Questions and Answers about Preparing Submissions for Food Health Claims
Table of Contents
- 1.0 General Information about Preparing a Submission for Food Health Claims in Canada
- 1.1 Why does Health Canada have Guiding Principles for health claims?
- 1.2 How many studies are required to substantiate a health claim?
- 1.3 How does Health Canada determine the final wording of a health claim?
- 1.4 If a company receives a favourable decision in response to its application for a health claim, can competitors be prevented from using the claim?
- 2.0 Information about the Characterization of the Food
- 3.0 Information about the Characterization of the Health Effect
- 3.1 Why is it important to characterize the health effect that is the subject of the health claim?
- 3.2 How should an outcome be selected to characterize the health effect?
- 3.3 What is meant by the terms methodological validity and biological validity?
- 3.4 What information is needed to demonstrate the biological validity of a biomarker?
- 3.5 What sort of information is needed to show that the proposed effect is beneficial to human health?
- 4.0 Information about the Retrieval, Synthesis and Validation of the Scientific Evidence
- 4.1 What is the best way to manage the results of the literature search?
- 4.2 Can unpublished data be included in the Evaluation of Claim Validity? How should it be handled?
- 4.3 What is meant by the term target population and what types of people are considered generally healthy?
- 4.4 Can studies in which the bioactive substance of interest was administered in capsule format be selected for inclusion in the evidence provided to support a proposed health claim?
- 4.5 One of the suggested criteria for literature filtering is the use of an appropriate control/placebo in intervention studies. What is the purpose of an appropriate control/placebo?
- 4.6 Are baseline-controlled studies appropriate for inclusion in the Evaluation of Claim Validity?
- 4.7 What is the purpose of the quality appraisal tool? What if studies that support the health claim are rated as lower quality?
- 4.8 What is a meta-analysis? Should a meta-analysis be performed on the studies selected for inclusion?
- 4.9 What is the minimum effective intake of a food and how should it be determined?
- 4.10 What sort of evidence is needed to show that the magnitude of the effect is physiologically meaningful/biologically relevant?
- 4.11 Where should information on intakes of the food in the target population come from?
1.0 General Information about Preparing a Submission for Food Health Claims in Canada
1.1 Why does Health Canada have Guiding Principles for health claims?
Health claims on product labels and in advertising convey information about food characteristics and related health benefits. This information has the potential to influence consumer food choices in a manner that could affect their health; therefore, it must be accurate and reliable.
Health Canada is responsible for the development of policies, regulations and standards that relate to health and safety. To ensure that food health claims are substantiated in a systematic, comprehensive and transparent manner, Health Canada has developed standards of evidence, also known as Guiding Principles. The process required to meet the standards of evidence may be perceived as resource intensive; however, it is necessary to help ensure health claims are justified based on the totality of the scientific evidence (not just the favourable evidence).
Evidence-based requirements to substantiate food health claims are intended to enhance consumer confidence in the health-related information provided by industry on product labels and in advertising. Health Canada’s standards of evidence are generally consistent with those of other scientific and regulatory authorities, including the European Food Safety Authority, Food Standards Australia New Zealand, the Cochrane Collaboration and the Agency for Health Care Research and Quality.
1.2 How many studies are required to substantiate a health claim?
The relevance, quality and strength of the studies, as well as other factors such as plausibility of the food-health relationship, are more important than the quantity of studies. A minimum of two independent, well-designed studies are needed to demonstrate reproducibility of the effect of the food.
1.3 How does Health Canada determine the final wording of a health claim?
Health Canada evaluates whether the wording proposed by the petitioner is i) an accurate reflection of the scientific evidence and ii) not misleading to consumers. If the proposed wording does not meet these standards, Health Canada revises the wording of the health claim based on the scientific evidence available and the wording of claims in other jurisdictions. If possible, consumer research is used to anticipate consumers’ understanding of the claim. The petitioner will have the opportunity to respond to Health Canada’s wording before it is finalized.
