Do traffic lights on food labels help us eat better?

There are a number of smartphone apps now available that rate the fat, saturated fat, sugar and salt content of foods as red, amber or green. The question is, do they help you choose nutritious foods? I covered this topic recently on my BLOG.

The Australian Government rejected a proposal for mandatory front-of-pack traffic light labels on foods recommended in ‘Labelling Logic’ (also known as the Blewett Report) in late 2011. This was disappointing to many staunch public health advocates who believed this was an effective way to improve the nations eating habits (of course it doesn’t automatically follow that more information translates to better choices because the influences on food choices are complex). Organisations such as the Obesity Policy Coalition, The Cancer Council and CHOICE were notable supporters. Not to be defeated, two of these organisations have gone ahead with their own traffic light labelling systems utilising smart phone apps. The Obesity Policy Coalition were first out of the blocks with an app called Food Tracker (http://www.opc.org.au/take-action/trafficlightfoodtrackerapp.aspx)in which you entered the numbers for total fat, saturated fat, sugar and sodium from a food label and it generated a colour rating of red, amber or green for each nutrient. This process of manually entering the numbers into the app is a disadvantage because it takes so long to do, and probably not practical to use while shopping. On the plus side, you can save your ratings in a ‘pantry’. Then came the even more tech-savvy app called FoodSwitch (http://www.bupa.com.au/foodswitch) from the health insurance giant BUPA in collaboration with the George Institute for Global Health (well known for campaigning against excessive salt in foods). FoodSwitch is more streamlined because it uses your smart phone camera to scan the barcode of products to instantly generate colour ratings for each nutrient, and sensibly includes an un-rated kilojoule content (albeit per 100g). “Healthier choices” are listed underneath your scanned food result, and you can keep recent results and organise foods into lists. Cleverly, if you scan a food that isn’t on their database, the app asks you to take a photo of the front of the pack and the Nutrition Information Panel (NIP) so it can be added. It is only available for iPhone for the moment, but the Android version is soon to be released with any teething troubles sorted out.

I took to my own pantry to test out FoodSwitch and these are the results, and my comments.



Sultana Bran

Fat GREEN, saturated fat GREEN, sugars AMBER, salt AMBER

“Healthier options”: recommended a brand lower in fibre as healthier (which is questionable), and one Coles cereal is recommended as healthier despite having the same traffic light ratings (and higher in kilojoules) so I’m not sure why this would be a “healthier choice”. Perhaps the advice should say “Similar or healthier choices”?



Quick barley and oats

Fat AMBER, Saturated fat AMBER, Sugars GREEN, Salt GREEN

“Cannot determine if a healthier choice is available”

I think this is possibly the best example of how a criteria for fat can create unwanted collateral damage: this cereal is just barley and oats yet it attracts two amber spots, denoting caution. I’d say ignore FoodSwitch – this is a super healthy breakfast cereal option.



Weet-Bix multigrain

Fat AMBER, Saturated fat AMBER, SUgars AMBER, Salt AMBER

The total fat is 1.7g per serve (3.6g/100g), of which 1.2g is saturated, which is a tiny amount. An amber rating for fat when the fat is coming from the five wholegrains themselves is misleading. I would still recommend this cereal as a healthy wholegrain cereal, but they probably could reduce the salt in this cereal to make it better.



Apricot and almond natural muesli (with the Heart Foundation Tick)

Fat AMBER, Saturated fat AMBER SUgars GREEN, Salt GREEN

“Healthier choices: We were unable to identify a similar product with a healthier profile”

I guess this means you’ve hit the jackpot and there’s no need to look further. Amber ratings for both fat and saturated fat demonstrate the unsuitability of using the same criteria across all food categories. The fat in this product is coming primarily from the 69% rolled oats content, plus small amounts of vegetable oil, sunflower seeds and almonds (which all contain healthy fats anyway). Gosh I hope people don’t swap this nutritious low GI muesli for some overly refined, over-sweetened high GI cereal disaster because of two amber dots.



Canola oil

Fat RED, Saturated fat RED, SUgars GREEN, Salt GREEN

“Healthier choices: choose cooking oils that are higher in monounsaturated and polyunsaturated fat”. Um, I already did- canola oil contains both and is recommended as a healthy oil by the Heart Foundation. Having two red dots could well discourage people eating it when it is healthy oil. An obvious failure of the app to properly assess the healthiness of this food.



Extra Virgin Olive oil

Fat RED, Saturated fat RED, Sugars GREEN, Salt GREEN

“Healthier choices: choose cooking oils that are higher in monounsaturated and polyunsaturated fat”. Yep, already did that. The traffic light system driving this app unhelpfully ‘red-lights’ foods rich in healthy oils.



Flora Buttery (margarine spread with the Heart Foundation Tick)

FAT RED, Saturated fat RED, Sugars GREEN, Salt GREEN

“Healthier choices”: were given as three cholesterol lowering spreads and three other spreads with exactly the same colour ratings including products with lower total fat contents (ie ‘extra light’ spreads). Margarine spreads contain unsaturated fats including essential omega-6 and omega-3: there’s no reason to recommend spreads with less of these good fats. “Light” and “extra light” spreads are not healthier and merely fill a marketing space generated by misguided fear of fat in the community.



