α of the scale valued at .82 and showed an association between the use of portion control strategies and reduced BMI .
The portion control practices questionnaire developed by Spence et al.  consists of three subscales: the measurement strategy (use of guidance, i.e. by means of suggested serving sizes), the eating strategy (practices such as eating slowly) and the purchasing strategies (buying or ordering smaller amounts). Cronbach’s α of all subscales was ≥.78, and an association was found between the eating strategy scale and reduced BMI, whereas both the eating and purchasing strategies were negatively associated with pizza consumption . These studies by Poelman et al. (2013) and Spence et al.  suggest that portion control strategies may improve food intake and support weight control. However, more research is needed into the validity of portion control behaviour scales and the associations with food intake and BMI [31, 71].
Clearly, there is considerable evidence for the portion-size effect. However, there is still a lot of work to be done with respect to interventions targeting portion size. Although we conclude that a promising start has been made in relation to interventions targeting portion size, it is not yet clear what type of interventions work best, for whom and in what context . Specific attention should be directed to the public acceptance of portion-size interventions and in particular portion-size policies. Based on the very limited number of studies thus far, targeting portion size seems a promising strategy for weight loss interventions but this should be explored further. Interventions should be developed and tested in real-life settings . It is of importance to include the whole spectrum of potential interventions, from providing information to eliminating choices, from voluntary changes to regulatory interventions . Moreover, valid measurements should be developed to evaluate the effects of interventions. While a few portion control scales have been developed, more research is needed into their validity and the associations with food intake as well as BMI.
We would like to thank Frédérique Rongen (MSc) for her assistance in preparing the manuscript.
Ingrid Steenhuis has received compensation from Brickhouse Academy for providing training to dieticians and other paramedics about behaviour change and the obesogenic environment (including portion size) and royalties from Scriptum Publishers and SWP Publishing for writing two books—one on SMARTsize and one on relapse.
Maartje Poelman has received royalties from Scriptum Publishers for writing a book on SMARTsize.
This article does not contain any studies with human or animal subjects performed by any of the authors.
This article is part of the Topical Collection on Etiology of Obesity