1.4 If a company receives a favourable decision in response to its application for a health claim, can competitors be prevented from using the claim?
Health Canada has not accepted a brand/product-specific health claim to date. So far, accepted health claims have been for natural components of foods or generic non-proprietary food ingredients; therefore, the conditions for using the claims can usually be met by other companies. As long as a competitor’s product meets the conditions for a specific health claim, it is allowed to use the accepted claim. If a petitioner is able to demonstrate that their product is different from that of competitors’ and that the differences are related to the claimed health effects, it may be appropriate to incorporate the unique features of the petitioner’s product into the conditions for carrying the claim.
2.0 Information about the Characterization of the Food
2.1 Why is it important to characterize the food that is the subject of the health claim?
The information requirements in the Characterization of the Food section are useful to help develop the literature search strategy and selection criteria required in the Evaluation of Claim Validity section. The food that is the subject of the proposed health claim should be described in sufficient detail to ensure that the foods tested in the studies selected for inclusion are equivalent. A detailed characterization of the food will also facilitate the development of conditions for foods to use the claim. Manufacturing and quality information should be provided for certain types of foods to ensure that characteristics important for the health effect are still present in the final product. For example, quality information such as stability data may be required to demonstrate that a bioactive ingredient is still effective at the end of the product’s shelf life.
2.2 How should Table 4, Information requirements for characterization of the food, be completed?
Table 4, Information requirements for characterization of the food, in the Characterization of the Food section, is not a template for petitioners to complete; rather it is a listing of the information needed by Health Canada to characterize the food that is the subject of the claim. Petitioners may wish to create their own table(s) to present the listed requirements. Alternatively, descriptive text with appropriate headings is acceptable.
The first item in Table 4, Dietary intake estimates, applies to all health claim submissions, regardless of the subject of the claim. The remaining requirements will depend on the subject of the proposed claim. Once the subject of the claim has been identified, the corresponding Information Requirements described in Table 4 should be provided. The subject of the health claim should fit into one of the following categories:
- Food category – A group of foods (for example, fruits, whole grains, dairy products) associated with a proposed health benefit.
- Whole (unprocessed) food – A specific food (for example, apple, strawberries, milk, tofu), rather than a bioactive substance in the food, is associated with a proposed health benefit.
- Foods containing an inherent bioactive substance(s) – A specific intrinsic constituent of a food (or a food category) is associated with a proposed health benefit (for example, galacturonic acid in apples, beta-glucan in oats, calcium in cheese, soy protein in tofu).
- Foods containing an added bioactive substance(s) (for example, yogurt with added L. casei 431, orange juice with added calcium hydroxide) – A food constituent(s) associated with a proposed health benefit has been isolated and added as an ingredient to the food.
3.0 Information about the Characterization of the Health Effect
3.1 Why is it important to characterize the health effect that is the subject of the health claim?
The information requirements in the Characterization of the Health Effect section are useful to help develop the literature search strategy and determine selection criteria required in the Evaluation of Claim Validity section. A detailed description of the health effect that is the subject of the proposed claim will help to ensure that the outcomes measured in the studies selected for inclusion are equivalent to that health effect.
All health claims, including function claims, should be supported by a specific and measurable health benefit or effect that is relevant to health. Non-specific or general claims (for example, “beneficial to health” or “supports immune health”) are subject to multiple interpretations and are potentially misleading. The only exceptions are claims about the general role of the nutrient as a factor in the maintenance of good health. More information about these types of claims can be found in the section on Nutrient Function Claims of the Canadian Food Inspection Agency’s Industry Labelling Tool.
3.2 How should an outcome be selected to characterize the health effect?
The rationale for the selection of a specific outcome will depend on the type of health effect that is being claimed. The selection of an outcome for each of the three types of health effects, therapeutic, disease risk reduction, and physiological function, is described below. For all types of health effects, in cases where the effect could be measured by more than one outcome, a rationale for the selection of one outcome as the primary outcome should be provided. A discussion of whether the food can be expected to impact other outcomes related to the health effect should be included.