Canned pink salmon

Fat AMBER, Saturated fat GREEN, Sugars GREEN, Salt AMBER

“Healthier choices: Medium Red Salmon”, but it is has the same rating for all nutrients so why is it healthier? I’d hate to think it is because the red salmon has 120kJ less per 100g (which is inconsequential).

This salmonis in brine so all of the fat is from the salmon itself. An amber rating is described as “OK” by FoodSwitch but logically you’d think amber means “slow down and think carefully”. Many Australians do not eat adequate omega-3 fats and canned salmon is a convenient and nutritious source. An amber rating for the fat in pink salmon is misleading and has the potential of someone swapping from salmon to tuna to get rid of the amber but missing out on a whole lot of beneficial omega-3s.



Light ice cream

Fat GREEN, Saturated fat AMBER, Sugars RED, Salt GREEN

“Healthier choices”: it recommended a Coles brand as healthier when it had the same ratings for all nutrients-perhaps it should say “similar or healthier choices”?



Sun-dried apricots

Fat GREEN, Saturated fat GREEN, Sugars AMBER, Salt GREEN

“Healthier choices: This is a dried fruit product and contains naturally occurring sugar”.



Fruit salad treats (canned fruit in juice)

Fat GREEN, Saturated fat GREEN, Sugars AMBER, Salt GREEN

Healthier choices: Fourteen alternatives all with the same colour ratings, including some with higher kilojoule ratings. Not sure why they’re “healthier”. Not sure what’s going on with this one. Canned fruit in juice is recommended as a fresh fruit alternative. Why would it attract an amber rating for sugars? It has scored the same rating as the dried apricots above when the sugars content is much lower than the dried fruit.



Conclusion

FoodSwitch is an elegant technical solution for obtaining an instant traffic light rating for foods, however there are several shortcomings to the whole traffic light system. The most obvious is, which nutrient is most important when a food has red, amber and green ratings for different nutrients? Do people generally know how to reconcile one against the other? There are a few issues with the recommendations from FoodSwitch too, such as why another food product is a “healthier choice” when it has an identical rating? A similar choice maybe…



The main problem with traffic light labelling systems generally is the nutrient criteria upon which they are based are scientifically flawed. The criteria are based on 100g of the food when this can give a skewed result for foods we eat in small amounts, such as oils and margarines we eat in 10-20g servings or even breakfast cereals we usually eat in 40g servings. Criteria also need to be category specific because foods are so diverse: what makes a healthier breakfast cereal is very different to what makes a healthier ice cream or frozen dinner (that’s why the Internationally respected Heart Foundation Tick program has different criteria for each food category). To use the same nutrient criteria for all foods based on 100g is fundamentally flawed.



To have a total fat criteria is meaningless because a food’s total fat content has no bearing on its healthiness: chocolate chip ice cream and avocado are both high in fat, but they are obviously not both unhealthy. Low fat diets are no longer generally recommended because we need to include healthy oils and foods containing them in our diets for optimal health. And this is where total fat cut-offs fail terribly. The community at large is already fat-phobic- to the detriment of their diet and health – and putting a big red dot on foods with naturally high oil content can only make this worse. Oh, and eating fat does not cause obesity: eating too many kilojoules you don’t use causes weight gain.



Having a saturated fat cut-off makes logical sense because this is a nutrient of public health importance (Australians eat too much saturated fat and need to eat less), however all foods naturally high healthy oils will score badly because having a cut-off fails to acknowledge that it is the ratio of saturated fats to unsaturated fats and not the absolute amount of saturated fat that determines whether a food is healthy. That’s why the canola and olive oil and the margarine spread (with the Heart Foundation Tick) above unfairly get a red rating for saturated fat from FoodSwitch. This is not good.



Although many people have a very negative perception of sugars, in public health terms it is not directly related to any adverse health outcomes. Indirectly it can add kilojoules we don’t need, but that’s about it. How much sugar you eat is unrelated to your risk of heart disease, cancer or even diabetes. And starches are just as bad as sugars when it comes to contributing to dental decay (and heart disease for that matter). I’m not saying it’s a good idea to eat lots of added sugar, but it is nutritionally naive to say just because a food is high in sugars it is unhealthy. Because food laboratories and the human body are unable to distinguish between natural or added sugars, food labels list total sugars and the colour rating is based on this. This means that foods naturally high in natural sugars such as dried fruit are unfairly given a red or amber light. To be fair, FoodSwitch says “dried fruit contains naturally occurring sugar” and we can only hope people understand it’s OK to eat them (albeit in sensible amounts).



The one major shortfall of traffic light labelling systems is they ignore the most important determinant of weight gain, and that is kilojoule content. To its credit, FoodSwitch includes the kilojoule content (albeit not colour rated, and expressed per 100g). Of course this raw number is meaningless to most people (but you can check out the Percent Daily Intake on some food products), but its inclusion is a step up from other traffic light systems.



Maybe its just as well the goverment said no to mandatory traffic light labels on foods…



Traffic light labels may not always stear you toward the healthiest choice