July 15, 2020

expert reaction to study looking at dietary guidelines, global health and environmental targets

A study, published in the British Medical Journal (BMJ), has aimed to link food-based dietary guidelines with global health and environmental targets.

Prof Francesco Cappuccio, Chair of Cardiovascular Medicine & Epidemiology and Head of the WHO Collaborating Centre for Nutrition, University of Warwick, said:

“This study collated data from 85 food based dietary national guidelines issued around the world based on cross-sectional nutritional surveys of different quality.  The authors then compared them to WHO and EAT-Lancet Commission recommendations by modelling their health and environmental impacts.  The authors concluded that most national guidelines could be healthier and more sustainable.

“This analysis is a timely reminder that nutritional guidelines should consider factors that go beyond the health gains they can provide, such as environmental impact (reduction in anthropogenic land use, global water conservation, reduction in global greenhouse gas emissions) and sustainability.  There are limitations though.  The analysis is based on cross-sectional studies with often unreliable and incomplete data. Cause-effect relationships are presumed, and control for confounding difficult.  The health impact could be underestimated in many circumstances due to the omission of sodium consumption in the model.  High sodium is estimated to cause death and disabilities comparable to those accounted for by low consumption of whole grains and low fruits.  Furthermore, estimates of impact are modelled on the results of efficacy studies.  However, to inform policy actions we would need to establish the feasibility (or effectiveness) of our recommendations considering barriers and facilitators for their implementation in different settings (socio-economic, cultural, geographic, political, etc). Nevertheless, the present study is a very good platform to build a debate on the development of new sustainable food policies.”

Prof Elizabeth Robinson, Professor of Environmental Economics, School of Agriculture, Policy, and Development, University of Reading, said:

“In the UK there are similar concerns to those raised by Springmann et al (2020) that our diets comprise too much sugar, salt, and unhealthy fats, in part due to our relatively high consumption of ‘highly processed’ foods.  The “Eatwell Guide” recommends a diet similar to that discussed in Springmann et al (2020), focusing on consumption of at least five portions of fruits and vegetables each day, and more consumption of high fibre starchy foods.

“However, as highlighted in the recent House of Lords Select Committee on Food, Poverty, Health and the Environment report, titled “Hungry for change: fixing the failures”, there are barriers to following the Eatwell Guide including cost, which may be too high for lower-income households as a result of poverty; and wider factors that can make it harder for all households to make healthy choices.  These factors, that might be physical, social, economic, or cultural, and are sometimes referred to as the “food environment”, are not addressed explicitly in Springmann et al (2020).  This is not a criticism.  An important first step, as taken by Springmann et al (2020), is to determine the health and environmental impacts of dietary changes.  A next step is to understand and recognise the complexities of ensuring that people have access to, and the ability to choose, a healthy and sustainable diet, and why they may not choose, or not be able to choose, to change their diets accordingly.

“In an era of climate change, many of us need to be eating a more plant-based diet, containing less meat, and probably we will be healthier for doing so.  For those of us that continue to consume red meat and dairy, choosing UK-sourced products is typically a more positive choice for the environment, for our farming community, and for animal welfare.

“The World Health Organisation has declared climate change to be the greatest threat to global health in the 21st century, and the Lancet Countdown has identified a number of win-win outcomes for global health and climate change, including a reduction in the consumption of ruminant meat such as beef and lamb.  Low-meat plant-rich diets can reduce global emissions, benefiting the planet; reduce colorectal cancer and cardiovascular disease risk, benefiting the individual; and consequently reduce hospital admissions, thereby reducing pressure on what in many countries are often stretched public health care systems, benefiting those countries directly.”

Prof Alan Dangour, Director of Centre on Climate Change and Planetary Health, London School of Hygiene & Tropical Medicine, said:

“This is an important analysis that reports a wide range of potential benefits that could accrue from creating national dietary guidelines that align more strongly with both public health and global sustainability targets.

“Major questions remain about the ability and desire of people to change their diets and adopt national recommendations, and the willingness of governments fully to implement their mandated role in public health and environmental stewardship.