3.2.1 Therapeutic Effect
If the proposed health effect is therapeutic, outcomes could include effects on risk factors for the development of a disease or biomarkers described in clinical practice guidelines.
3.2.2 Disease Risk Reduction
If the proposed health effect is the reduction of disease risk, the outcome selected should be a clinical measurement of the disease or health condition; for example, myocardial infarction could be selected as an outcome/endpoint for heart disease.
If a large sample size and/or long periods of time are required for a sufficient number of clinical events to be achieved, trials that measure the health effect directly may not exist. In such cases, it may be appropriate to evaluate the change in a biomarker (or risk factor) in place of the clinical measurement of the disease or health condition. The rationale provided for using a biomarker should explain why studies measuring the effect of the food on the health effect are not available. If the scientific evidence for a food-health relationship is based on biomarker data, petitioners will need to explain why the wording of the health claim should reflect the disease, rather than the biomarker. Information on the biological validity of the biomarker will also be needed (see Question 3.3 and Question 3.4).
3.2.3 Physiological Function
For a claim about a beneficial physiological function (that is, a function claim), the selected outcome should be related to maintaining or improving the normal body function. Function claims may also be related to physical or cognitive performance (for example, increased performance in different types of physical activities, or alertness, respectively) or self-reported perceptions of wellness or health-related ratings (for example, satiety).
3.3 What is meant by the terms methodological validity and biological validity?
Methodological validity refers to the accuracy and reliability of a method to measure the outcome of interest. Petitioners should provide information to show that the method(s) used to measure the outcome(s) are accurate, reproducible, sensitive and specific.
Biological validity refers to the nature and strength of the evidence that links the biomarker with the health effect. If the scientific evidence for a disease risk reduction health effect is based on biomarker data, then information on the biological validity of the biomarker is needed (see Question 3.4).
3.4 What information is needed to demonstrate the biological validity of a biomarker?
If a generally accepted biomarker is used as a surrogate for a clinical disease outcome, then the basis for its acceptance should be provided. For example, for a biomarker that has been accepted by other regulatory or scientific organizations to substantiate a disease risk reduction claim, the scientific basis of acceptance and the validation process used by the organization should be provided. Examples of well-established disease biomarkers for heart disease include blood pressure, blood total cholesterol, and blood low-density lipoprotein (LDL) concentration.
Evidence to support the biological validity of a biomarker is particularly important if the biomarker is not a well-established surrogate for the health effect. In this case, the following information should be provided to demonstrate biological validity:
- An explanation of the role of the biomarker in the causal pathway(s) of the target outcome (disease);
- A description of the expected mechanism of action of the test food (intervention) on the surrogate marker;
- Evidence that the effect of the test food on the biomarker predicts its effect on the target outcome;
- A description of the biomarker evaluation processFootnote 1 (that is, analytical validation, quantification, utilization);
- A description of the context of use of the biomarker to-date; and
- Published peer-reviewed papers supporting the above.
Before relying on a biomarker that is not well-established to substantiate a disease risk reduction claim, petitioners are encouraged to contact the Food Directorate of Health Canada to set up a pre-submission meeting (firstname.lastname@example.org).
3.5 What sort of information is needed to show that the proposed effect is beneficial to human health?
The type of evidence needed to show that the proposed effect is beneficial to human health (or that there is cause for concern about the health effect) will depend on the type of health effect that is being claimed. The evidence needed for each of the three types of health effects, therapeutic, disease risk reduction, and physiological function, is described below.
3.5.1 Therapeutic Effect
A submission for a therapeutic effect should include the following information to show that the effect is beneficial:
- Evidence that the therapeutic effect is a recognized predictor of reduced risk for the disease or adverse health condition;
- Evidence that the relationship of the therapeutic effect to the development of the disease or adverse health condition is biologically plausible; and
- Published peer-reviewed papers supporting the above.
