Read time: 20 minutes (3300 words)
What to eat?
That used to be a question that only needed to be asked on the rare occasions when you ate out. Traditionally, eating had always been a cultural thing. People learned what to eat at the family table, guided by seasonally available foods most of which were produced locally or nationally. Each country had its traditional foods, everything seemed to be simple.
But at some point in the last century, partly as a result of rising cardiovascular disease, but perhaps also because of the increasing commercialisation of the food chain, governments across the developed world took it upon themselves to start advising their populations how best to eat. In the USA, the first Dietary Guidelines for Americans was published in 1980 and it has been updated every five years ever since. In the UK similar recommendations began in 1994.
By 2010 more than a quarter of a century of government programmes should have produced some measurable results. So what did the data show? Here are some graphs from the USA:
Things were looking pretty bad. But what was behind the worsening national health? To find out a report was commissioned “The State of US Health, 1990-2010: burden of diseases, injuries, and risk factors” because…
“Understanding the major health problems in the United States and how they are changing over time is critical for informing national health policy.
I couldn’t agree more. So, what did they find?
Incredibly, dietary risk factors were identified as the biggest contributor to premature deaths, outstripping smoking and obesity. Diet also came out as the leading contributor to the number of years spent living with chronic health problems / disability (graph not shown)
With such clear evidence of the importance of diet, and the failure of 30 years of the standard advice, most governments have dug in their heals, doubled down and simply shouted their message louder, much to the dismay of nutritional researchers many of whom were starting to believe that the dietary guidelines might actually be contributing to the rising tide of chronic disease they were designed to address.
So what is it that these researchers have discovered? What kind of changes do they think we need to make to national health policies? And perhaps, more importantly for readers of this blog, how can we use their ideas in our own lives? Lets take a look:
The importance of overall dietary patterns
Nutritional science is shifting away from a focus on single nutrients (fat, carbs, specific vitamins etc) to looking at whole foods and overall dietary patterns. Fats are a good example: Once demonised as obesogenic due to their inate energy density, it is now realised that their metabolic effects depend on source; extra-virgin olive oil, coconut oil, fish oils and dairy fats – can have very different effects in the body, and not always for obvious reasons.
Likewise, the simplistic calorie counting paradigm for obesity is being challenged as foods are shown to have complex effects on long term weight change, via hunger, satiety, reward, metabolic responses, liver fat synthesis, metabolic expenditure and the microbiome. [Mozaffarian, 2017]
It is now understood that foods, rather than simply their fat content, have important effects on multiple aspects of cardiovascular health such as blood pressure (BP), glucose-insulin homeostasis, lipoprotein concentrations and function, oxidative stress, inflammation, endothelial health, hepatic function, adipocyte metabolism and cardiac function. A narrow focus on LDL cholesterol fails to capture these nuances. (see our post on cardiovascular risk here)
Looking more widely, foods are rarely eaten in isolation, leading researchers to look at groups of foods that are commonly eaten together. A recent example from Australia ‘Associations between Dietary Patterns and Blood Pressure in a Clinical Sample of Overweight Adults’ [Ndanuko, 2017]. Among their participants they identified six major dietary patterns of foods commonly eaten together: “nuts, seeds, fruit, and fish,” “milk and meat,” “breads, cereals, and snacks,” “cereal-based products, fats, and oils,” “alcohol, eggs, and legumes,” and “savoury sauces, condiments, and meat.” The main measured outcomes were resting blood pressure measured over a 12 month period. They found that a dietary pattern rich in nuts, seeds, fruit, and fish was inversely associated with blood pressure.
That’s just one example of trying to approach nutrition more holistically although how useful this particular approach will be in the long run remains to be seen. More interesting, perhaps, are the underlying principles that are emerging in nutritional science. These are the ideas that need to be embodied in national policy if it is to do better than it has in the recent past…
1. Not all calories are equal
The calorie-centric approach to food policy led to some extraordinarily bad public health decisions. The 2012 US National School Lunch Programs, for example, bans whole milk, but allows sugar-sweetened chocolate skimmed milk, as it has fewer calories, fat and saturated fat. This ‘nutrient demonising’ approach ignored the wide range of epidemiological evidence that identified no long term harms associated with full fat milk consumption.
Longitudinal studies suggest no harms of whole-fat milk for obesity, diabetes mellitus, or cardiovascular disease in adults28,29,51,57,58; that dairy fat may have potential benefits for diabetes mellitus59–61; that people switching to low-fat dairy products compensate elsewhere in their diet by increasing consumption of carbohydrates29; and that children who habitually drink low-fat milk gain more weight, and those who drink whole-fat milk gain less weight, over time. – Mozaffarian, Circulation (2016)
The power of foods compared to single nutrients was demonstrated in the PREDIMED trial where including modest amounts of nuts or olive oil in the diet was more effective at reducing heart disease than following an intensive low fat regimen. Under the single nutrient mindset nuts and olive oil have been considered unhealthy because of their high fat and calorie content, but it is now appreciated that accompanying nutrients in the whole food modify the effects that might be expected from considering just the basic macronutrient content.
2. real foods
Olive oil provides a good study for our changing understanding of whole foods. Early studies put olive oil’s heart protective effects down to its high content of monounsaturated fats, however recent meta-analyses have concluded that there is little evidence to suggest that monounsaturated fatty acids are any better for the heart than saturated fats [Kuipers, 2011]. So there must be something else in olive oil. One clue is that almost all of its benefits are enhanced when extra virgin olive oil is used instead of the more refined versions, suggesting that it is the phytonutrients present in the raw oil that are beneficial. Recent studies have found that the polyphenols in olive oil have an antioxidant effect on LDL lipoproteins whilst improving the functioning of HDL cholesterol [Berrougui, 2015]
In the case of nuts, it appears that they somehow reduce appetite proportionally to their consumption, so snacking on them ad libitum does not usually lead to weight gain.
Another example is cheese, which would be expected to raise LDL cholesterol based on its fat content but does not, yet an equal amount of dairy fat consumed as butter does. The food matrix seems all important and requires a more nuanced approach than simple nutrient profiling can achieve.
3. Small changes in the right direction
What is also remarkable is that quite small shifts in the right direction can have significant effects: just 30g per day of mixed nuts led to a 30% reduction in heart disease in the PREDIMED trial with few other changes in the habitual diet.
Similarly, eating just two portions of oily fish per week (salmon, sardine, mackeral) have been shown to improve pregnancy and infant outcomes, depression and stroke incidence. You may have seen headlines a while back, claiming ‘eating just two portions of oily fish per week can protect you from a junk food diet‘. (Actually the research behind this demonstrated the ability of omega 3 PUFAs to reverse aspects of immune disfunction associated with obesity in mice [Ndanuko, 2017]- but that didn’t make such a catchy headline!)
The above graph shows a study of the association between fish consumption and stroke risk in the elderly (Archives of internal medicine, 2006). It is notable that including a relatively modest amount of fish in the diet – eating it weekly instead of monthly – is associated with a significant reduction in stroke incidence.
This should not be surprising as the average intake of omega-3 fatty acids has decreased to less than 20 % of what was present in common diets 150 years ago. Less than 5% of the population gets the amounts required for good health, which according to some researchers, makes these fats the most therapeutic of all the essential nutrients [Kaur, 2014]
4. More of a good thing is not necessarily better
There is an understandable tendency to think that if something is good for you then more of it must be better, but this is frequently not true. In many cases the major benefits come with the initial addition of a food to the diet, whereas further increases lead to reducing returns. This was evident in the stroke study above, where there was essentially no additional benefit eating fish more often than one to four times per week.
As we showed in our post on coffee consumption, increasing benefits are seen up to five cups per day, but the greatest gain is shifting from fewer than one cup per day to two or three. Similarly, eating larger quantities of nuts does not appear to bring significant additional benefits above 30g per day.
In the case of public health policy when presumed benefits do not materialise (as has been the case with the last thirty years of low-fat recommendations), there has been a move to double down. The latest revision of the UK Government’s healthy eating guidelines in 2016 further marginalised dietary fat, whilst recommendations to eat more fruit and vegetables has escalated from a recommended five a day to seven, and most recently ten portions per day. However, recent large scale observational studies such as PURE (video here) have found that benefit plateaus above 3 a day: a much more realistic target, especially for poorer households, as fruit and veg can be particularly expensive ‘luxuries’.
