Canadian Food Inspection Agency
Guide to Food Labelling and Advertising

Section VI: Nutrient Content Claims
Section 6.3

6.3 Compliance for Nutrient Content Declarations and Claims

6.3.1 General Principles

1. It is the responsibility of industry to ensure that labelling information accurately reflects the nutrient content of the product;
2. Manufacturers should have good quality control of the formulation of the product to minimize variability;
3. The production lot should be properly sampled and analyzed by trained staff using recognized methods of measurement. The analyst may determine how best to collect and analyze products in order to ensure accuracy of the declared values; and
4. Canadian Food Inspection Agency compliance action, when necessary, will consider not only laboratory results, but also factors such as the past compliance history, the health risk, the potential economic loss to the consumer and the company's quality control over the analyses and processes.

6.3.2 Definitions

1. A lot is a collection of primary containers or units of the same size, type and style produced under conditions as nearly uniform as possible, designated by a common container code or marking, or in the absence of any common container code or marking, a day's production.
2. A sample is the unit of analysis. It shall consist of individual consumer units selected randomly from a lot; the consumer units may be composite and analyzed as a single sample, or may be analyzed individually and the results averaged.

6.3.3 Nutrient Definitions

The basis for nutrient definitions is found in Calculating the Nutrients, Section 6.4.

6.3.4 Methods of Analysis

Nutrients should be measured using internationally accepted methods of analysis such as AOAC methods* or Health Products and Food Branch (HPB), Health Canada acceptable methods. (* Official Methods of Analysis of the Association of Official Analytical Chemists, 16th Edition, AOAC, Arlington, Virginia, 22209, U.S.A.)

6.3.5 Tolerances for Nutrient Declaration (Amended 16/03/98)

(i) Type I Nutrients: Where a nutrient is subject to a regulatory minimum (e.g., when vitamin C is added to fruit flavoured drinks), or a regulatory maximum (e.g., low fat), or has been specifically highlighted by a claim (e.g., 25 percent reduced in Calories), the total amount of the nutrient including any amount due to natural occurrence, is classified as a "Type I Nutrient". Included in this group are foods for special dietary use.

For Type I nutrients, five sample units should be drawn at random and analyzed as a composite or separately. If the five samples are analysed separately, the test results will be averaged to give a mean result. (The size of a sample unit will vary depending upon the nutrient analyzed, the methodology used and the food; on average, a 250 g sample may be sufficient).

When a minimum value is required by Regulation, or stipulated in a guideline, the lot will be deemed to be out of compliance if the result of the analysis is less than the minimum required value. The lot will also be out of compliance if the result of the analysis is less than 90% of the declared value, or if any one sample has less than 30% of the declared value.

When a maximum value is prescribed by Regulation, or stipulated in a guideline, the lot will be deemed to be out of compliance if the result of the analysis is greater than the maximum value prescribed. The lot will also be out of compliance if the result of the analysis is more than 110% of the declared value, or if any one sample has more than 170% of the declared value.

In the case of fat-free claims, a tolerance of 0.1g per serving and per reference amount is allowed. The lot will be deemed to be out of compliance if the result of the analysis is greater than 0.5g per serving and per reference amount.

Reasonable overages of the added vitamins and minerals within good manufacturing practice should be present to ensure that the required level of vitamins and minerals are maintained within the expected shelf life of the food.

(ii) Type II Nutrients: A nutrient which is present naturally in the food and for which there is no regulated minimum or maximum is classified as a "Type II Nutrient" (an example is vitamin C in orange juice).

For Type II nutrients, 12 sample units (see sample unit standards specified above) should be drawn at random and then combined to make a composite to be analyzed.

For vitamin, mineral, protein, carbohydrate, dietary fibre, polyunsaturated fatty acids, monounsaturated fatty acids, or potassium, the nutrient content of the composite must be at least equal to 80 percent of the value declared on the label.

For energy, sugars, fat, saturated fatty acids, cholesterol or sodium, the lot is deemed to be unsatisfactory if the result exceeds 120 percent of the declared value.

