Food packaging is manufactured from a range of materials, and bulk packaging material is often modified due to the use of substances such as adhesives, protective coatings and printing inks. Several thousand chemicals are used in the manufacture of food packaging and other chemicals that come into contact with food during its production and processing.
To gain an understanding of the risk posed by chemical migration from packaging into food, FSANZ has evaluated information on the hazard characteristics of chemicals used in the production of food packaging, as well as estimated dietary exposure to these chemicals due to migration into food. Use of the threshold of toxicological concern (TTC) concept has been valuable for this work, as well as a number of analytical surveys.
The TTC approach is a risk based screening tool that categorises chemicals into various threshold levels of safe expected exposure, depending on a chemical’s structure. The TTC threshold applicable to a specific chemical is compared to its estimated dietary exposure. Estimated dietary exposure below the TTC threshold indicates no safety concern, while exposure above the threshold indicates that appropriate toxicity data on the chemical, or a close structural relative, are needed in order to perform a safety assessment.
Analysis of a USFDA database of over 1300 food contact substances showed that for 86% of the substances estimated dietary exposure is below the lowest TTC for non-genotoxic substances. This figure increased to 97% taking account of a higher threshold value derived in a recent re-evaluation of the TTC for this class of compounds. For many of the chemicals with estimated dietary exposure exceeding their respective TTC thresholds, specific toxicity data were located in various databases and the published literature that support the safety of those chemicals. For some chemicals, supporting toxicity data may not be publicly available, or toxicity data on structurally related substances was used for safety assessment.
A conclusion of low risk based on the TTC analysis is supported by the results of a number of analytical surveys investigating the presence of packaging chemicals in Australian and New Zealand foods. These surveys have demonstrated that dietary exposures to chemicals migrating from packaging into food are low and below levels of concern.
The screening methodology used in the 24th Australian Total Diet Study (ATDS) to assess 30 food packaging chemicals in Australian foods and beverages identified two chemicals for which more data were needed to enable a more refined risk assessment. These chemicals were the phthalates diethylhexyl phthalate (DEHP) and diisononyl phthalate (DINP). However, a targeted follow up survey of DEHP and DINP plus five additional plasticisers in Australian foods found that estimated dietary exposures are below the tolerable daily intakes (TDIs) for these substances and not of concern for human health.
Potential risks from the migration of chemicals in recycled paperboard, particularly mineral oil hydrocarbons (MOH), into food are not yet well characterised and research is ongoing internationally. However, FSANZ has recently completed a survey of Australian food packaging and foods and did not find widespread migration of MOH into foods. MOH was only detected at quantifiable levels in two food samples, and there was no evidence of a public health and safety concern.
A recent survey of packaging chemicals including phthalates, printing inks and photoinitiators in New Zealand foods also found that dietary exposures to these chemicals are low and not of concern for human health.
The overall conclusion based on the available data is that the human health risk posed by chemical migration from packaging into food available in Australia and New Zealand is low.
Table of contents
Executive summary 1
1 Introduction 4
2 Risk assessment of food packaging chemicals 4
3 Summary of previous work 4
3.1 Threshold of Toxicological Concern analysis 4
3.2 Surveys of specific packaging chemicals 5
3.2.1 24th Australian Total Diet Study 5
3.2.2 Additional surveys of packaging chemicals 6
4 Phthalates and related plasticisers 6
4.1 Plasticisers survey 6
4.1.1 Introduction 6
4.1.2 Methods 6
4.1.3 Hazard summary 7
4.1.4 Summary of analytical survey results 8
4.2 Dietary exposure assessment of Plasticisers 11
4.2.1 Approach to estimating dietary exposures to plasticisers 11
4.2.2 Summary of dietary exposure assessment results 11
4.3 Risk characterisation and conclusion 15
5 Mineral oils 15
5.1 Mineral oil hydrocarbons survey 15
5.2.1 Introduction 15
5.2.2 Hazard summary 15
5.2.3 Methods 15
5.2.4 Summary of analytical survey results 16
5.2.5 Risk characterisation and conclusion 18
6 Survey of packaging chemicals in New Zealand foods 19
7 Conclusions 19
8 References 21
Appendix 1: Analytical results for Plasticisers 24
Appendix 2: Estimating dietary exposures for the Plasticisers Survey 41
How the dietary modelling was conducted 41
Number of respondents in each of the population groups assessed 41
Food consumption data 42
Construction of the model diet for 9 month old infants 42
Respondents versus consumers 43
Treatment of analytical values below the Limit of Reporting (LOR) 44
Food mapping 44
Food contribution calculations 45
Assumptions and limitations in dietary modelling 45
Appendix 3: Median plasticiser concentrations (mg/kg) in foods for dietary exposure assessment 47
Appendix 4: Estimated dietary exposures to plasticisers 52
Appendix 5: Major food contributors to plasticiser dietary exposures, derived using median concentrations 55
Appendix 6: Mapping 69
Food packaging is manufactured from a range of materials including glass, paper/paperboard, plastics and metals. Bulk packaging material is often modified due to the use of substances such as adhesives, protective coatings and printing inks. Several thousand chemicals are used in the manufacture of food packaging and other chemicals that come into contact with food during its production and processing. Chemicals used in the production of food contact materials include solvents, monomers, cross-linking agents, catalysts, plasticisers and antioxidants/stabilisers.
FSANZ has conducted a range of activities in order to gain an understanding of the risks posed by chemical migration from packaging into food (CMPF) for consumers in Australia and New Zealand.
This work includes analysis of information from United States Food and Drug Administration (USFDA) and European Food Safety Authority (EFSA) databases and publications as well as information in the published scientific literature. In addition, FSANZ has conducted a number of surveys investigating levels of packaging chemicals in foods and beverages available in Australia.
Much of this work has previously been reported by FSANZ, most recently in a Call for Submissions published in June 2016. Since that time FSANZ has conducted further survey work assessing the levels of plasticisers and mineral oils in Australian foods. In addition, the New Zealand Ministry for Primary Industries (MPI) has completed a survey of packaging chemicals in New Zealand foods.
This document provides a summary of these recent surveys, as well as an overview of FSANZ’s previous work on CMPF.
2 Risk assessment of food packaging chemicals
The scientific principles that apply to the risk assessment of other chemicals in food, such as food additives, contaminants and processing aids, also apply to food packaging chemicals. An overview of these risk assessment principles is presented in the FSANZ document Risk Analysis in Food Regulation (FSANZ 2013), while detailed information is provided in FAO/WHO (2009).
The core principle in food chemical risk assessment is that risk is a function of both the intrinsic hazard characteristics of the chemical (i.e. its toxicological properties) and dietary exposure to the chemical from consuming food and beverages.
3 Summary of previous work
FSANZ has evaluated the risks associated with chemical migration from packaging into food using a number of approaches and sources of information. These included a threshold of toxicological concern analysis of a United States Federal Drug Administration (USFDA) database of food contact substances, as well as FSANZ surveys of the Australian and New Zealand food supply. This work is briefly summarised below, while more detailed information can be found in Supporting Document 3 of the Call for Submissions for P1034.
3.1 Threshold of Toxicological Concern analysis
The Threshold of Toxicological Concern (TTC) approach is a risk-based tool for screening of chemicals for which only limited hazard data are available. It can be used to distinguish those chemicals with no appreciable human health risk from those for which further data are needed for risk assessment (Kroes et al. 2000; EFSA/WHO 2016).
The TTC approach allows chemicals to be categorised into various threshold levels of safe expected exposure, dependent upon structure. In general chemicals have been categorised into one of three structural classes (class I, II, III), assigned respective TTC values of 0.03, 0.009 and 0.0015 mg/kg bw/day.
In SD3 of the Call for Submissions for P1034, these TTC values were compared against a USFDA database1 containing dietary exposure information on more than 1300 food contact substances. The estimated dietary exposures for 86% of the substances (1119/1302) were below the lowest TTC value for non-genotoxic substances (structural class III: 0.0015 mg/kg bw/day). This figure increased to 97% (1260/1302) taking account of a higher threshold of 0.004 mg/kg bw/day for structural class III substances derived in a recent re-evaluation of the TTC for this class of compounds (Leeman et al. 2014).
For many of the chemicals with estimated dietary exposure exceeding their respective TTC thresholds, specific toxicity data were located in various databases and the published literature that support the safety of those chemicals. For a substantial fraction of chemicals in the database it is expected that the supporting toxicity data are unpublished, while for some chemicals the USFDA would have used read-across and/or quantitative structure-activity relationship (QSAR) approaches for safety assessment.
Overall, it was concluded that estimated dietary exposures to the majority of food contact substances are below the lowest TTC for non-genotoxic substances. This implies that repeat-dose toxicity data would not be required for the majority of substances to support a conclusion of negligible risk for the specific food contact use(s).
3.2 Surveys of specific packaging chemicals
A conclusion of low risk from the TTC analysis described above is supported by the findings of a number of analytical surveys investigating the presence of specific packaging chemicals in Australian foods. This information is summarised below, while more detailed information can be found in Supporting Document 3 of the call for submissions on the P1034 proposal.
3.2.1 24th Australian Total Diet Study
In 2016 FSANZ published the second phase of the 24th Australian Total Diet Study (ATDS) which investigated the presence of 30 packaging chemicals in a total of 81 typically consumed Australian food and beverages (FSANZ 2016). This survey took a screening approach to identify whether any of the detected chemicals were of potential health and safety concern that would require further investigation. This assessment was based on a conservative, worst case exposure scenario known as the Theoretical Maximum Daily Exposure (TMDE). The TMDE for each detected chemical was calculated based on an assumption that 50% of foods and beverages consumed contain the chemical, and that the concentration of the chemical is the maximum level detected in any food sample.
Packaging chemicals assessed in Phase 2 of the 24th ATDS included bisphenol A (BPA), epoxidised soybean oil (ESBO), di-2-ethylhexyl adipate (DEHA), 14 phthalates, two perfluorinated chemicals and 11 chemicals used in printing inks. These chemicals have a range of food packaging uses including plasticisers (plastic softeners), lid-sealing agents, moisture/oil-resistant coatings and labelling. The chemicals included in the study were selected based on consultation with internal and external stakeholders, reviews of Australian and international literature and data from previous international incidents.
For all but two of the detected chemicals the TMDEs were below internationally recognised safe levels, supporting a conclusion of negligible to low health risk.
The exceptions were the phthalates diethylhexyl phthalate (DEHP) and diisononyl phthalate (DINP). For these substances the TMDE calculated for screening purposes exceeded the EFSA tolerable daily intake (TDI2) for DEHP by a factor of 4 and the EFSA TDI for DINP by 9-fold. As a result of these findings FSANZ conducted a follow-up survey of a wider range of foods to allow better estimates of dietary exposure to DEHP and DINP that could be used for risk characterisation. A summary of the results of this survey can be found in Section 4 of this report.
3.2.2 Additional surveys of packaging chemicals
In addition to the 24th ATDS, FSANZ has previously conducted two targeted surveys on the presence of food packaging chemicals in Australian food and beverages. The first of these surveys focused on BPA (FSANZ 2010) while the second assessed levels of ESBO, phthalates, perfluorinated compounds, semicarbazide, acrylonitrile and vinyl chloride (FSANZ 2011). BPA and ESBO were detected at low levels in a small proportion of foods analysed, however no public health and safety risks were identified. None of the other chemicals were detected in any of the tested foods.
A recent survey of canned and bottled fruit products also found that levels of tin were well below the ML and no public health and safety concerns were identified (FSANZ 2015).
4 Phthalates and related plasticisers
4.1 Plasticisers survey
The plasticisers’ survey was conducted as a follow-up to Phase 2 of the 24th ATDS. The ATDS was undertaken as a screening study, which found that further survey work needed to be done on two phthalate plasticisers, DEHP and DINP, to enable a refined dietary exposure assessment to be conducted and to determine whether they posed any public health and safety concerns. In addition to DEHP and DINP, this survey examined the levels of several other phthalates, butyl benzyl phthalate (BBP), dibutyl phthalate (DBP), and diisodecyl phthalate (DIDP), the adipate di(2-ethylhexyl) adipate (DEHA), and the citrate acetyltributylcitrate (ATBC).
The survey was conducted as part of the Implementation Sub-Committee for Food Regulation’s (ISFR) Coordinated Food Survey Plan, which is developed to facilitate bi-nationally coordinated food surveillance activities to inform food regulation policy, standards development and compliance priorities.
Food sample purchasing, preparation and analysis were undertaken from April to June 2016 in accordance with detailed instructions outlined in a survey procedures manual.
A total of 65 foods and beverages were sampled from five Australian jurisdictions including the Australian Capital Territory, New South Wales, South Australia, Victoria and Western Australia. Foods sampled included: baked beans; biscuits and cake; canned tomatoes; crisps and confectionery; dry noodles; frozen and takeaway meals; infant foods and formulas; jam; meat products; milk; oils (olive and vegetable oils); olives; peanut butter; sugar; and various cereal products (e.g. breads, breakfast cereals and bars). Foods were sampled in a variety of different packaging types including various types of plastics (rigid and flexible), metal cans, cardboard (carton board and folding cartons), paper and glass.
The food sample list was developed to enable a robust estimate of dietary exposure for Australian consumers. The purchasing of food samples took place in each participating jurisdiction in April and May 2016. Foods were sampled from a range of different retail outlets representing general availability in the Australian market. The following factors were considered in developing the food list:
Ensuring coverage of a broad range of foods representative of typical packaged food and beverage consumption patterns for the Australian population
Known uses of plasticisers in food packaging
Results from phase 2 of the 24th ATDS
International reports and surveys investigating plasticiser migration into food.
Food sample purchases were sent to the National Association for Testing Authorities (NATA) accredited food preparation laboratory, the National Measurement Institute (NMI) in Melbourne, as soon as practicable after purchase. Before analysis, all samples were prepared to a table ready state, including cooking (where required) and removal of inedible portions. Perishable foods were all prepared within 48 hours of purchase. All other foods were prepared within a week of purchase.
In preparation for analysis, three individual samples for each food and jurisdiction were combined into a single composite sample for laboratory analysis (to produce one composite sample per jurisdiction). Composite samples were sent from NMI to Eurofins WEJ Contaminants in Hamburg, Germany. The extraction and clean-up of plasticisers was carried out with solid phase extraction cartridges. Quantitative determination was conducted using LC-ESI-MS/MS3 in positive ion mode via two MS/MS-transitions. This method is accredited by the regional Dutch authority4 and in accordance with International Organisation for Standardisation (ISO) standards. The plasticiser concentrations were calculated using internal standards. The Limit of Detection (LOD)5 and Limit of Quantitation (LOQ)6 for each of the seven plasticisers are listed in Table 1 below. The Limit of Reporting (LOR)7 for the results provided by the laboratory was equal to the LOD.
Table 1 – Limits of detection and quantitation for the plasticisers in the survey
4.1.3 Hazard summary
Some phthalates and adipates have been the subject of concern in regard to their potential for adverse reproductive and developmental effects, as shown in laboratory animal studies. TDI values have been established for all seven substances included in this survey, as summarised in Table 2.
