The Five Food Groups concept has been a cornerstone of nutrition education for decades. But are they based on good science and logic? Do the Five Food Groups make sense?
Food groups and ‘distinguishing nutrients’
The concept of food groups has been central to nutrition education in many countries for decades. It’s a simple idea: different types of foods make distinctive contributions to the diet so eating a variety of foods from the different food groups should ensure adequate intakes of essential nutrients.
Although all whole foods contain small amounts of a wide variety of nutrients, the distinctive contribution of a food group is due to 4-6 ‘distinguishing nutrients’ found in abundance. As foods in the same food group share distinguishing nutrients they can be exchanged without compromising nutrient intake. This nutritional equivalence allows flexibility in food choice.
Vegetables and fruits: two food groups or one?
In Australia, fruits and vegetables appear in two separate food groups but the rationale for two food groups is not at all clear. The dietary modelling conducted for the latest Australian Guide to Healthy Eating highlights the distinguishing nutrients of fruits and vegetables:
Vegetables: Vitamin C, fibre, folate, vitamin B6, potassium, beta-carotene
Fruit: Vitamin C, fibre, folate, vitamin B6, potassium.
Aren’t the distinguishing nutrients for both food groups almost exactly the same? Don’t these food groups do the same job? In early versions of the Five Food Groups vegetables and fruits appeared in one food group but separation into two groups was suggested in 1994 in a NHMRC report called The Core Food Groups. The report stated that ‘It was necessary to group fruit and vegetables separately because of their differing nutrient profiles’. Really?
Consolidating these similar food groups into one would appear to make sense and would provide more flexibility for those who find it difficult to eat five serves of vegetables each day, which is most of us.
Image: source
Grain foods: distinguished by fortification
The dietary modelling for the new food guide demonstrates the distinctive and very substantial nutritional contribution of the grains food group. In the Foundation Diet for women aged 19-30 years, grains provide 62% of the thiamin, 55% of the folate, 46% of the iron, 39% of the iodine, 38% of the fibre, 32% of the niacin and 28% of the riboflavin. That’s impressive. By far the majority of these nutrients come from wholegrain cereals, though I suspect these were actually fortified wholegrain cereals.
Compare the nutritional content of 600kJ of the following grain foods, expressed as a percentage of the relevant Nutrient Reference Value.
Grain White rice Brown rice W/meal pasta ‘Weetbix’
Thiamin 6% 9% 31% 94%
Folate 2% 6% 3% 88%
Iron 7% 8% 32% 65%
Fibre 2% 6% 25% 19%
Niacin 13% 24% 29% 77%
Riboflavin 1% 3% 9% 78%
This table demonstrates some key points:
• refined grains are relatively nutrient-poor
• wholegrains are slightly better, but
• only fortified grains are nutrient-dense.
Fortification strongly shapes the nutritional contribution of the grains food group. As we enter the era of carbohydrate restriction there will be an increased focus on nutrient-rich carbohydrate foods and in the case of grains that means more emphasis on the fortified foods – breads and breakfast cereals – and less on white rice and plain pasta.
Dairy foods: should they be a group?
It has been argued that dairy foods hold a privileged position in Five food Groups and could be grouped with meats with which they share distinguishing nutrients, such as protein and vitamin B12.
The latest Harvard Healthy Eating Plate went even further, excluding dairy products altogether. This seems to be a step too far as the rationale was fairly weak. The Harvard team point out that dairy products are rich in saturated fat, but why not just recommend lower fat dairy foods? They argue that high intakes of calcium do not translate directly into stronger bones and that other factors are relevant, such as vitamin D and physical activity. Fair enough, but aren’t they all pieces of the same puzzle?
Harvard concedes that “Moderate consumption of milk or other dairy products—one to two servings a day—is fine, and likely has some benefits for children.”
