Health Canada's Revised Exposure Assessment of Acrylamide in Food

August 2012

Table of Contents

• Data Collected as part of the Acrylamide Monitoring Program
• Dietary Exposure Assessment
  • Table 1 - Probabilistic dietary exposure estimates for acrylamide for select age-gender groups
  • Contribution of Different Foods to Acrylamide Exposure
  • Table 2 - Top contributors to average total dietary acrylamide exposure
• Risk Characterization
  • Table 3 - Margins of exposure (MOE) between dietary exposure to acrylamide and a NOAEL of 200 µg/kg bw per day and a BMDL₁₀ of 180 µg/kg bw per day
  • Uncertainties
• Next Steps
• References
• Appendix 1 - Concentrations of acrylamide in food products, on an as-consumed basis, available in Canada (µg/kg of food or parts per billion (ppb))

Data Collected as part of the Acrylamide Monitoring Program

In 2009, Health Canada initiated a broader acrylamide monitoring program to provide a more up to date and accurate estimate of dietary exposure to acrylamide in Canada, assess the potential risk from acrylamide exposure and monitor the effectiveness of any reduction strategies which may have been implemented by industry. Health Canada completed phase 1 of the program in 2009 which included acrylamide testing results in bread (toasted and untoasted), coffee, French fries, cereals, snack foods, baby foods and many other food commodities. In 2010, Health Canada completed phase 2 of the monitoring program, which was designed to include samples from more food brands, capture additional generic brand names and include more samples from commodities that have shown highly variable acrylamide levels.

This new data has enabled Health Canada to set a "new baseline" for occurrence of acrylamide in various foods and serves as a basis upon which its assessment of dietary exposure to acrylamide can be based and updated regularly.

The cumulative results from phases 1 and 2 of Health Canada's acrylamide monitoring program are included in Appendix 1. Note that the data presented in Appendix 1 represent only a certain proportion of the products that are available on the Canadian market and do not represent all brands and product types. No particular inference should be drawn from the presence or absence of any brand from this survey.

The data in Appendix 1 are presented as the levels of acrylamide in the food when the food is in a prepared, ready-to-consume form. Where possible, foods were prepared according to manufacturers' specific instructions concerning cooking times, cooking temperatures, dilution factors, etc. Acrylamide concentrations in instant coffee, grain beverage, cocoa, infant formula and dry cereal samples were measured in the dry powder. However, acrylamide levels for instant coffee and grain beverage samples are presented in Appendix 1 after applying a dilution factor based on the amounts of powder and water used to prepare the beverage.  Toasted bread was prepared with a cafeteria belt toaster. Light, medium, and dark toasting levels were determined empirically and correspond to the settings used to generate the respective levels of browning.

Dietary Exposure Assessment

Prior to the establishment of the acrylamide monitoring program, Health Canada had conducted a limited number of small surveys to determine the levels of acrylamide in certain food commodities. Using this data, a preliminary exposure assessment was conducted. The mean adult exposure to acrylamide from food was estimated to be between 0.3 and 0.4 micrograms per kilogram body weight per day (µg/kg bw per day). The preliminary assessment used mean acrylamide concentration values for each commodity and mean food consumption figures to calculate a single exposure value for the population, rather than a range of values, in what is called a deterministic assessment. The results of Health Canada's preliminary deterministic exposure assessment are similar to other deterministic dietary acrylamide exposure estimates conducted in other countries such as Sweden and the United States.

Using the much larger database of acrylamide occurrence data in foods collected in Canada as part of phases 1 and 2 of the acrylamide monitoring program and food consumption figures from the Canadian Community Health Survey (CCHS) - Cycle 2.2 on Nutrition (Statistics Canada, 2004), Health Canada used computer modeling to conduct a probabilistic dietary exposure assessment for acrylamide. A probabilistic exposure assessment combines all of the individual contaminant concentrations measured in samples of each particular food along with individual consumption data from the CCHS for the same foods. Specifically, an acrylamide concentration for each food listed in Appendix 1 was randomly applied to the relevant food consumption rates reported by each individual respondent in the CCHS. The result is a distribution of possible dietary exposures to the substance in question. Note that if acrylamide levels were reported as less than the analytical instrument's limit of detection (LOD), the level of acrylamide in the food was conservatively set to the LOD.

Whenever possible, market share data obtained from market research results collected in 2009 by the AC Nielsen Company of Canada was factored into the assessment. If more than one brand of a certain type of food was sampled, the proportion of each brand's share of the market (based on volume of sales) was used to determine the probability of a random selection of an acrylamide concentration from that particular brand and type of food. For example, if cookies of brand X represent 25 percent of the market share for all cookies, the selection of a random acrylamide level in cookies would be taken from the results found in brand X 25 percent of the time. Market share data was only used to estimate exposure when the Health Canada acrylamide occurrence data for a particular food was available for brands that collectively represent the majority of the market. This was the case for chocolate chip cookies, chocolate cream cookies, ginger cookies, prune juice, instant coffee, frozen French fries, arrowroot cookies, peanut butter, corn chips, and potato chips.

For each respondent in the CCHS survey, the estimated exposure to acrylamide from each food was summed to give an estimate of the total dietary acrylamide intake. The result is a distribution of total dietary exposure to acrylamide within a population. Five-hundred different exposure scenarios were generated for each individual (five-hundred iterations of the randomized assignment of acrylamide levels), from which mean and 90^th percentile exposure estimates were calculated. The median value of the mean exposure estimates and the median value of the 90^th percentile exposure estimates calculated for all individuals within a certain age group are reported in Table 1.

Results of the probabilistic exposure assessment show that children (from 1 to 8 years of age) ingest higher amounts of acrylamide through the diet, on a body weight basis, than other age groups. Mean acrylamide intakes range from 0.356 to 0.609 µg/kg bw per day for age groups from 1 to 18 years and from 0.157 to 0.288 µg/kg bw per day for adults (19 to 71+ years). Mean adult exposure estimates calculated by probabilistic methods are slightly lower than the results from Health Canada's preliminary exposure assessment. Exposure among infants < 1 year of age was lower than that of children and within the range of adult exposures.

Contribution of Different Foods to Acrylamide Exposure

Foods contributing most to the average total dietary acrylamide exposure for the general population above the age of 1 year are provided in Table 2. The results shown here are similar to those found in the preliminary deterministic exposure assessment which utilized data collected prior to the launch of the acrylamide monitoring program.

Oven-baked French fries prepared from frozen products purchased at retail were found to contribute very little to acrylamide intake, accounting for approximately 1 percent of the overall dietary exposure for the total population. This is significantly lower than the contribution from restaurant French fries, generally prepared by deep frying, primarily due to the lower frequency of consumption of oven-baked French fries prepared at home from frozen products as reported by CCHS participants. Lower acrylamide levels in oven-baked frozen French fries prepared at home and the weight applied from market share data were also contributing factors.

Risk Characterization

In February 2010, the Joint Food and Agriculture Organization/World Health Organization Expert Committee on Food Additives (JECFA) reviewed the dietary exposure and toxicological data for acrylamide (JECFA, 2010). JECFA calculated Margins Of Exposure (MOE) using: (1) a No Observed Adverse Effect Level (NOAEL) of 200 µg/kg of body weight per day based on morphological changes in nerves in rats; (2) a Benchmark Dose Lower Confidence Limit (BMDL) associated with a 10% response (BMDL₁₀) rate for mammary tumours in rats of 310 µg/kg bw per day; and (3) a BMDL₁₀ rate for Harderian gland tumours in male mice of 180 µg/kg bw per day.

MOE values were calculated by comparing the NOAEL and BMDL₁₀ values to the dietary exposure estimate for acrylamide according to the following equation:

In general, the larger the MOE value, the lower the safety concern. However, the interpretation of MOE values in the context of an associated potential risk to human health is based on considerations of the chemical in question, the nature of the hazard it poses, any scientific uncertainties, and the expert judgment of the risk assessors.

The JECFA assessment estimated a mean dietary exposure to acrylamide of 1 µg/kg bw per day for the general population and a high exposure level of 4 µg/kg bw per day using a single point estimate approach. MOE values for mean and high exposure consumers using the most sensitive non-carcinogenic end-point (a NOAEL) were 200 and 50, respectively. MOEs using the lower BMDL₁₀ value for induction of tumours in mice were 180 for mean acrylamide exposures and 45 for high-end exposures. JECFA concluded that for a compound that is both genotoxic and carcinogenic, these MOEs for acrylamide indicate a human health concern.

The MOEs calculated in this assessment for various age groups using the mean and 90^th percentile dietary exposure estimates, the NOAEL and the lowest BMDL₁₀ value considered by JECFA, are provided in Table 3.

MOE values from mean and high-end exposure to acrylamide were lowest among young children (from 1 to 8 years of age), and highest for the adult age groups (above 19 years) and infants under 1 year. Overall, dietary exposure to acrylamide in Canada is estimated to be lower than that reported by JECFA which results in MOEs that are higher than those estimated by JECFA. Nonetheless, Health Canada concurs with the opinion of JECFA that dietary exposure to acrylamide represents a potential human health concern.

Uncertainties

Exposure estimates were derived based on the potential ingestion of foods that were included in phases 1 and 2 of the survey. Surveys conducted by other countries have found very low levels of acrylamide in some foods that were not included in this assessment, such as seafood and dairy products. Although the available data indicates that these foods would not significantly contribute to the total dietary exposure to acrylamide, exposure may have been underestimated in this assessment as not all foods that may contain acrylamide were included. Health Canada will continue to update its estimate of the occurrence of acrylamide in a wide range of foods consumed by Canadians along with the corresponding exposure estimates.

Next Steps

Health Canada recently met with members of the food industry and other stakeholders to provide an update of its assessment on acrylamide in food. Health Canada continues to encourage the food industry to further pursue reduction efforts for acrylamide in processed foods. Acrylamide reduction strategies are available through the  Codex Alimentarius Code of Practice for the Reduction of Acrylamide in Foods (CAC/RCP 67-2009) and the  Food Drink Europe Acrylamide Toolbox 2011. Health Canada is aware that certain segments of the Canadian food industry have implemented measures to reduce acrylamide in their products and continue to conduct research on new methods to further minimize acrylamide occurrence.

On March 14, 2012 Health Canada amended the Food and Drug Regulations to permit the use of asparaginase in certain food products. Asparaginase is an enzyme that will reduce the amount of the amino acid asparagine, which is naturally present in certain foods. Under specific cooking conditions and for certain foods, asparagine can react with other natural components of the food and form acrylamide. Reducing the amount of asparagine in these foods will reduce the amount of acrylamide formed during cooking.

Health Canada also continues to work with health authorities in other countries to better understand how acrylamide is formed in various foods, what foods contain the highest amounts of acrylamide, and what impact acrylamide has on human health. Health Canada plans to continue to collect information on the levels of acrylamide in foods that have not been sampled to date by Health Canada and/or in foods that potentially contain elevated levels of acrylamide. Future data collection endeavors will also serve to determine potential trends in acrylamide occurrence in order to assess the impact of the food industry's efforts in applying reduction strategies.

Health Canada also recommends practicing some simple techniques to minimize acrylamide formation in foods prepared at home and to consume a variety of foods according to Eating Well with Canada's Food Guide.

References

Joint FAO/WHO Expert Committee on Food Additives (JECFA) 2010.  Summary and conclusions, seventy-second meeting, Rome, 16-25 February 2010.

Statistics Canada 2004.  Canadian Community Health Survey - Cycle 2.2 Nutrition (CCHS). Detailed information for 2004 (Cycle 2.2), Ottawa (ON): Statistics Canada.

Appendix 1 - Concentrations of acrylamide in food products, on an as-consumed basis, available in Canada (µg/kg of food or parts per billion (ppb)).

This table is considered a snapshot of a small proportion of the products that are available on the Canadian market and do not represent all brands and product types. No particular inference should be drawn from the presence or absence of any brand from this survey. The absence of a particular brand simply means that it was not sampled and does not mean that it would have either similar or dissimilar levels of acrylamide as those in brands of the same food listed below.  Differences in acrylamide levels between foods and brands do not necessarily reflect differences in exposure or potential risk that would be faced by consumers, and should not be used as indices of healthy product choices.