This week I was shown two labels of canned sockeye salmon. These products were from different brand names but otherwise appeared identical:
- Ingredients: Wild Sockeye Salmon
- Serving Size: 55grams
- Per the person consuming these products – reports they appear the same, with no obvious differences
But the reported protein content on one product was 11g and the other was 20g. So what’s going on here? Isn’t salmon salmon? How can brand of salmon contain 11g of protein while another brand contains 20g?
Is the information out of date?
I could see that the Gold Seal label is older (as they don’t report potassium). So I went to their website to see if the nutrition information has changed. And they still report 20g of protein in their unsalted sockeye salmon product. So that didn’t appear to be the reason.
How much protein does the Canadian Nutrient File report?
My second step in solving this mystery was to check out the Canadian Nutrient File. What did they report the protein content of canned salmon to be?
They had a few different listings for canned sockeye salmon – so I checked them all out (see table below). I calculated the protein amount in 55g based on the amount of protein in 100g.
| Description | CNF Number | Protein per 100g | Data Source | Protein per 55g |
| Fish, salmon, sockeye (red), canned, drained solids with bone, unsalted | 3226 | 20.47 | USDA | 11.25 |
| Fish, salmon, sockeye (red), canned, drained solids, salted | 3223 | 23.59 | USDA | 12.97 |
| Fish, salmon, sockeye (red), canned, drained solids, without skin and bones | 7080 | 26.33 | USDA | 14.48 |
Fish, salmon, sockeye (red), canned, total can contents | 7078 | 20.63 | USDA | 11.34 |
So as per the Canadian nutrient file, 11g of protein is more likely correct and than 20g. But is this just a one off? Or has this sort of thing been previously reported?
What do we know about labeling accuracy for protein?
I couldn’t find any studies from Canada. I found a few studies from the USA and Australia.
In one American study – they reported that snacks foods labels were fairly accurate in terms of their protein reporting. In another study, exploring low/no protein foods, the authors reported that several 0g protein products contained protein-containing ingredients. The authors concluded that these labels were likely inaccurate as it is unlikely that a product with protein containing ingredients contains no protein at all.
In a 2006 Australian study, the authors reported significant concerns with the accuracy of food labels. 350 foods were sampled. Protein content was analyzed in the lab using very good methodologies (4 samples per food!). The authors reported that 20% of food items had a more than +/-20% off from the reported protein content. That’s 1/5 products!
Where does the information on a nutrition label come from?
I had always assumed that the amount of a nutrient listed on a label was determined by a lab analysis of the specific food. The Canadian Food Inspection Agency, acknowledges that this is the most accurate method. However, their guidance documents clearly state that food producers may also calculate the amount of a nutrient in food from nutrient databases or by making comparisons with other similar products.
How accurate do labels need to be?
For protein (and other minerals – such as potassium) CFIA has adopted a 1-sided limit. This means that in order for labels to be compliant with CFIA regulations, the product must contain 80% or more protein. There is no limit on under-reporting – other than “good manufacturing practice”.
Lets wrap our heads around this with our cans of salmon:
- if it says 11g and actually has 20g, it contains 180% of the declared protein content. Is it compliant? YES
- if it says 20g and actually has 11g, it contains 55% of the declared protein content. Is it compliant? NO
Below is a copy and paste from the CFIA document that I am reading to make this claim.
Q2. What would be a reasonable excess or deficiency? For example, if the vitamin content of a sample must be at least 80% of declared amount, how much variation above the declared amount is reasonable?
Canadian Food Inspection Agency, Nutrition Labeling Compliance Test, Part 2 Analysis and Feedback
Most agreed with the proposal that nutrient values with a minimum requirement, may exceed labelled amounts “within good manufacturing practices” and that nutrient values with a maximum requirement may be less than labelled amounts “within good manufacturing practices”. One-sided tolerances were considered appropriate in circumstances where health concerns are relevant primarily to over-consumption or to under-consumption, because they ensure that the claimed amount of the nutrient is present in the food without unreasonably varying beyond what is declared to ensure the declared level is present. A few did not favour one-sided tolerances. It was suggested that there should be upper limits for carbohydrate values, of concern to consumers with diabetes, as well as upper limits for protein values, of concern to those following a special diet low in protein. Also, there was concern that the one-sided tolerances would lead to misleading labels. An industry representative suggested a range between 80% and 120% of the label amount. The CFIA acknowledges the merits of a two-sided approach, i.e. an acceptable range of 80%-120% of the label amount; however, since this range may be difficult to achieve for many manufacturers, a one-sided approach, either minimum or maximum provides the greatest health benefit.
Therefore, the proposed approach to allow nutrient values with a minimum requirement to exceed the label value “within good manufacturing practices” and those with a maximum requirement to be less than labelled amounts “within good manufacturing practices” will be retained with the proviso that there is no risk to health.
Take Aways
Back to the salmon – I think the easy answer is that one of those labels is wrong. Based on the Canadian Nutrient File, it is highly likely that the can claiming 20g of protein is incorrect. As I am assuming this is a case of over-reporting (by 45%) it is not compliant with CFIA labeling regulations.
However, in the event that I am wrong and both salmons contains 20g of protein, then both products would be compliant. In Canada, we don’t set limits on under-reporting for minerals such as protein and potassium on food labels. Our label regulations only have 1-sided limits.
Thanks for a great post – very thought provoking!
I noticed further inaccuracies on the Gold Seal salmon label. The overall kcal reported on the gold seal salmon also appears incorrect. 4.5 g fat (40.5 kcal) + 20 g protein (80 kcal) = 120.5 kcal per 55 g serving, not the 90 kcal reported.
It makes me wonder if the serving size was shifted to a smaller portion for some reason; I feel that previous labelling for canned fish more so focused on a ~3 oz 120 g serving size) but the remainder of the label wasn’t updated appropriately?
I hadn’t noticed that! Thanks for doing the math. I think you are on to something – it makes sense that the error could be a change in portion size without updates to the nutrition information. In my hastiest of searches – I tried to see if there was any legacy data available in Food Data Central (https://fdc.nal.usda.gov/fdc-app.html#/?query=gold%20seal) but unfortunately there wasn’t. That being said as more information becomes available in the Global Branded Products Database, which can be accessed through food central, hopefully tracking these tips of things and how label information is changing overtime will become easier!