Our research on diet

We study how our environments influence what we eat, and which interventions could change this so that our diets can be healthier.


Why is improving diet important?

Nine of the 20 leading risk factors for ill-health and early death in the UK are dietary (Global Burden of Disease Study 2010), and are caused by:

  • eating a diet low in fruits (6th leading risk factor), nuts & seeds (8th), vegetables (13th), fibre (16th), whole grains (18th), seafood omega-3 fatty acids (15th) or polyunsaturated fatty acids (19th);
  • eating a diet high in sodium (10th) or processed meat (14th).

In addition, high body-mass index, which both diet and physical inactivity may contribute to, is the 3rd leading cause of ill-health and early death in the UK (Global Burden of Disease Study 2010).

Click on the images below to find out more about the work we have been doing in this area:


Key Studies:

  • How do consumers perceive smaller vs. larger bottles of sugary drinks?

    Sixteen consumers of sugary drinks were interviewed about their experiences of using four different bottle sizes of cola: 1500 ml, 1000 ml, 500 ml, and 250 ml.

    Participants perceived their consumption pace and amount to be greater with the smallest (250ml) compared to larger bottles. The smallest bottles were described as increasing the frequency of drinking occasions and encouraging consumption of numerous bottles in succession.

    Various factors associated with the smallest bottles were described as enabling consumption, like the perceived convenience and portability and the greater number of smaller vs. larger bottles available. Further work is needed to experimentally test the use of small bottles as a method for decreasing sugary drink consumption.

    Perceived impact of smaller compared with larger-sized bottles of sugar-sweetened beverages on consumption: a qualitative analysis. Mantzari, Hollands, Pechey, Jebb & Marteau, 2017

    Access full text


  • Can interventions in workplace cafeterias modify employees’ diets?

    Reducing excessive consumption of food and soft drinks is core to tackling the high rates of overweight and obesity in the UK and elsewhere. It is estimated that about one third of our daily energy intake is consumed while at work, with most of the food consumed not brought from home.

    This is a protocol for a pilot stepped wedge randomised controlled trial to estimate the potential impact of three sets of physical micro-environment interventions to reduce energy purchased in workplace cafeterias: 1) size of portions, packages and tableware; 2) availability of healthier vs. less healthy options; and 3) energy labelling.

    The results of this pilot study will inform the design and procedures for a larger trial.

    Physical micro-environment interventions for healthier eating in the workplace: protocol for a stepped wedge randomised controlled pilot trial. Vasiljevic, Cartwright, Pechey, Hollands, Couturier, Jebb & Marteau, 2017.

    Access full text


  • Does bottle size affect the amount of cola consumed at home?

    We assessed the feasibility of running a trial on the impact of bottle size on in-home consumption of cola, in which households received a set amount of cola each week for four weeks, in bottles of one of four sizes (1500 ml, 1000 ml, 500 ml, or 250 ml). 16 households were tested and we found the procedures to be both feasible and acceptable.

    Issues associated with the study design were identified, which may impact the validity of the primary outcome (consumption). These include consumption driven by study-determined supply and a failure to capture out-of-home consumption. Methods to avoid these would be needed before conducting a definitive trial.

    Impact of bottle size on in-home consumption of sugar-sweetened beverages: a feasibility and acceptability study. Mantzari, Hollands, Pechey, Jebb & Marteau, 2017.

    Access full text


  • Does presenting images of negative health outcomes lead to healthier food choices?

    Composite for BHRU websiteWe examined the impact on people’s choices of presenting food images paired with positive or negative images of the health consequences of eating those foods.

    Presenting images of negative health outcomes led to more healthy food choices, irrespective of whether they were paired with images of energy-dense snack foods or of fruit. Images of positive health outcomes did not alter food choices.

    This work provides insights relevant to health communication interventions where aversive visual images are used to alter the consumption of products that impact on human health.

    Pairing images of unhealthy and healthy foods with images of negative and positive health consequences: impact on attitudes and food choice. Hollands & Marteau. 2016

    Access full text


  • How does the public judge nudging to cut consumption of sugary drinks?

    How acceptable do people find nudges such as reducing bottle sizes of sugary drinks to prevent obesity? Does highlighting the non-conscious nature of nudging affect their acceptability?

    The results of our study, conducted with over 2000 UK and US participants, show that most people find such “nudges” to be acceptable interventions to prevent obesity. Although, highlighting the non-conscious nature of nudges does not alter their acceptability, the study found that taxing sugary drinks was only acceptable to a minority.

    But for both nudging and taxing, the acceptability of the intervention increased the more effective participants judged them to be. This suggests people are prepared to trade off dislike of an intervention for achieving a valued goal, such as tackling obesity.

    Public acceptability in the UK and USA of nudging to reduce obesity: the example of reducing sugar-sweetened beverages consumption. Petrescu, Hollands, Couturier, Ng & Marteau, 2016.

    Access full text.


  • Is acting on intentions to change diet socially patterned?

    Bringing together data from previous studies, we found no evidence that those from more deprived groups are less likely to act on their intentions to change their behaviour, including changing their diet.

    However, we found some differences between more and less deprived groups in terms of the influence of perceived control over behaviour (self-efficacy) on actual behaviour:

    • for those from less deprived groups, higher perceived control over their behaviour was associated with healthier behaviour;
    • for those from more deprived groups, perceived control over behaviour had no relation to health behaviour.


    This suggests interventions that do not require active engagement from people may be more effective at reducing health inequalities, as responses to these may be less likely to depend on attributes like perceived control over behaviour.

    Is the intention–behaviour gap greater amongst the more deprived? A meta-analysis of five studies on physical activity, diet, and medication adherence in smoking cessation. Vasiljevic, Ng, Griffin, Sutton & Marteau. 2016.

    Access full text.


  • What are the policy options to “Downsize” portions, packages and tableware?

    BMJ-infographic_portions_v5In this paper we outline a range of actions with the potential to avoid the excess consumption of food and drink that consistently occurs when people are presented with larger portions, packages and tableware size.

    The actions focus on reducing the size, availability and appeal of such larger sized portions, packages and tableware.
    The analysis considers: what interventions might work; where these interventions can be made; how they might be achieved; and who needs to act to enable the changes. This builds on our recent Cochrane Portion Size review.

    Downsizing: policy options to reduce portion sizes to help tackle obesity. Marteau, Hollands, Shemilt & Jebb. 2015.

    Access full text.




  • Can portion, package and tableware size increase consumption?

    shutterstock_41447758.jpgOur systematic review has produced the most conclusive evidence to date that people consistently consume more food and drink when offered larger-sized portions, packages or tableware than when offered smaller-sized versions.


    The size of this effect, based on combining data from 61 randomised controlled trials (6,711 participants), suggests that if sustained reductions in exposure to large sizes could be achieved across the whole diet, this could reduce average daily energy consumed from food by up to 16% (279 kcals) among UK adults.


    Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Hollands, Shemilt, Marteau, Jebb, Lewis, Wei, Higgins & Ogilvie. 2015.

    Access full text.

    • To view Cochrane UK coverage of this Cochrane review in the ‘Evidently Cochrane’ blog (from 15/09), click here.
    • To join the Twitter conversation about this Cochrane review, follow: @BHRUCambridge and #PortionSize



  • Can emoticons or colours on labels influence our snack choice?

    Chocolate smiling face on white backgroundIn an online experiment, neither an emoticon label (smiling or frowning face) nor a label colour (red, green or white) influenced people’s choice between a chocolate or a cereal bar.

    However, frowning faces on cereal bars did lead people to rate these as less tasty and less healthy. These results highlight the need for more research on the impact of emoticons on labelling.

    Making food labels social: The impact of colour of nutritional labels and injunctive norms on perceptions and choice of snack foods. Vasiljevic, Pechey & Marteau. 2015.

    Access full text.


  • Do less deprived households have healthier shopping baskets?

    Midsection of mature man carrying basket full of vegetables at homeWe looked at the purchases of 25,000 British households in 2010 to get a detailed picture of socioeconomic differences in the purchases of food and drink. Our research found that:

    • more deprived households purchased a larger proportion of energy from less healthy foods and drinks than less deprived households, and vice versa for healthier foods and drinks.


    This pattern of purchasing could contribute to the health differences we see between socioeconomic groups.

    Socioeconomic differences in purchases of more vs. less healthy foods and beverages: Analysis of over 25,000 British households in 2010. Pechey, Jebb, Kelly, Almiron-Roig, Conde, Nakamura, Shemilt, Suhrcke, & Marteau. 2013.

    Access full text



Publications related to diet: