Junk food nation

Eating more junk foods such as soft drinks, packaged snacks and sugary cereals is associated with a higher risk of over 30 different physical and mental health problems according to an umbrella review by Australian and international researchers.

The researchers combined the results of 45 previous meta-analyses on the topic, and while this kind of study cannot prove the junk food is causing the health problems, they say there was consistent evidence that ultra-processed foods (junk foods) are associated with death of any cause and specific conditions including type 2 diabetes, obesity, anxiety, sleep problems and wheezing.

The findings, published by The BMJ, show that diets high in ultra-processed food may be harmful to many body systems and underscore the need for urgent measures that target and aim to reduce dietary exposure to these products and better understand the mechanisms linking them to poor health.

Ultra-processed foods, including packaged baked goods and snacks, fizzy drinks, sugary cereals, and ready-to-eat or heat products, undergo multiple industrial processes and often contain colours, emulsifiers, flavours, and other additives. These products also tend to be high in added sugar, fat, and/or salt, but are low in vitamins and fibre.

They can account for up to 58% of total daily energy intake in some high income countries, and have rapidly increased in many low and middle income nations in recent decades.

Many previous studies and meta-analyses have linked highly processed food to poor health, but no comprehensive review has yet provided a broad assessment of the evidence in this area.

To bridge this gap, researchers carried out an umbrella review (a high-level evidence summary) of 45 distinct pooled meta-analyses from 14 review articles associating ultra-processed foods with adverse health outcomes.

The review articles were all published in the past three years and involved almost 10 million participants. None were funded by companies involved in the production of ultra-processed foods.

Estimates of exposure to ultra-processed foods were obtained from a combination of food frequency questionnaires, 24 hour dietary recalls, and dietary history and were measured as higher versus lower consumption, additional servings per day, or a 10% increment.

The researchers graded the evidence as convincing, highly suggestive, suggestive, weak, or no evidence. They also assessed the quality of evidence as high, moderate, low, or very low.

Overall, the results show that higher exposure to ultra-processed foods was consistently associated with an increased risk of 32 adverse health outcomes.

Convincing evidence showed that higher ultra-processed food intake was associated with around a 50% increased risk of cardiovascular disease related death, a 48-53% higher risk of anxiety and common mental disorders, and a 12% greater risk of type 2 diabetes.

Highly suggestive evidence also indicated that higher ultra-processed food intake was associated with a 21% greater risk of death from any cause, a 40-66% increased risk of heart disease related death, obesity, type 2 diabetes, and sleep problems, and a 22% increased risk of depression.

Evidence for the associations of ultra-processed food exposure with asthma, gastrointestinal health, some cancers and cardiometabolic risk factors, such as high blood fats and low levels of ‘good’ cholesterol, remains limited.

The researchers acknowledge that umbrella reviews can only provide high-level overviews and they can’t rule out the possibility that other unmeasured factors and variations in assessing ultra-processed food intake may have influenced their results.

However, their use of rigorous and prespecified systematic methods to evaluate the credibility and quality of the analyses suggests that the results withstand scrutiny.

As such, they conclude: “These findings support urgent mechanistic research and public health actions that seek to target and minimise ultra-processed food consumption for improved population health.”

Ultra-processed foods damage health and shorten life, say researchers in a linked editorial. So what can be done to control and reduce their production and consumption, which is rising worldwide?

They point out that reformulation does not eliminate harm, and profitability discourages manufacturers from switching to make nutritious foods, so public policies and action on ultra-processed foods are essential.

These include front-of-pack labels, restricting advertising and prohibiting sales in or near schools and hospitals, and fiscal and other measures that make unprocessed or minimally processed foods and freshly prepared meals as accessible and available as, and cheaper than, ultra-processed foods.

It is now time for United Nations agencies, with member states, to develop and implement a framework convention on ultra-processed foods similar to the framework on tobacco, and promote examples of best practice, they write.

Finally, they say multidisciplinary investigations “are needed to identify the most effective ways to control and reduce ultra-processing and to quantify and track the cost-benefits and other effects of all such policies and actions on human health and welfare, society, culture, employment, and the environment.”

Expert Reaction

Dr Charlotte Gupta, a postdoctoral research fellow at the Appleton Institute, Central Queensland University, said “while foods like ice-cream, chips, lollies, fast food, soft drinks, chocolate, ready-made meals, and biscuits are delicious, new research has identified just how damaging they can be to our long-term health.

These foods are termed ultra-processed foods, which are products made up of foods that have undergone significant processing and no longer resemble the raw ingredients. It is common knowledge that these foods aren’t considered ‘healthy’, but this new research has shed light on how damaging they can actually be.

By reviewing data from 14 reviews with over 10 million participants, researchers found strong evidence that greater exposure to ultra-processed foods in our diet increases the risk of cardiometabolic health problems, mental health disorders, and early death. Overall this research provides convincing evidence that ultra-processed foods can lead to serious health problems and therefore public campaigns should focus on education about reducing intake.