3.5.2 Disease Risk Reduction
The relationship between disease risk reduction and benefits to human health is usually clear; therefore, only a short discussion is adequate to demonstrate that the proposed effect is beneficial to human health.
3.5.3 Physiological Function
The evidence required to show that a physiological function resulting from the consumption of a food is beneficial to health in the context of a healthy diet will vary depending on the function. For example, an improvement in the absorption of iron, an essential nutrient, would be considered beneficial to health for the general population. An improvement in lactose digestion would be considered beneficial to health primarily in those with lactose maldigestion. Since this condition is prevalent in some population subgroups, such an effect could be considered an acceptable function claim when the target group is identified as part of the claim. In other cases, like improvement in immune function for which the underlying mechanisms are complex, the use of a few selected immune markers is generally insufficient to demonstrate a health benefit. Corroborating evidence based on an improvement in a clinical outcome would also be required.
4.0 Information about the Retrieval, Synthesis and Validation of the Scientific Evidence
4.1 What is the best way to manage the results of the literature search?
The results of the literature search conducted for Step 2 of the Evaluation of Claim Validity can be downloaded using reference management software, such as RefWorks or Endnote. Results from each database can be saved in a different folder. A composite folder can be created and the software’s “duplicates” function used to identify and remove duplicate references. The results of the title, abstract, and full-text filtering should be saved to new folders. Results of hand searching can be uploaded or entered manually into a separate folder and then combined with the final composite folder containing the articles that met the inclusion criteria at the full-text filtering stage. The software can then be used to generate the counts and reference lists required for the submission in Steps 2-5 of the Evaluation of Claim Validity, such as Table 9, Results of literature filtering and Table 10, List of references that met the inclusion criteria at the full-text filtering stage.
4.2 Can unpublished data be included in the Evaluation of Claim Validity? How should it be handled?
Unpublished data may be included at Step 2 of the Evaluation of Claim Validity. Unpublished studies should undergo the same filtering, data extraction and quality appraisal steps as published studies. If unpublished studies are included, full-text reports should be provided. The level of detail provided in unpublished study reports should be at least equivalent to that of published studies.
4.3 What is meant by the term target population and what types of people are considered generally healthy?
Target population refers to the group of people that the health claim is directed at. The evidence obtained from the studied populations should be generalizable to the target population. Typically, the target population for food health claims is free-living, generally healthy adults. The guidance in Step 3 of the Evaluation of Claim Validity suggests that only studies with free-living, generally healthy adults should be included to ensure that the results are generalizable to this target population. This is because there is often reason to think that the effect of the food might be different in hospitalized or diseased individuals.
Generally healthy refers to people that do not have a diagnosed disease. The presence of risk factors for disease, such as elevated blood cholesterol, hypertension or obesity may be acceptable, particularly if the risk factor is not directly related to the food/health effect in question. If all of the evidence for the beneficial effect of a food comes from studies in individuals with a risk factor, then the wording of the health claim may need to indicate that the food-health relationship is applicable to that group only, unless the condition is widespread in the Canadian population.
4.4 Can studies in which the bioactive substance of interest was administered in capsule format be selected for inclusion in the evidence provided to support a proposed health claim?
Ideally, the criteria for inclusion developed in Step 3 of the Evaluation of Claim Validity will require that the product format tested is the same or very similar to the product format that will be marketed with the claim if it is accepted. However, it is possible that evidence testing one product format (for example, a capsule or cookie) may be used as part of the evidence that supports claim validity for another product format (for example, a beverage) but generally not on its own without additional information. In this case, petitioners should have a clear understanding of how the bioactive substance is expected to work before generalizing the evidence based on one product format to another format. The extent to which the evidence for one format can be generalized to another will depend on several factors. Ultimately, it will be the responsibility of the petitioner to demonstrate that the effect of the bioactive ingredient can be expected to be similar if the studies that support the effect tested a different format.
Examples of factors that should be considered include, but are not limited to, the following:
- Bioavailability – Could the bioactive ingredient in food format be less available for absorption than in capsule format because it is bound in the food matrix or because of interactions with other food constituents?
- Stability – How could various food matrices and processing affect the stability of the bioactive substance?
- Characteristics of the bioactive substance that are important for the health effect – Could these characteristics be significantly modified by different food matrices or different types of processing when the bioactive substance is added to foods?
- Survival through the digestion process – If the health effect of the bioactive substance occurs primarily in the large intestine and the substance is protected from gastric digestion when tested in capsule format, will the health effect be preserved when consumed in a different format or when subject to different processing methods?
- Target site and mode of action of the bioactive substance – Does it need to be absorbed into the systemic circulation to reach the target site? If absorption and distribution are not needed, evidence to support bioavailability of the substance in different product forms may not be required.
4.5 One of the suggested criteria for literature filtering is the use of an appropriate control/placebo in intervention studies. What is the purpose of an appropriate control/placebo?
The purpose of the control treatment is to distinguish the effects caused by the food of interest from those caused by other factors. If a control treatment does not allow the effect of the food to be isolated from those of other factors, it is not appropriate for substantiation of a health claim and the study should be excluded during the literature filtering in Step 4 of the Evaluation of Claim Validity. For example, suppose a study investigated the effects of beta-glucan fibre on cholesterol lowering. If the treatment group received a test product containing beta-glucan fibre in addition to their usual diet, then the control group should have received an isocaloric control product without beta-glucan fibre, but with a similar amount of total fibre. At the very least, the diet of the control group should have been adjusted to contain a similar amount of calories and total fibre and the control group should be instructed to avoid foods containing beta-glucan fibre. If measures such as these were not taken, it is not possible to isolate the effect of beta-glucan fibre on cholesterol-lowering from the effect of other factors.
4.6 Are baseline-controlled studies appropriate for inclusion in the Evaluation of Claim Validity?
Baseline-controlled studies are those in which outcomes are measured in the same participants before and after an intervention. The two measures are compared to estimate the effect of the intervention. An example of a baseline-controlled study would be a study that investigated the cholesterol-lowering effect of oats in only one group by measuring baseline cholesterol levels in participants, administering oats, and then measuring cholesterol levels at the end of the treatment, comparing these levels to baseline cholesterol levels.
Baseline-controlled studies should be excluded at Step 4 of the Evaluation of Claim Validity since the effect of the intervention cannot be isolated from the effect of other influences that might have occurred between the baseline and end measurements. There is no internal control in this type of study since changes from baseline are compared with an estimate of what would have happened to the subject in the absence of treatment. While evidence from studies without an internal control is not strong enough for decision-making purposes, it can be used by petitioners as exploratory research to help inform the design of randomized controlled trials.
4.7 What is the purpose of the quality appraisal tool? What if studies that support the health claim are rated as lower quality?
The purpose of the quality appraisal tool in Step 7 of the Evaluation of Claim Validity is to help discriminate between studies that have a low or high risk of bias. “Risk of bias” refers to the likelihood that study results and inferences deviate from the truth. Since bias can affect the magnitude and/or the direction of an effect, it is important to identify the studies that are at higher risk of reporting misleading results.
The quality rating is not intended to exclude studies from consideration. Rather, once the quality has been assessed for each study, petitioners are asked to incorporate the quality rating into the subsequent steps of the evaluation of claim validity. For example, as a part of Step 9 of the Evaluation of Claim Validity, to demonstrate causality petitioners are asked to evaluate the following according to study design and quality: i) the consistency in the direction of the effect and ii) the strength of the association. If differences are noted (for example, only lower quality studies show a positive effect), the petitioner is given the opportunity to provide an explanation.