“The weaknesses of national dietary guidelines suggest that governments are failing to recognise the combined impact of the food system on both public health and the environment, and are likely to be dangerously ill-prepared for the major impacts that are forecast under future climate change.”

Prof Chris Reynolds, Professor of Animal and Dairy Science, University of Reading, said:

“As emphasized by Springmann et al (2020), there is widespread consensus that a reduction in the total global consumption of meat and dairy is important for climate change mitigation and reduced environmental impact of food production, and that for some people there will be health benefits from reducing their consumption of red and processed meat.  The carbon footprint of beef and lamb is to a large extent attributable to emissions of methane, a greenhouse gas with a short atmospheric half-life compared to CO2 and other greenhouse gases, and very much depends on the production system used, including how intensive it is, and how the animals are fed and managed.  In the UK the livestock sector is relatively effective in managing emissions compared with many other countries, using production systems with high animal welfare standards, and in recent years has seen substantial improvements in the sustainability of its livestock production systems.”

Dr Sarah Berry, Senior Lecturer in the Department of Nutritional Sciences, King’s College London, said:

“Everyone has a different opinion on what constitutes a healthy diet and there is a general lack of consensus, even amongst some experts, on the health effects of some foods. This is because our individual responses to food are highly variable and importantly food is very complex. Additionally nutritional scientists are making huge advances in understanding the health effects of foods and, therefore, the evidence is continuously evolving. We now know that categorising single foods into larger food groups is oversimplistic and can give highly variable and often conflicting results. For example, dairy as a food group includes cheese, butter and milk. The evidence to support a favourable health effect of cheese is strong, yet the sum of the current evidence shows that butter has unfavourable health effects.

“Additionally the health effects of all foods are dependent on the rest of the diet and the food that they may otherwise be displacing (displacement effect). For example, replacing biscuits for cheese would have a beneficial impact on health, whilst replacing pulses and nuts for cheese would have a negative impact on health. Therefore examining the health effects of food based dietary guideless, which use broad food groups, is challenging and any output is somewhat limited within the context of the complexity of food and dietary patterns.

“For example, the current study has not been able to unravel the true level of health effects of following food based dietary guidelines within the complexity of current dietary patterns. Further, although this was not the main aim of the research, it is important to note that the potential impact of food displacement and the differential health effects of single food items within a food group used for the guidelines are not able to be addressed within the current modelling. Additionally, by only comparing national and global dietary food based guidelines with the EAT-Lancet recommendations, the study is potentially underestimating the true beneficial health effects that can be achieved through diets. For example, there is a huge body of epidemiological evidence to show that particular dietary patterns, such as the Mediterranean diet, are beneficial for health. In most instances it is the subtleties of the dietary patterns than confer the most health effects rather than just food group changes.

“It would have been advantageous to compare the health impact of adherence to dietary guidelines versus established dietary patterns with a wealth of epidemiological evidence, rather than the ‘theoretical’ EAT-Lancet diet. Additionally, the study does not examine the barriers to people’s adherence to food based guidelines. Whilst this was outside the scope of the current study, this is key to understanding the level of adherence to food based dietary guidelines; if it is not feasible to follow these for the majority of the population due to access to food, or budget or cultural reasons, then they are not suitable for the population.

“Currently there is much debate as to the value of national and international food-based dietary guidelines. Further, it is often, in my opinion unfairly, claimed that national and international dietary guidelines are partly responsible for the growing global obesity burden and diet-associated ill-health. However, an important point that this research confirms, for the first time at a national and a global level, is that very few of us actually adhere to national and global dietary guidelines. For example, less than half the 85 countries examined fulfilled any of their food-based guidelines and none of the countries fulfilled all of their food-based guidelines. Given that this research demonstrates the predicted value of both national and global food based dietary recommendations at a population level in reducing diet related ill-health, this highlights to me that a big part of the problem is that we aren’t following dietary guidelines.

“The practical dietary recommendations that arise from this research are aligned with previous published research from large nutritional studies. In particular, we should be consuming more fibre-rich wholegrain foods and less processed meats. For example, to meet global dietary recommendations we need to increase our whole-grain intake by 241% and reduce our processed meat intake by 56%.

“When developing dietary guidelines, we must consider the practicality of implementing these at an individual level based on individual dietary and cultural preferences, access to food and budget. Therefore, although guidelines such as those laid out in the EAT-Lancet report may be more beneficial than current dietary recommendations at both a health and environmental level, the EAT-Lancet guidelines are not achievable for a large proportion of individuals across the globe.

“A major strength of this study is the inclusion of dietary guidelines from a large cross section of the world encompassing different demographics and different dietary requirements and health related issues. Many previous studies examining the effectiveness of dietary guidelines have tended to focus on countries where just non-communicable disease or just communicable disease are the main health issue, whereas this research encompasses nations where both non-communicable disease and communicable disease cause considerable ill-health.”

Prof Michael Lee, Head of Department and Site, Sustainable Agriculture Sciences, Rothamsted Research, and Chair in Sustainable Livestock Systems, University of Bristol, said:

“The authors’ attention to uncertainty is commendable given the number of methodological limitations surrounding diet-level environmental impacts; however, although data of environmental footprints are drawn from national averages, they do not represent the complexity of food production or availability. Agricultural and dietary improvement interventions are delivered at the regional-local level, which is where policy needs to be focused and not at international or national levels, which is the focus of the current paper. There is a missed opportunity to greatly enhance our understanding of sustainable food systems (via recommended dietary guidelines) as the study reuses already-published environmental databases without any novel consideration of critical sources of uncertainty such as, for instance, individual nutrients’ digestibility and bioavailability. This is a major limitation of the EAT-Lancet report, which failed to consider nutritional value and therefore if consumed meticulously in perpetuity, as proposed without supplements, would result in nutritional deficiency especially for key micronutrients, which is a major global health concern, especially within LMIC. This hidden hunger (micro-nutrient deficiency) risk was not considered or mentioned in the current study despite the considerable section identifying limitations of the approach, nor did it reflect the Food-Based Dietary Guidelines’ response to increased milk consumption to improve global health, which was at loggerheads with the EAT-Lancet recommendations, which was co-authored by the lead author.

“Furthermore, unlike WHO and EAT-Lancet recommendations that are based on mathematically desirable diets, national dietary guidelines must also consider practical constraints facing both food producers and consumers to ensure that the information is socially relevant. This includes, but is not limited to, producers’ livelihoods, consumers’ affordability, national food security, and ease to understand by the general public. It is therefore unsurprising that WHO and EAT-Lancet recommendations perform better according to the authors’ purely nutritional and environmental criteria, whereas a more pertinent question may be how we can shift to healthier diets without compromising other, and equally important, goals of society. Making national dietary guidelines mathematically more accurate without recognising these real-world constraints, in my view, does not solve the very core of this complex and multifaceted problem.

“Finally the study opts not to join global discussions on cutting edge issues in the field of agri-food systems-analysis; for example, there is no acknowledgement of recent debates on the short atmospheric lifecycle of methane compared to nitrous oxide, which question the global reliability of commonly used approaches for quantifying the sustainability of food supply-chains. It would appear that these potential weaknesses were not acknowledged by the authors and it goes against their current stance of advocating the EAT-Lancet findings, which may not be totally balanced when comparing food groups or their environmental credentials.”

Prof Janet Cade, Professor of Nutritional Epidemiology and Public Health, University of Leeds, said:

“This is an important paper, aiming to link Food Based Dietary Guidelines (FBDG) with health and environmental impact. Key findings highlighted that if countries adopted FBDG they could reduce premature mortality by 15% (95%CI 13% to 16%) and reduce greenhouse gas emissions by 13% (regional range -34% to 35%). A lot of care had gone in to converting country specific FBDG key messages into quantitative values. A broad set of health and environmental indicators were used.

“Unfortunately, and at the moment inevitably, this high level analysis has to use proxies for both dietary consumption and healthiness. We need better, more standardised approaches to measuring food intake at a national level. Many countries do have National Diet Surveys, however, they all use different methods making them hard to combine. The Food Balance Sheets from the FAO used in this study, however, do not actually measure intake, but rather food availability in a country.

“Interestingly, within the national FBDG, the least clear guidance was provided for processed meat. This is unexpected, since the majority of systematic reviews on the topic suggest higher risk of non-communicable disease with higher processed meat. Improving the quality and clarity of FBDG will help direct public policy to improve health and the environment. Most countries seem a long way from having guidelines which are adequate in these respects; and even further away in terms of population dietary intakes. How to achieve this change in the population is the next critical question to be answered.

“The impact of increasing whole grains on premature mortality was as large as reducing prevalence of obesity in many regions (fig. 4), highlighting the need to consider the composition of our dietary patterns, including fibre intake. The analysis undertaken here has a potential for over-adjustment, since BMI is likely to be on the causal pathway as opposed to an independent risk factor. Without BMI in the model, effects of food groups would be larger.

“Overall, a strong analysis doing the best they could with the data available. Data collection to help answer these questions needs to adapt and draw from new technologies to move the international evidence base further forwards.”

Dr Duane Mellor, Registered Dietitian and Senior Teaching Fellow, Aston Medical School, Aston University, said:

“This is an interesting paper which compares the food based guidelines for the population in a country to food available in that country for a number of countries across the world. What it does not do is look at what people are eating and how many people are eating in a way that matches the guidelines or could be seen to be sustainable.

“Perhaps this is most useful at this level, to support governments to shape their food systems and so the food available to the public so that health and environmental goals are easier for the average consumer to achieve. What is does not do but suggests in its discussion is measure current diets, which it claims from the data presented are neither healthy nor environmentally sustainable, this can only be said about the food supply at a national level. This also assumes that a global approach to a food system issue is plausible, based on variations in agricultural practice, climate and cultural practices linked to food, this assumption promoted by the EAT-Lancet model used in this paper appears at best ignorant of the diversity of traditional food behaviours globally, many of which are associated with long and healthy diets, such as those in parts of the Mediterranean and Okinawa in Japan.

“With respect to the diet of people in the UK, we are a country which has a diverse multicultural food cultures, which could be improved by increasing intakes of vegetables, pulses, fruit, nuts, seeds etc. as this paper suggests, but perhaps we need to support communities in finding healthier ways of eating they can enjoy, are affordable and also sustainable. Work like this is important for influencing food supply and systems at a national level, but it needs to be linked to working with communities to improve health and reduce food insecurity in a sustainable manner.”

‘The healthiness and sustainability of national and global food based dietary guidelines: modelling study’ by Marco Springmann et al. was be published in the BMJ at 23:30 UK time on Wednesday 15 July 2020.

DOI: 10.1136 bmj.m2322

Declared interests

Prof Francesco Cappuccio: “Head of WHO Collaborating Centre for Nutrition (UNK-257) at the University of Warwick, unpaid.”

Prof Elizabeth Robinson: “She was specialist advisor to the House of Lords Select Committee on Food, Poverty, Health and the Environment report, titled “Hungry for change: fixing the failures”.  She is co-author of the annual Lancet Countdown report.”

Prof Alan Dangour: “I am receipt of competitive funding from the Wellcome Trust, DFID, NIHR, GCRF and others to conduct research on the links between the environment, food systems and health.  I am the Director of the LSHTM Centre on Climate Change and Planetary Health.”

Dr Sarah Berry: “I receive funding and am paid consultancy from ZOE Global Ltd.”

Prof Michael Lee: “We do not have any conflict of interests, Rothamsted Research is an independent research organisation which receives the vast majority of its research income from public funding.”

Prof Janet Cade: “My research group has been undertaking similar work exploring impact of dietary intake on GHGE.  I am a Director of Dietary Assessment Ltd.”

Dr Duane Mellor: “None.”

None others received.