For both Type I and Type II nutrients, the amount of vitamin, mineral, protein, carbohydrate, starch, dietary fibre, polyunsaturated fatty acids, monounsaturated fatty acids or potassium, may vary over labelled amounts within good manufacturing practices. The amount of energy, sugars, fat, saturated fatty acids, cholesterol or sodium may vary under labelled amounts within good manufacturing practice.

Rationale: Due to the inherent heterogeneous nature of food products, it is recognized that any declaration of the nutrient content is subject to variance. The content will fluctuate from container to container, lot to lot, and season to season, particularly for a naturally occurring nutrient. The concept of two different systems for assessing the compliance of nutrients subject to a regulated minimum, a regulated maximum, or which have been specifically highlighted by a claim (Type I) and naturally-occurring (Type II) nutrients, is to address the two levels of variability known to exist.

For a Type I nutrient, where the nutrient content is declared at the stipulated minimum or maximum, that stipulated level is established as the tolerance for the product. Where the declaration is made below the maximum level or above the minimum level, an additional 10 percent tolerance is applied to the declared value.

The option of testing a composite sample or testing individual samples is provided to address the cost of analysis. It is recognized that for some products, the sample-to-sample variability may be insignificant, whereas with other products, this variability may be a major factor. It is, therefore, the responsibility of industry to determine how best to collect and analyze their products to ensure the accuracy of declared values.

For the naturally occurring Type II nutrients, the manufacturer does not usually have control over the variability of the nutrient from container to container. A larger tolerance of 20 percent of the declared value is allowed in recognition of this.

The sampling plan calls for the testing of a composite sample and does not attempt to monitor this variability. The main objective is to ensure that the sample taken is representative of the lot and measures, as closely as possible, the lot average.

Note: These tolerances have taken into consideration variability among containers within a lot, variability in mean nutrient values between different lots, as well as within laboratory and between laboratory tolerances.

6.3.6 Databases

For compliance with nutrition labelling regulations, the use of industry-maintained databases for some food products may be acceptable. Health Canada will assist industry in evaluating databases to determine their appropriateness for nutrition labelling. Guidelines on the measurement of nutrients in Canadian foods for the purposes of nutrition labelling are provided in Annex 3.

Products using nutrient values from a database for labelling and advertising, that are found not to be in compliance, may avoid enforcement action provided it can be demonstrated that production techniques were consistent with good manufacturing practice and corrective action has been initiated. The use of a database does not eliminate the need for and industry's responsibility for periodic analysis of the product to ensure that nutrient values are still within the limits of the database.

In the design of the sampling plan for gathering data for a database, all major sources of variation should be considered. Each sample should be a substantial composite to ensure that it is representative of the lot. The methods of analysis must have been validated for the food and performed by experienced analysts in a laboratory where there is an established system of quality assurance.

A submission of a database must be accompanied by a complete disclosure of the identity of the food, the reliability of the analysis, the variability encountered in the results, the method of analysis used, along with the name of the laboratory and analyst or responsible manager. The information should include individual results for all samples with the number of replicates and the standard error or standard deviation stated.

Rationale: The use of databases is viewed by some as a potential cost-saving tool to the industry, by reducing the need for vigorous laboratory testing. However, the reliability of databases to determine natural variability of a nutrient in a food and for meeting compliance requirements for nutrition labelling must be assessed carefully and cautiously on a case-by-case basis.

6.3.7 Use of the Canadian Nutrient File

The use of tables of food composition, such as the Canadian Nutrient File, for the purpose of labelling is to be avoided, since these values may not be representative of products currently on the market. An exception would be where a database as described above has been created and accepted for use in the Canadian Nutrient File.

6.3.8 Commodity-Specific Compliance Policy for Nutrient Content Declaration and Claims

A compliance policy for minimum protein standards and nutrition labelling of meat and poultry products containing phosphate salts or water is found in Section 8.10.3.8. This policy is consistent with the compliance policy for nutrient declaration with some differences in tolerance levels to reflect the differences in the purpose of the regulatory requirement for these commodities. It is intended that this policy appear in the commodity specific sections planned for this compendium.