4.8 What is a meta-analysis? Should a meta-analysis be performed on the studies selected for inclusion?
Step 8 of the Evaluation of Claim Validity mentions meta-analysis. A meta-analysis is a type of systematic review that applies statistical methods to combine and summarize the results of several studies. Meta-analyses can be very useful; they can provide a summary estimate of the effect size of a food or help compare the effect of a food across different studies. Petitioners are required to follow systematic methods to assemble and interpret the results of relevant studies; however, it stops short of requiring petitioners to perform a formal meta-analysis which is not mandatory for a health claim submission. Meta-analyses are a significant undertaking and can be difficult to do well. If a company possesses the resources to perform or commission a meta-analysis of the studies included in the health claim submission, Health Canada encourages the company to submit it as part of their application.
4.9 What is the minimum effective intake of a food and how should it be determined?
In Step 9c of the Evaluation of Claim Validity petitioners are asked to propose and justify the minimum amount of the food eligible to carry the claim. For the purposes of a health claim submission, minimum effective intake refers to the smallest amount of intake of the food that gives a biologically relevant change (see Question 4.10) in the health outcome of interest (Scenario I). On the other hand, when there is evidence (from human or animal dose-response studies or biological plausibility) to suggest that a plateau in the proposed health effect is reached with increasing intakes of the food, it may be appropriate to propose the optimum effective intake rather than the minimum effective intake (Scenario II).
If it is appropriate to combine the results of the studies selected for inclusion, a crude dose-response curve can be developed using the data from these studies to identify an intake that corresponds to a biologically relevant magnitude of change in the health outcome of interest. For example, if a reduction of at least x % of LDL-cholesterol is considered biologically relevant, the amount of beta-glucan that causes this magnitude of change on the dose-response curve should be proposed as the minimum effective intake in an application for a cholesterol-lowering claim for beta-glucan.
Alternatively, the lowest amount of the food that caused a biologically relevant improvement in the health outcome of interest could be identified from the studies selected for inclusion. If this amount consistently leads to an improvement in the health outcome across studies, it can be proposed as the minimum effective intake. If this amount caused an improvement in one study only, and the amounts in the other studies that caused improvements were substantially higher, it might be preferable to propose a higher amount for which there are consistent results.
When there is evidence for a plateau in the effect, the optimum effective intake could be proposed. This amount is best determined from a human dose-response study that tests different levels of the food. When such a study (or studies) is available, the lowest intake that did not result in additional improvement in health outcome should be proposed as the optimum effective intake. The change in health outcome should be biologically relevant and the proposed optimum effective intake should be compared with other studies where only a single intake was used.
4.10 What sort of evidence is needed to show that the magnitude of the effect is physiologically meaningful/biologically relevant?
In Step 11 of the Evaluation of Claim Validity, a discussion on the physiological meaningfulness, or biological relevance, of the effect of the food in the included studies is required. A biologically relevant effect is defined as an important and meaningful change that has consequences for human health. While general standards do not exist to define the quantitative changes (such as the size of an effect) that would constitute biological relevance, accepted norms in the relevant research field should be considered and explained.
The biological change that would be considered relevant should be defined before the studies are initiated. The size of such changes should be used to design studies with sufficient statistical power to be able to detect effects of such size.
4.11 Where should information on intakes of the food in the target population come from?
In Step 12 of the Evaluation of Claim Validity, a discussion on the feasibility of consuming an effective amount of the food is required. This discussion should include estimates of current intake of the food, expected intake of the food if the health claim is approved, and potential changes in usual dietary patterns. Current intake information can come from a variety of sources, such as the Canadian Community Health Survey (CCHS). It can also come from research conducted by the petitioner. For example, expected intakes and potential changes in intake can be estimated using i) computer modelling or ii) surveillance data from other jurisdictions where a similar claim was accepted.
- Date modified: