The double burden of food insecurity

There are many approaches to describing food security. One of the definitions coined by the World Summit on Food Security⁽¹³⁾ characterises food security as a situation in which, ‘… all people at all times have physical, social and economic access to sufficient, safe, and nutritious food to meet their dietary needs and food preferences for an active and healthy life’. The proposed definition highlights the four dimensions of food security: availability (related to the supply chain), access (physical, social or economic), utilisation (nutrient use by the body) and stability^{(13,Reference Jones, Ngure and Pelto14)} . Therefore, food insecurity is a complex social issue that requires a multidisciplinary approach with insights and experiences shared by policymakers, retailers, healthcare providers and people directly affected.

Historically, food insecurity was measured using hunger or anthropometric measures such as underweight status or wasting and stunting in children (on an individual level) and the prevalence of undernourishment at the population level⁽¹⁵⁾. These are still valid approaches in low-income countries, where food insecurity tends to be unequivocal with hunger and undernutrition. In contrast, in middle- to high-income countries, in parallel to hunger, obesity is increasingly becoming a strong predictor of food insecurity^{(15–Reference Carvajal-Aldaz, Cucalon and Ordonez17)}. It is commonly overlooked, that by definition, overconsumption is also a form of malnutrition, and despite excessive energy consumption, it may result in nutrient deficiencies (‘hidden hunger’), both contributing to poor physical and mental health as well as the development of chronic diseases⁽¹⁵⁾. Although the link between food insecurity and obesity may seem counterintuitive at first, there are several factors that may explain this relationship, which will be discussed later in this review.

The food system in the UK is burdened with all forms of malnutrition, including people experiencing hunger, but also those who are living with obesity. Hence, system-wide interventions focusing solely on individuals' responsibility and consumer education are insufficient in improving population health. To tackle the obesogenic environment, a whole system approach is needed, which ‘…promotes moving from individualistic health problems, and simple linear models towards understanding the complexity through systems, organisations and environments’^{(18–Reference Hodgins, Griffiths, Scriven and Hodgins20)}.

The ‘food system’ approach

The recent cost-of-living crisis has amplified the problem of food insecurity in the UK⁽²¹⁾. According to the Scottish Health Survey⁽²²⁾, 9 % of adults were worried about running out of food, 6 % ate less and 3 % ran out of food. In 2019/2021 combined, the highest levels of food insecurity were among single parents (34 % worried they would run out of food) and single adults under age 65 (10 % worried they would run out of food)⁽²²⁾. The latest tracking figures published by the Food Foundation are even more stark⁽²³⁾. From an online survey of over 10 000 adults in the UK, 18 % or 9⋅3 million households experienced food insecurity in January 2023⁽²³⁾. These food-insecure households are more likely to be cutting back on purchasing healthy foods such as fruit, vegetables and fish. The report highlighted that in households with an adult limited a lot by disability, food insecurity was three times higher; half of the households with low income, receiving universal credit, report food insecurity; and non-white ethnic groups are at higher risk of food insecurity than white ethnic groups.

Increasing food prices create a huge challenge for those with lower income. The cost of a basic basket of food has increased by approximately 20–25 % between April 2022 and February 2023, and food inflation has risen by 18⋅2 % in the 12 months up to February 2023⁽²⁴⁾. The poorest fifth of UK households would need to spend 50 % of their disposable income on food to meet the cost of the UK government-recommended healthy diet, in contrast to just 11 % for the richest fifth⁽²⁵⁾. Healthier foods are twice as expensive per kJ, compared to less healthy foods⁽²⁵⁾.