5. Balancing risks and benefits
Official advice on fish consumption has undergone a volta-face in the last decade. Initial concerns about mercury intake lead the WHO to advise pregnant women to avoid fish consumption. However it soon became apparent that women that ate more fish had children with better brain function and the advice was reversed. Pregnant women are now encouraged to eat two portions of oily fish per week as the benefits to them and their offspring outweigh the risk posed by mercury. Indeed in fish eating populations, like the Spanish, high levels of toenail mercury concentration are good markers of cardiovascular protection! [Downer, 2017]
In the case of red meat, epidemiological associations with heart disease and cancer are common, and have led to regular advice to reduce intake, but the picture is unexpectedly complicated with some major studies finding little effect – such as the UK EPIC study – see our post UK vegetarians DON’T live longer than meat eaters – or even a protective effect such as the recent study looking at cardiovascular disease across 42 EU countries (look out for our post on this soon).
6. Food displacement
One of the problems with dietary recommendations is that when you add a new food or eat more of one kind of food, you inevitably eat less of another, and vice versa. This makes nutritional science particularly difficult as the basic tenet of only changing one variable at a time is almost impossible to achieve. Consequently, many nutritional conclusions should be approached with more caution than newspaper headlines would have you think.
In public health policy we saw food displacement play out in a big way with the advice to reduce intakes of fat. This led to the public switching a proportion of their consumption from fats to carbohydrates, which almost certainly contributed to the current obesity epidemic.
The health consequences of reducing saturated fat in the diet depends entirely on which food replaces it. Recent evidence from randomised controlled trials and epidemiological data indicates that replacing saturated fat with refined starches or sugar worsens cardiovascular outcomes, whereas replacing them with omega-3, (but not omega 6 PUFAs), improves such outcomes [Kuipers, 2011]. It would appear that the effect of advising the public to eat less saturated fat led to them eating more carbohydrates instead. A positive message to eat more nuts and fish would probably have produced the desired outcome.
In the case of hot drinks, coffee, tea and coco have all been shown to have health benefits but there are only so many hot drinks one can get through in a day! The phytonutrients in each of these beverages exert their own range of beneficial effects so there is an argument for trying to include a range of them in your diet. In fact a recent study looking at liver health found that drinking just one cup of any herb tea each day was as beneficial as drinking multiple cups of coffee.
7. Diet is more than the sum of the parts
Foods are rarely eaten in isolation, but are combined and varied over the short and long term in overall dietary patterns. Eating one food in the context of another can alter how it is digested, absorbed and its final metabolic effects. A simple example of this is using salad dressings: consuming oil with salad increases the absorption of beta carotenes and other fat-soluble vitamins.
Habitual food associations can also affect the apparent health credentials of a food, particularly in observational studies where such associations can be hard to tease apart. For example, a recently published Harvard study found that fat from cheese and butter were associated with an increased risk of diabetes, but fat from full fat yogurt was associated with a decreased risk [Guasch-Ferré, 2017]. Why would this be? The authors make several points: cheeses and butter are habitually eaten with carbohydrates like bread and crackers, but yoghurt is not, so it may be that cheese and butter are damned by the company they keep. Also, people that eat yoghurt tend to have healthier lifestyles, including smoking less. Finally, there are functional differences – yoghurt often contains live bacteria which may modify the microbiome in ways cheese and butter do not.
Confounders like these make conclusions around meat and especially red meat, particularly problematic.
8. Dietary Diversity
From what has been said above, it might appear that simply eating a wide variety of foods from across the major food groups would be a prudent option. Indeed this is often what is meant by that most vague of terms ‘a balanced diet’. Diet Diversity Scores (DDS) are used to assess the adequacy of diets in certain studies [see UN Standing Committee on Nutrition report] and health policies around the world often focus on dietary diversity. For example, Japan advises consumption of 30 different food items per day whilst the US advocates consumption of a variety of nutrient-dense foods and beverages within and among 5 basic food groups, with an item from each food group consumed daily [the 5 USDA food groups are: cereals, vegetables, fruit, dairy, and protein source foods (meat, fish, poultry, eggs, nuts, beans)]
In developing countries where malnutrition is common, DDS is a good indicator of nutritional adequacy. When diets are based on a monotonous staple crop almost any diversification brings improved nutritional benefits [e.g. Kennedy, 2007]. In developed economies a high DDS can also reflect a healthier diet, but in many cases it is associated with excessive energy intake and obesity [e.g. Zhang, 2017] which leads researchers to argue that public health messages should give less emphasis to dietary diversity.
In short increased dietary diversity is a necessary starting point, but as diets get more diverse the emphasis needs to become more nuanced and focus on the foods that will really make a difference. On this point a particular bug bear of mine is the ubiquity of wheat based products in the typical western diet. Wheat accounts for close to one quarter of the British calorie intake, mostly in the refined, white form – this would be my first choice for swapping out for more nutrient dense foods such as eggs, nuts, veg and seafood.