As several phthalates have adverse effects on the male reproductive system, consideration was given to whether a combined risk assessment should be performed for these substances. However, male reproductive effects are not the most sensitive effect for some of the substances included in the survey, and the TDIs for each substance are based on different critical effects. Therefore it was not considered appropriate to conduct a combined risk assessment for the phthalates included in the survey.
Table 2 – TDIs for the plasticisers in the survey
TDI (mg/kg bw/day)
Testicular toxicity in rats
Liver and kidney effects in rats
Liver effects in dogs
Fetotoxicity in rats
Developmental effects in rats
Testicular toxicity and reduced anogenital distance in rats
Reduced body weight in rats
* NOAEL: No observed adverse effect level
4.1.4 Summary of analytical survey results
A summary of results for the plasticisers’ survey is provided in One or more plasticisers were detected above the LOQ in 47 out of 65 foods tested. As indicated in One or more plasticisers were detected above the LOQ in 47 out of 65 foods tested. As indicated in Table 3, DEHP was found in the highest number of foods, with 28 of 65 foods reported with concentrations above the LOQ. ATBC (23/65 foods), DEHA (17/65 foods) and DINP (11/65 foods) were also detected in a relatively high number of foods at concentrations above the LOQ. DBP (3/65 foods), DIDP (1/65 foods) and BBP (0/65 foods) were found in relatively few or no foods at concentrations above the LOQ. , DEHP was found in the highest number of foods, with 28 of 65 foods reported with concentrations above the LOQ. ATBC (23/65 foods), DEHA (17/65 foods) and DINP (11/65 foods) were also detected in a relatively high number of foods at concentrations above the LOQ. DBP (3/65 foods), DIDP (1/65 foods) and BBP (0/65 foods) were found in relatively few or no foods at concentrations above the LOQ. below. Further detailed results tables are included in Appendix 1.
One or more plasticisers were detected above the LOQ in 47 out of 65 foods tested. As indicated in One or more plasticisers were detected above the LOQ in 47 out of 65 foods tested. As indicated in One or more plasticisers were detected above the LOQ in 47 out of 65 foods tested. As indicated in Table 3, DEHP was found in the highest number of foods, with 28 of 65 foods reported with concentrations above the LOQ. ATBC (23/65 foods), DEHA (17/65 foods) and DINP (11/65 foods) were also detected in a relatively high number of foods at concentrations above the LOQ. DBP (3/65 foods), DIDP (1/65 foods) and BBP (0/65 foods) were found in relatively few or no foods at concentrations above the LOQ. , DEHP was found in the highest number of foods, with 28 of 65 foods reported with concentrations above the LOQ. ATBC (23/65 foods), DEHA (17/65 foods) and DINP (11/65 foods) were also detected in a relatively high number of foods at concentrations above the LOQ. DBP (3/65 foods), DIDP (1/65 foods) and BBP (0/65 foods) were found in relatively few or no foods at concentrations above the LOQ. , DEHP was found in the highest number of foods, with 28 of 65 foods reported with concentrations above the LOQ. ATBC (23/65 foods), DEHA (17/65 foods) and DINP (11/65 foods) were also detected in a relatively high number of foods at concentrations above the LOQ. DBP (3/65 foods), DIDP (1/65 foods) and BBP (0/65 foods) were found in relatively few or no foods at concentrations above the LOQ.
The highest concentrations for individual plasticisers were reported in fatty foods including sundried tomatoes (DEHP up to 78 mg/kg and ATBC up to 30 mg/kg), hamburgers (DINP up to 13 mg/kg) and olives (ATBC up to 9.2 mg/kg). These results are not unexpected, as phthalates are known to be fat soluble, lipophilic compounds (Cao, 2010). It should be noted that, generally, these higher concentrations were limited to a relatively small proportion of all samples. Overall, the survey results indicate that the levels of these seven plasticisers in a broad range of Australian foods were generally low, with the majority of individual measured results (91%) below the LOQ.
Table 3 – Summary results for plasticisers