Increasing food prices are not just regulated by industry and the top British retailers; hence a more holistic approach needs to be considered. The UK food system is intricate and includes a wide spectrum of interconnected stakeholder networks, involved in production, through processing, sales, consumption and waste management^{(Reference Hasnain, Ingram and Zurek26)}. Each network has its own agendas and priorities, hence aligning common goals can be extremely challenging. For this reason, a ‘food systems’ approach is a preferential outlook, as it allows to the exploration of different perspectives and trade-offs of the transformations, to help achieve a compromise between often competing priorities of the food systems actors (e.g. profit-orientated retailers v. health-seeking consumers)^{(Reference Hasnain, Ingram and Zurek26)}. It also helps to understand the multifaceted factors that drive food system components, e.g. geographical, social, political, legal, fiscal, environmental, etc. Food systems mapping has become increasingly popular in recent years because it can support finding complementary, and potentially synergistic solutions to a healthier and more sustainable food environment^{(Reference Hasnain, Ingram and Zurek26,Reference Béné, Oosterveer and Lamotte27)} .

The paradox of food insecurity and obesity

Despite many UK government reports on tackling obesity, the trends for obesity (defined as BMI >30 kg/m²) in men and women are increasing in Scotland⁽²⁸⁾ and England⁽²⁹⁾. In Scotland, 67 % of adults are living with overweight and obesity and nearly a quarter of children starting primary school were at risk of overweight and obesity⁽²⁸⁾. People living with obesity are more likely to experience a range of serious and chronic health issues including CVD such as stroke and heart disease and type 2 diabetes^{(Reference Jayedi, Soltani and Motlagh30,Reference Powell-Wiley, Poirier and Burke31)} .

Obesity is disproportionately represented in socially disadvantaged groups, for those living in areas of deprivation, a trend that has become more marked over the past 60 years^{(Reference Bann, Johnson and Li32)} and more recently during the coronavirus disease-2019 pandemic^(22,28,29) . In general, the statistics in 2021 indicate worrying patterns, with the prevalence of obesity and overweight increasing across the age groups, to age 55–64, with more adult males than females living with overweight^(28,29) . Children living in the poorest areas are four times more likely than children from the richest areas to be severely obese when they arrive at primary school⁽¹²⁾. They are five times more likely to be severely obese when they leave it. Sixteen per cent of people in the lowest income group live with diabetes: more than twice the percentage of those in the highest income group⁽³³⁾.

So why is it that those who struggle to afford food are also the most affected by the obesity epidemic? A number of researchers have explored this seemingly contradicting link between food insecurity and increased risk of obesity^{(Reference Nettle, Andrews and Bateson16,Reference Carvajal-Aldaz, Cucalon and Ordonez17,Reference Johnstone and Lonnie34–Reference Larson and Story37)} . The paradox highlights the social bias towards an assumption that food insecurity is experienced by people with low body weight. The peculiarity of the dysfunctional food system is that obesity can co-exist with hunger and/or low-quality diet. Most of the literature has consistently shown that there is a significant association between food insecurity and obesity, specifically in women from high-income countries^{(Reference Carvajal-Aldaz, Cucalon and Ordonez17,Reference Adams, Grummer-Strawn and Chavez38,Reference Martin and Lippert39)} . In the United States, people living with food insecurity have a 32 % higher odds of obesity compared to food-secure individuals^{(Reference Pan, Sherry and Njai40)}. Several hypotheses were proposed to explain this link: (1) low food security is associated with obesity because of the high-energy, palatable food consumed by low food secure populations, or (2) low food security is associated with obesity because of the limited knowledge, time and resources that low food-secure populations experience to engage in healthful eating and exercise^{(Reference Dhurandhar41–Reference Drewnowski and Darmon43)}.

These punitive explanations however exacerbate the weight stigma, placing the responsibility on the individual while the obesity issue is much more complex. Factors linking food insecurity and obesity are listed in Table 1 (list not extensive or in priority).

Table 1. Factors contributing to obesity among those living with food insecurity^{(Reference Carvajal-Aldaz, Cucalon and Ordonez17,18,Reference Hodgins, Griffiths, Scriven and Hodgins20,Reference Adams, Grummer-Strawn and Chavez38,Reference Pan, Sherry and Njai40,Reference Kendall, Olson and Frongillo42)}

There is growing scientific consensus that certain dietary behaviours are associated with obesity, including higher consumption of sugar-sweetened beverages, fast food and other foods with higher energy density and lower consumption of fruits, vegetables and whole grains^{(Reference Rauber, da Costa Louzada and Steele44–Reference Rauber, Steele and Louzada46)}. At the same time, there is evidence that low-income households tend to consume more of these ‘obesogenic’ foods and beverages and fewer ‘health-enhancing’ foods and beverages^{(Reference Drewnowski and Darmon43,Reference Sawyer, van Lenthe and Kamphuis47)} . The confluence of these factors hardly seems a paradox. Studies have shown that energy-dense, nutrient-poor foods cost less and that cost is an effective driver of behaviour^{(Reference Sawyer, van Lenthe and Kamphuis47)}. In addition to monetary cost, time cost may influence eating behaviours. For example, the U.S. Department of Agriculture's Thrifty Food Plan, a menu plan demonstrating ways to limit financial costs while optimising nutrition, has been estimated to require more than twice the number of hours of food preparation time that the average American food preparer spends⁽⁴⁸⁾.

Furthermore, it is not uncommon that access to healthy and nutritious foods is restricted by geographical factors. The most deprived areas are often characterised as ‘food deserts’ (locations with little access to nutritious foods)⁽⁴⁹⁾. It has been estimated, that about one in ten deprived areas in England and Wales can be classified as a ‘food desert’⁽⁵⁰⁾ suggesting that restricted access to a food store is an important barrier to eating healthily among people living in those areas. More recently, a new term has been popularised – ‘food swamps’^{(Reference Rose, Bodor and Swalm51)}, which is supposed to be an even better predictor of obesity than ‘food deserts’^{(Reference Cooksey-Stowers, Schwartz and Brownell52)}. ‘Food swamps’ describe neighbourhoods with an abundance of less healthy food options (e.g. takeaways, fast-food chains, convenience stores, etc.)^{(Reference Cooksey-Stowers, Schwartz and Brownell52)}. Evidence from the UK Biobank study shows that income and increased exposure to fast-food outlets were independently associated with BMI, body fat, obesity and frequent processed meat consumption^{(Reference Burgoine, Sarkar and Webster53)}.

The uncertainty about food access can generate stress and anxiety which may trigger physiological and psychological mechanisms underpinning overeating and low-quality diet^{(Reference Anversa, Muthmainah and Sketriene54)}. For example, food insecurity may be associated with compensatory feeding practices (i.e. high-energy supplements) by parents concerned that their children are not getting enough to eat^{(Reference Feinberg, Kavanagh and Young55)}.

Lastly, some unhealthy behaviours can be related to poor mental health and other associated health conditions^{(Reference Conry, Morgan and Curry56)}. Low socioeconomic status (SES) is a risk factor for a variety of disorders, including mental and behavioural problems, which may trigger a lifelong chain of physical illnesses^{(Reference Hudson57,Reference Kivimäki, Batty and Pentti58)} . Kivimäki et al.^{(Reference Kivimäki, Batty and Pentti58)} reported that low SES was associated with nearly one-third of the studied diseases or health conditions (eighteen out of fifty-six). Multimorbidity (the presence of two or more long-term health conditions) was shown for sixteen of those diseases in this vulnerable population group. Interestingly, the study revealed that the disease cascade in those with low SES is strongly interconnected with mental health problems. For instance, the risk of the co-occurrence of obesity with mood disorders (e.g. depression or anxiety) was nearly five times higher among participants with low SES, in comparison to higher SES^{(Reference Kivimäki, Batty and Pentti58)}. Similar findings were reported by the latest Great Britain Census 2021 data⁽⁵⁹⁾. Adults with moderate-to-severe depressive symptoms had higher odds of food insecurity (3⋅1 higher odds) than those with no-to-mild depressive symptoms⁽⁵⁹⁾. Both studies^{(Reference Hudson57,Reference Kivimäki, Batty and Pentti58)} were of cross-sectional nature, so the link is not causative and can be interpreted in a bidirectional manner; does obesity promote ill mental health or poor mental health contributes to weight gain? Conversely, longitudinal data reported a worrying trend, that multimorbidity is currently observed at younger ages, when compared with previous generations, with income and BMI being key determinants of the morbidity trajectories, understood as the co-occurrence of multiple diseases within the same individual developed over a period of time^{(Reference Canizares, Hogg-Johnson and Gignac60,Reference Cezard, McHale and Sullivan61)} . Perhaps the co-occurrence of diseases and poor mental health explains why interventions focused on education and behaviour change in those from disadvantaged groups are not as effective when compared to groups with higher income^{(Reference Bull, Dombrowski and McCleary62)} and more holistic health approaches are needed in those with the lowest incomes.

It is interesting that interventions that focus on food and resources for individuals and education to improve knowledge for low SES populations can have limited impact. At worst, they can even widen inequalities. For example, a Mexican intervention based on cash and in-kind transfers for low-income families did not improve the diet quality and body weight status of the recipients as anticipated^{(Reference Leroy, Gadsden and González de Cossío63)}. On the contrary, the additional monetary resources contributed to weight gain, as a result of increased energy consumption rather than purchases of more expensive foods such as fruit, vegetables or fish^{(Reference Leroy, Gadsden and González de Cossío63)}.

What about diet sustainability?

Sustainable diets are defined as ‘dietary patterns that promote all dimensions of individuals’ health and wellbeing; have low environmental pressure and impact; are accessible, affordable, safe and equitable; and are culturally acceptable’⁽⁶⁴⁾. Care needs to be taken as not all healthy eating patterns are guaranteed to be environmentally sustainable, and conversely, a diet that has a low environmental impact is not necessarily nutritionally adequate. Both aspects need to be considered for planetary health^{(Reference Willett, Rockström and Loken65)}. The One Blue Dot report⁽⁶⁶⁾ from the British Dietetic Association details the need to support a change in eating habits that are both healthy and sustainable for the environment. Healthy, sustainable eating patterns have been associated with improved health outcomes, such as reduced risk of obesity and reduced rates of diabetes and heart disease and could thus result in reductions in total mortality by 6–16 %⁽⁶⁶⁾. The UK Committee for Climate Change Net Zero report⁽⁶⁷⁾ included specific dietary recommendations that UK intakes of ruminant meat (beef and lamb) and dairy should be reduced by 20 % and evidence to help consumers translate this into behaviour change are necessary.

Sustainability is a priority for each of the devolved governments in the UK. In Scotland, the Good Food Nation policy, first published in 2014, set out an aspiration to produce food that is ‘tasty to eat and nutritious, fresh and environmentally sustainable’⁽⁶⁸⁾. The Carbon Trust⁽⁶⁹⁾ analysis of the Eatwell Guide shows a lower environmental impact than the current UK diet, supporting this approach. The Carbon Trust estimates that if individuals moved from current eating patterns to the Eatwell Guide recommendations, a 31 % reduction in greenhouse gas emissions, 17 % saving on water use and a 34 % reduction in land use could be achieved⁽⁶⁹⁾. The National Food Strategy⁽¹²⁾ fully recognises these complex issues in the UK, where obesity and diet inequalities exist, and food system solutions must consider environmental impact.

Sustainable diets are being perceived as more expensive and inconvenient by consumers^{(Reference van Bussel, Kuijsten and Mars70)}. There is still a limited understanding of what behaviours can be classified as sustainable, with the main connotations relating to the environmental impact of food production, local and organic food choices and ethical considerations^{(Reference van Bussel, Kuijsten and Mars70)}. Especially in the context of the cost-of-living crisis, it becomes increasingly apparent that food affordability is becoming a key determinant of food choices and sustainability is of lower importance^(21,25,71) . However, in our discussions with patient and public involvement advocate a theme of food waste is becoming apparent. Although reducing food waste is not initially associated with sustainable behaviours by the participants, families with low SES reported that the ability to purchase the desired amount of fresh produce (e.g. by weight rather than pre-packed) would allow them to buy the required amounts of food to prepare a healthier meal, reduce food waste and save money. It is often forgotten that one of the dimensions of a sustainable food system is ‘food and nutrition’ which consist of the following metrics: food security, food safety, food waste and nutrition^{(Reference Béné, Prager and Achicanoy72)}. Hence, improving diet quality, reducing waste and targeting obesity among groups with low SES is in itself, a way of improving the sustainability of the whole food system.

The value of co-production

Co-production was conceptualised in the 1970s referring to the fact that service users were not recognised in the delivery of services^{(Reference Parks, Baker and Kiser73)}. Subsequently, Edgar Cahn, a civil rights law professor, developed the concept further by creating a system of time banks, with input from the volunteers, who were also service users. Cahn's work highlighted the importance of service users in the delivery of services and provided examples of how co-production could be achieved^{(Reference Needham74)}. Since then, co-production has become an increasingly popular concept in the field of service delivery and has been used to improve the efficacy of services and reduce costs. It has also been used to ensure that service users are more actively involved in the process to foster greater collaboration between service providers and service users^{(Reference Needham74)}. Although, initially grounded in the health research settings, currently co-production is being applied to other research fields that are seeking to improve health and well-being, e.g. health promoting retail environment. In the FIO Food project, the term ‘service users’ refer to consumers, while ‘service providers’ translates to retailers and policymakers, with all parties expected to have an interest in the results of the research.

There are several definitions of co-production^{(Reference Smith, Budworth and Grindey75)}, however, for the purpose of the present paper, we will use the definition by Slattery et al.^{(Reference Slattery, Saeri and Bragge76)} who describes it as a ‘meaningful end-user engagement in research design and includes instances of engagement that occur across all stages of the research process and range in intensity from relatively passive to highly active and involved’. To simplify, co-producing means people who use services, members of the public and professionals working together in a ‘partnership’ to produce research, meaning that research is carried out ‘with’ or ‘by’ members of the public, rather than ‘to’, ‘about’ or ‘for them’⁽⁷⁷⁾. There are other terms used in the literature, such as patient-oriented research, integrated knowledge translation research, lay representation, citizen engagement or co-design, which often intersect with the co-production term, but are not necessarily equivalent to it^{(Reference Boote, Baird and Sutton78)}.

Patient and public involvement is the most advanced form of co-production. Boote et al.^{(Reference Russell, Fudge and Greenhalgh79)} identified three distinct levels of involvement: (1) consultation (a process in which researchers seek public feedback on key aspects of the study); (2) collaboration (a continuous partnership between researchers and the public throughout the research process) and (3) public-led (where the public designs and oversees the research, and researchers are only invited to take part at public invitation). In the next section, we will describe how this concept was incorporated into our research.

The food inequalities and obesity: FIO Food project

The ‘Food Insecurity in People Living with Obesity’ (FIO Food) is a 3-year (2022–2025), £1⋅6m research project funded by the Transforming the UK Food System – Strategic Priority Fund^{(Reference Lonnie, Hunter and Stone80)}. The overall aim of the FIO Food project is to improve environmentally sustainable and healthier food choices in the UK food system, and to provide actionable evidence for policy on retail strategies to address dietary inequalities in two vulnerable groups (people living with obesity and food insecurity). Reducing obesity levels has been a public health priority in the UK for decades but we have not yet managed to achieve that goal^{(Reference Theis and White81)}. This is partly due to the range of factors that influence body weight. The FIO Food project supports the previously discussed whole-systems approach to consider this problem. A key challenge facing people living with obesity is being able to afford a healthy, balanced diet. Nutritionally poor and energy-dense foods that are often ultra-processed, are cheaper and more readily available^(24,25) . To start to address this challenge, we need better evidence on how to support healthier food purchasing patterns to improve their health and wellbeing, while considering environmental impact and sustainability.

Families on low incomes are more likely to be food insecure and they spend a greater proportion (three-quarters) of their monthly food budget in supermarkets⁽²³⁾. Supermarket promotions, advertising and online product placement decisions can impact this group's access to healthy foods. Our research is bringing together food-insecure people living with obesity, consumers, retailers, policymakers and academics to co-develop and test strategies that can support future transformative potential in the food system. To achieve this, we have designed an innovative four-part project (see Fig. 1).

Fig. 1. Project outline.