- See our post The balanced diet argument – Just another straw man
With regards to carbohydrate-rich foods more widely, they now represent more than 50% of the world’s calories. Evidence for cardiometabolic harm falls chiefly on refined products leading researchers to conclude
Based on their adverse effects and pervasiveness in modern diets, reducing refined grains, starches, and added sugars is a major dietary priority for cardiometabolic health
9. Sustainability and pleasure
An important part of any national food policy recommendation has to be not only that the advice is sound, but that it is do-able, something that people can enjoy and make a way of life. Low fat diets have not only proved to be ineffective, but are notoriously dull and hard to stick to. A good diet should be a pleasurable, sociable, adaptable, satisfying and delicious.
Building a diet around fish, meat, eggs, seasonal fruit, vegetables, nuts, seafoods and full fat dairy, with tea, coffee, moderate wine and dark chocolate is eminently doable and backed by a ton of research.
Getting there at a national level will require action on multiple fronts, including the personal, social, agricultural, industrial, governmental and global.
Campaigns and organisations such as The Weston A Price Foundation, Pasture for Life, The Oxford Real Farming Conference and Jamie Oliver’s push for more home cooking are all part of this bigger picture.
10. Traditional Diets
We began this article looking at how diets had once been culturally driven, with people learning what to eat at the family table, guided by seasonally available local foods. Is there a case for a return to such traditional eating? I think so.
There is now considerable evidence that pre-industrial or traditional diets did not promote the chronic degenerative diseases that only became prevalent in the late 20th century. Furthermore, clear health issues are associated with overconsumption of industrialised foods such as refined grains, sugar and hydrogenated and industrial vegetable oils – which are often found together in popular convenience foods. More traditional, home cooked foods starting with raw ingredients tend to be higher in polyphenols and bitter compounds which have well documented health benefits. Amongst the dietary patterns that most closely advocate for a pre-industrial diet are the Paleo and Mediterranean diets.
A recent study in the Journal of Nutrition [Whalen, 2017] investigated associations of these two diet pattern scores, with all-cause and cause-specific mortality from the REGARDS (REasons for Geographic and Racial Differences in Stroke) study. This was a longitudinal cohort of black and white men and women ≥45 y of age who were following their habitual diets.
In short, participants whose diets more closely resembled paleolithic or mediterranean eating patterns had significantly reduced mortality over the 6.5 years of the study. Similar benefits have been enumerated in other studies adding to a growing body of evidence that eschewing industrial foods in favour of traditional ones confers significant health benefits.
What do these healthy dietary patterns look like?
The Mediterranean diet pattern is summarised well by Lopez-Garcia et al [American Journal of Nutrition, 2013]:
[The Mediterranean diet pattern is characterised by…] high consumption of fruit and vegetables, a substantial intake of proteins from plant sources (legumes and nuts), and a high fat intake, mostly from MUFAs. There is also a moderate to relatively high fish consumption and, in contrast, a low consumption of meat and meat products. Alcohol intake is moderate, usually as red wine consumed with meals. In addition, olive oil* is the predominant fat for cooking and dressing salads, and sautéing and stir frying are the cooking techniques characteristic of this diet. Finally, because this is a palatable dietary pattern, adherence to a Mediterranean-style diet can be easily achieved by the general population in and outside the Mediterranean basin
(*It should be noted that the benefits of olive oil are primarily associated with the unrefined Extra Virgin variety)
The paleolithic diet pattern is summarised here by Kristine Whalen [American Journal of Epidemiology 2014].
The Paleolithic diet pattern is characterized by a wide diversity of fruits and vegetables, lean meats, eggs, and nuts; it excludes grains, dairy products, refined fats, and sugar, and is very low in salt.
In conclusion, it appears to me that the public health message needs to move towards promoting such dietary patterns if we want to see major improvements in public health. But don’t hold your breath. Despite many nutritional researchers pushing for such reform the pace of change is glacial and seems dominated by the needs to keep the industrial food manufacturers on board. In the meantime, family meals, like the traditional Sunday roast here in the UK, are increasingly becoming a thing of the past.
In the meantime, there is nothing to stop each of us from implementing these changes for ourselves and our families. For ideas on how you can start to improve your diet and enjoy your food more, check out our nutrition posts and recipe pages.
Jointly written by Afifah Hamilton and Keir Watson