Image:source
Meats and alternatives: the illogical food group
Unfortunately, the logic of distinguishing nutrients that underpins the Five Food Groups falls apart when we get to the meat, fish, poultry and ‘alternatives’ food group. Our current Dietary Guidelines for Australians report describes the distinguishing nutrients of this food group as protein, bioavailable iron, zinc, vitamin B12 and long-chain omega 3s. This is valid for red meats, poultry and fish but not for legumes and nuts.
The marked lack of nutrition equivalence in the meats and alternatives food group has been documented. Although legumes and nuts contain iron its absorption from these foods is very low, about 1-2%. Phytates, found in abundance in legumes and nuts, are also potent inhibitors of zinc absorption. And legumes and nuts contain no long-chain omega 3s or vitamin B12. However, legumes and nuts contain dietary fibre, unlike the animal-sourced foods in this group. These are very different foods.
Despite these inconsistencies the composition of this food group won’t change because it reflects an ideological view strongly held by some of the nutrition profession. It’s not driven by science; it’s all about us.
Legumes can at least claim to be a protein food, which is more than can be said for nuts. The recent dietary modelling for the next food guide noted that nuts ‘have a different protein to energy ratio compared with other components of the meats group’, a delicate way of saying that the major macronutrient in nuts is not protein, but fat.
Healthy fats: the missing food group?
If nuts and seeds are a poor fit in the meats food group, where do they belong? Is it too radical to suggest that they belong in their own food group, together with the oils that are derived from them and foods made from these oils, such as unsaturated margarines, salad dressings and mayonnaise? Such a food group would certainly have its own distinguishing nutrients – fat-soluble nutrients. The dietary modelling conducted for the latest food guide (p276) highlights that although these foods contributed just 10.2% of dietary energy in a Foundation Diet they provided 47% of the linoleic acid, 50% of the alpha-linolenic acid, 44% of the retinol, 29% of the vitamin E and 26% of the vitamin D. That’s a substantial and very distinctive contribution.
Shouldn’t there be at least one food group designed to encourage adequate intakes of fat-soluble nutrients? Fats were included as a food group in Australia’s Five Food Groups until 1998 when fat phobia overwhelmed our nutrition authorities. They still haven’t quite recovered.
I am excited about your comments! We know that much of the five food groups represents the politics of food in the 1920-30s. Diary products were given a special group status to economically assist the dairy industry. This concept was further developed by Clements when he conducted the 1936-1938 Final Report on Nutrition which was designed to prove to the Australian population they were not malnourished during the depression. It is a very worthy concept that we re-examine the groupings and move forward. I will support this approach.
Hi Dianne. Nice historical insights. Thanks for filling in the gaps for us. I couldn’t agree more that we need to ‘re-examine the groupings and move forward’. The trouble is that there is a deep vein of conservatism in nutrition that won’t allow this to happen. In the recent dietary modelling for the development of the new food guide nuts and seeds were separated from the meats and modelled as an independent group. But when the design for the guide was put together they bundled nuts and seeds back in with the meats. There is no logic to it; it’s just the way it’s always been done. Can’t go changing things Dianne. There is plenty of food politics at play in 2013. Regards, Bill.
I have only just discovered your site, and am really enjoying reading your posts. I have only read a few, but here I would like to add a point of view to your discussion re combining fruit and vegetable groups. Whilst there are indeed common nutrients, you have only mentioned the micros. What about macros, namely carbohydrate? With the exception of a few vegetables, the majority of fruits are much more energy-dense (a quick scroll through NUTTAB suggests >10% CHO in fruits vs. <5% for most veg), especially when considering dried and canned fruits, which are still 'allowed' as a serve of fruit. One of the great things about having the two separate groups is that, again with the exception of the starchy veg, we can 'fill up' on veggies with minimal contribution to our daily caloric intake. Imagine if we recommend 5 serves of fruit OR veg, and have people opting mostly for fruit – is this not in part misleading? I feel that the density of key nutrients should be considered – for instance folate is more predominant in certain vegetables than in fruits. Just my two cents worth!
And a good two cents worth it is, Rachel. Of course you are right – the energy densities of fruits and vegetables differ considerably. But the question I would ask is: how relevant is energy density to obesity? Based on research published in the last decade, I don’t think it’s that important (see my earlier post ‘Do calorie-rich foods make you fat?’ June 2012).
Should we advise against the consumption of nuts because they are energy dense? A healthy food is a healthy food, irrespective of its energy density. If you target energy density you end up recommending a low fat/high carbohydrate diet, which we now know is less than optimal for chronic disease prevention. Regards, Bill
Thank you for your reply. I had a read of your earlier post – another interesting one! It does make it quite difficult when we have a national document which shapes so much of our practice. It is even more challenging for those of us who have not been practicing in the field for very long and who have really only been exposed to these documents (rather than a wider evidence base) and encouraged – by senior and respected professionals – to use them as a basis for our assessments and recommendations. And further that much of our educational material is based on these national recommendations. Personally I would find it a bit confronting to decide to disregard national guidelines when making recommendations.
And with regard to what we recommend in the way of healthy foods vs. energy density – I do agree with your comments. We aren’t going to advise against a healthy foods just because they are higher in calories. Having said that, we wouldn’t promote excessive consumption of, say, nuts just because they are good for us. We can still have too much of a good thing!
For me, the simple messages of balance, moderation and looking at the overall diet (in terms of both calories and nutrients) seems to be the best approach.
Hi Rachel. I understand that it may be difficult for any dietitian-nutritionist to decide on a practice that’s inconsistent with national guidelines. So, it is incumbent on our national authorities to ensure that their advice is up-to-date. But is it?
Who knows what our new Australian Dietary Guidelines (expected to be released this week) will have to say? The draft released for consultation in December 2011 was appalling. Despite the systematic reviews finding no links between dietary fat intake or energy density and body weight, the text was full of comments about the need to limit energy-dense foods to address overweight. The same old advice – without an evidence base.
The section on saturated fat and heart disease was even worse – a relevant literature review wasn’t even conducted. After four years of stonewalling I am told the authorities finally agreed with the Heart Foundation’s position that saturated fats should be replaced by unsaturated fats (not carbohydrate). This has been known since 2009 yet it was very nearly excluded from guidelines being released in 2013.
The review of the Dietary Guidelines has taken FIVE YEARS! I suspect that the guidelines will be 2-3 years out of date before they are even released. I no longer look to the Dietary Guidelines for guidance. I read a lot, go to conferences and make up my own mind. That way I have a chance of being up-to-date, at least in my areas of expertise. Regards, Bill.
Thank you for this article Bill – these issues with the food grouping system need to be highlighted.
Agree totally that the meat grouping doesn’t make much sense. Just how do macadamia nuts and beef belong in the same group? Perhaps as well as conservatism there is an element of political correctness as well in that the model tries too hard to incorporate vegetarian diets?
Would another approach be to consider vegetarian and/or vegan diets as a completely separate issue with their own modelling system – rather than trying to incorporate vegan diets into the modelling by having nuts and legumes incorporated into a “meat and alternatives” group? Maybe the results would be quite different if the modelling for vegan diets was started from square one, with legumes as a food group, and sources of fat (nuts, oils/spreads, avocado) as another.
Hi Paul. I agree with everything you have said. The issue of vegetarian diets is a tough one. There is often a lot of passion at play and nutrition science can play second fiddle. Separate modelling of vegan and vegetarian diets would certainly highlight some nutritional challenges and how they may be overcome. But equating beef and macadamias is silly. 92% of the calories in macadamias come from fat! Regards, Bill
Hi Bill, I have only just come across your website and am looking forward to reading more of your articles! I am particularly interested in your mention of the dairy food group and its links to the Dairy Industry. I was intrigued when Harvard put out their food plate and I am looking forward to researching more into this area (you may have possibly already written an article on this!). I certainly do not get my 3 serves of dairy a day. Thanks for your insight!