However, for some people, such as shiftworkers working at night (e.g., nurses, doctors, firefighters, taxi drivers, miners, hospitality workers), there is a lack of availability of fresh foods or time to prepare any food, and so ultra-processed foods have to be relied on (e.g. from the vending machine in the hospital). This highlights the need for not only individuals to try reducing ultra-processed foods in our diet, but also for public health actions to improve access to healthier foods.”

Dr Alan Barclay, an Honorary Associate at the University of Sydney, said  this “latest umbrella review of the evidence from observational studies (cohort, case-control and cross-sectional designs) linking consumption of ultra-processed foods (as defined by the NOVA food classification system) to adverse health conditions has just been published. Like its predecessors, it has unsurprisingly found positive associations between high consumption of foods categorised as ultra-processed with cancer, mental, respiratory, cardiovascular, gastrointestinal, and metabolic health conditions. In other words, the more ultra-processed foods consumed, they higher the risk of developing these conditions in these observational studies. The authors rated the evidence as mostly very low to low quality.

While these associations are interesting and warrant further high quality research, they do not and cannot provide evidence of causality.

Many experts in food and nutrition consider the NOVA system to be a simple and crude system for classifying foods into categories on the basis of their degree of processing and ingredients, in stark contrast to many existing food-classification systems based on the food’s nutrient content.

While efforts have been made in recent years to improve the ability of health professionals and researchers to classify foods using the NOVA system, many of the included studies were based on older definitions and methodologies. Furthermore, very few of the studies included in the umbrella review were specifically designed to assess the ultra-processed food content of their populations respective diets, and consequently, validation data are scarce.

By their very nature, observational studies are renowned for being confounded by numerous factors – both known and unknown. A recent European study, for example found an association with ultra-processed foods and head, neck cancer and oesophageal adenocarcinoma. However, they also included a negative control in their analysis and found that ultra-processed food consumption was also positively associated with accidental death (e.g. falls, transport accidents, accidental drowning).

There is of course no reason why high ultra-processed food consumption would cause accidents. Wisely, the authors reasoned that ‘Any evidence of an association between UPF consumption and accidental deaths would suggest that our main results may be biased by the same factors that biased the negative control outcome results.’ Other researchers need to take note.

Prospective cohort studies specifically designed and validated to use the NOVA system need to be conducted for rare conditions and when feasible and ethical, randomised controlled trials specifically designed to use the NOVA system need to be conducted. Until then, calls to incorporate the ultra-processed food concept into dietary and other public health guidelines are premature and may have unexpected social and health consequences.”

Dr Evangeline Mantzioris, the Program Director of Nutrition and Food sciences at the University of South Australia, said this “recent umbrella review, which provides us with high level evidence, has looked at the impact of ultra-processed foods (UPF) on health. This umbrella review combined the results from 14 different meta-analyses, which includes just under 10 million people.

UPF include packaged baked goods and snacks, soft drinks, sugary cereals, and ready-to-eat or heat foods. UPF undergo multiple industrial processes, often leading to a destruction of the food matrix. They also often contain colours, emulsifiers, flavours, and other additives. UPF also tend to be high in added sugar, saturated and/or trans fats, and/or salt, but are low in anti-oxidants, phytonutrients, vitamins, minerals and fibre. Given that in Australia, around 50% of our dietary energy may come from UPF, these results are of great interest and significance to us all.

From these analyses there is convincing evidence that a higher UPF intake was associated with about a 50% increased risk of a heart disease related death, a 48-53% higher risk of anxiety and common mental disorders, and a 12% greater risk of type 2 diabetes.

This review also found highly suggestive evidence that higher UPF intake was associated with a 21% greater risk of death from any cause, a 40-66% increased risk of heart disease related death, obesity, type 2 diabetes, and sleep problems, and a 22% increased risk of depression.

Additionally, greater intakes of UPF were related to further higher risks of these adverse health outcomes.

This research highlights the importance of people limiting their intake of UPF and replacing it with whole foods.”

Dr Daisy Coyle, a Research Fellow and Accredited Practising Dietitian at The George Institute for Global Health, said the “Recent research, published in the BMJ, highlights a troubling reality: ultra-processed foods pose a threat to human health with intakes linked to more than 30 health outcomes.

The statistics are staggering – these foods may double your risk of dying from heart disease or from developing a mental health disorder.

Ultra-processed foods, laden with additives and sometimes lacking in essential nutrients, have become ubiquitous in the Australian diet. In fact, they make up almost half of what we buy at the supermarket. While not all ultra-processed foods are linked to poor health outcomes, many are, particularly sugary drinks and processed meats.

Despite mounting evidence of their harmful effects, effective measures to curb our consumption of these products have been lacking. Existing nutrition policies in Australia aren’t enough to tackle this problem. It’s important that we start developing solutions that target the most harmful ultra-processed foods.

We need to consider both policies that pull consumers away from ultra-processed foods, like advertising restrictions, warning labels and bans in schools as well as policies that support consumers toward healthier alternatives such as through subsidies for fresh, nutritious foods.”

Melanie McGrice, an Advanced Accredited Practising Dietitian with a Masters degree in nutrition and founding member of the Early Life Nutrition Coalition, said “In my role as a dietitian, I see so many people reliant on ultra-processed foods. Some of the key factors impacting consumption of ultra-processed foods include perceived convenience, social influences, emotional eating and advertising. I’ve seen extraordinary results for individuals who I have worked with as a result of decreasing intakes of ultra-processed foods. Given the difference that it makes, it is an obvious area for intervention to help improve the health and vitality of Australians, especially in young children.”

Helen Truby, a Professorial Research Fellow at the University of Queensland, observed the “latest data from the Australian Bureau of Statistics shows that 1 in 20 Australians have diabetes and 81% of Australian adults have at least one health condition, with mental health conditions including anxiety being common. The food we eat has long been known to be important in preventing some of the long-term conditions that the majority of Australians experience.

This review concentrates only on the relationship between consumption of ‘ultra-processed foods’ and a range of health outcomes. The strength of this review is the consistent direction of the relationships between consumption of ultra-processed foods and poorer health outcomes including mental health. Indeed, for Type 2 diabetes there was a ‘dose-response’ which means that the more ultra-processed food that was eaten the higher the risk of diabetes.

However, the evidence is not always strong and relationships do not prove causality. The authors call for much-needed research to provide direct cause-and-effect evidence, for example, can symptoms of anxiety be reduced or improved by swapping ultra-processed foods with unprocessed food? The volume of evidence in this review would suggest that it would do no harm to swap out the pop tart for a piece of whole grain toast. ”

Clare Collins, a Laureate Professor in Nutrition and Dietetics at the University of Newcastle and Co-Director of the Food and Nutrition Research Program at the Hunter Medical Research Institute, said  the “results published today from an Umbrella Review of previous systematic reviews provide evidence that it is time for government to take stronger action regarding the availability of ultra-processed foods (UPF). There has been a steady increase in UPF intake around the world which parallels the rise in diet-related burden of disease.

The paper published in The BMJ today, utilised data from 45 separate pooled analyses examining associations between UPF intakes and health outcomes. They presented a summary figure of the evidence grading and credibility of that evidence. This highlights the strongest dose-response relationship, rated as convincing/moderate quality, was for high intakes of UPF and type 2 diabetes, followed by combined mental health outcomes and anxiety outcomes, which were rated as convincing/low quality. There were many concerning findings, particularly for all-cause mortality which was rated as suggestive/moderate quality. These findings cannot be ignored.

The studies are observational, which means cause and effect cannot be proven and that the research evidence gets downgraded, compared to intervention studies. The problem is that it is not ethical to do an intervention study lasting for many years where you feed people lots of UPF every day and wait for them to get sick and die.

This analysis which includes almost 10 million people should be a call to governments around the world for a moratorium on UPF. However, it should not be governments (and that means we taxpayers) who should be funding research on the adverse health effects from UPF, rather the UPF food industry creating the foods should be paying for this research. Government resources should be directed to strategies that implement what we already know about the protective effects of consuming a healthy diet that is minimally processed and ensuring those who experience food insecurity are given every opportunity to be food secure and have dietary patterns that align with their county’s national food and nutrition recommendations.”

Professor Mark L Wahlqvist, an Emeritus Professor, Head of Medicine at Monash University and Monash Medical Centre and Past President of the International Union of Nutritional Sciences notes “The publication by the BMJ of a compilation by several Australian and international investigators of the now extensive database of so-called ultra-processed food (UFP) intakes and their apparent adverse health outcomes is a landmark flexion point for nutritional and health science and policy, duly editorialised by Monteiro, Martínez-Steele, and Cannon.

It is at least relevant to ‘all-cause mortality, cardiovascular disease-related mortality, common mental disorder outcomes, overweight and obesity, and type 2 diabetes’. It comes, however, with caveats in the definition and public interpretation, arbitrariness, and acceptance of the UFP terminology and science. This is especially so insofar as cross-cultural, socio-demographic and food system considerations are taken into account.

Why do we find ourselves in this dilemma of an inability to trust our own food choices as ones for nutritious foods without resorting to yet another categorisation? It is not only that the UFP foods are discredited for their healthiness on account of their ingredients and manufactural methodology. It is because we are increasingly distant from the point of origin, handling, transport, storage, packaging (generating nanoplastic endocrine disruptors), retailing, preparation cooking and consumption environs of our food. Its real benefits, risks and costs are less available with consequences for not only health, but our household economics, and sustainable livelihoods.

The packaging itself, not just the UPF, and the measurements of food composition and nutritional value do not, for example, even include those of food structure, let alone its enormous chemical complexity. The answer to this is not to be perplexed by the negative assessment of food and food systems, but to have the facility to choose it for its positive value.

As long as dietary guidelines have been formulated, they have asserted the simple and overarching principle to have a biodiverse dietary pattern and to be sufficiently active to eat enough without adversely affecting body composition, brain, blood, bone, muscle and fat! This understanding can be taught throughout the educational system, and reinforced in public policy.”