EUROPEAN COMMISSION

Brussels, 14.10.2020

SWD(2020) 251 final

COMMISSION STAFF WORKING DOCUMENT

FITNESS CHECK

on endocrine disruptors

Accompanying the document

COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

Chemicals Strategy for Sustainability



Towards a Toxic-Free Environment

{COM(2020) 667 final} - {SWD(2020) 225 final} - {SWD(2020) 247 final} - {SWD(2020) 248 final} - {SWD(2020) 249 final} - {SWD(2020) 250 final}

Table of contents

1 INTRODUCTION    

1.1    Background to the Fitness Check    

1.2    Objectives of the Fitness Check    

2    METHODOLOGY    

2.1    Scope of the Fitness Check    

2.2    Fitness Check questions and baseline    

2.3    Mapping of of regulatory provisions and review of key documents    

2.4    Case studies    

2.5    Consultation activities    

3    STATE OF PLAY    

4    COHERENCE    

4.1    Data requirements    

4.2    Hazard identification    

4.3    Risk assessment    

4.4    Risk management    

4.5    Coherence with non-EU jurisdictions    

5    EFFECTIVENESS    

5.1    Current process of ED identification    

5.2    Biocidal Products and Plant Protection Products Regulations    

5.3    REACH    

5.4    Other sector-specific and product-specific legislation    

5.5    Communication    

5.6    Monitoring data    

5.7    Human health and environmental impact evidence and indicators    

5.8    Market surveillance    

5.9    Functioning of the internal market, competitiveness and innovation    

6    EFFICIENCY    

7    RELEVANCE    

7.1    Relevance to citizens and stakeholders    

7.2    Ability to adapt to the evolving societal and economic context    

7.3    Ability to adapt to scientific progress    

8    EU ADDED VALUE    

8.1    Joining forces and efficient use of resources    

8.2    Preserving the functioning of the single market    

8.3    Legislative measures alignment between Europe and global regimen    

9    CONCLUSIONS    

10    REFERENCES    

11    GLOSSARY    

Annex 1. Procedural information    

1    Lead DG and internal references    

2    Organisation and timing    

3    Exceptions to the better regulation guidelines    

4    Consultation of the Regulatory Scrutiny Board    

5    Evidence, sources and quality    

Annex 2. Synopsis report: stakeholder consultation activities    

1    Summary    

2    Introduction and approach    

3    Stakeholder groups covered by the consultation activities    

4    Outcome of the public consultation    

4.1    Respondents    

4.2    How well informed do citizens feel?    

4.3    Public views on effectiveness    

4.4    Public views on efficiency    

4.5    Summary of public views on regulatory testing and animal welfare    

4.6    Public views on EU-added value    

4.7    Public views on relevance    

4.8    Additional comments made by citizens    

5    Outcome of the stakeholder consultation    

5.1    Structure of survey    

5.2    Respondents    

5.3    Familiarity of stakeholders with relevant legislation    

5.4    Stakeholder views on coherence (overall statistics)    

5.5    Stakeholder views on effectiveness (overall statistics)    

5.6    Stakeholder views on efficiency (overall statistics)    

5.7    Stakeholder views on relevance (overall statistics)    

5.8    Stakeholder views on EU-added value (overall statistics)    

5.9    Breakdown of stakeholder views according to stakeholder group    

6    Outcome of the SME consultation    

6.1    Respondents    

6.2    Legislation affecting respondents    

6.3    Information in the company about endocrine disruptors    

6.4    SME views on coherence    

6.5    SME views on effectiveness    

6.6    SME views on efficiency    

6.7    SME views on EU added value    

6.8    Additional comments made by SMEs    

Annex 3. Methodology    

1    Summary    

2    Workflow    

3    Preliminary analysis of current situation    

3.1    Questionnaire to policy DGs    

3.2    Identification of Regulatory Reviews/Better Regulation activities    

3.3    Definition of scope    

3.4    Formulation of Fitness Check questions    

4    Extraction and review of available evidence    

4.1    Mapping of provisions    

4.2    Policy and scientific sources of evidence    

4.3    Evidence base and limitations    

4.4    Case studies    

4.5    Evidence base and limitations    

4.6    Economic case studies    

5    Consultation activities    

5.1    External consultation activities    

5.2    Evidence base and limitations    

Annex 4. Mapping of provisions and regulatory processes for identifying and managing Endocrine Disruptors    

1    Biocidal Products    

2    Plant Protection Products and residues    

3    REACH    

4    Classification, Labelling and Packaging (CLP)    

5    POPs Regulation    

6    Toys    

7    Food legislation    

7.1    Food Contact Materials    

7.2    Contaminants in feed and food    

7.3    Food additives    

8    Cosmetic Products    

9    Medical devices and in vitro diagnostics    

10    Human and veterinary medicines    

11    Occupational Safety and Health legislation (OSH)    

12    Water    

13    Chemical-product-waste interface    

14    Other in-scope legislation    

14.1    Detergents    

14.2    Fertilising Products    

14.3    Ecolabel Regulation    

14.4    General Product Safety Directive    

14.5    Seveso III Directive and Industrial Emission Directive    

14.6    Directive on air quality    

14.7    WTO rules    

15    References    

Annex 5A. Case Study: 3-benzylidene camphor (3-BC)    

1    Case study factsheet: overview of concerns for human health and environment and main regulatory intervention    

2    Substance profile    

2.1    Exposure pathways of concern    

2.2    Effects on human health and the environment    

3    3-BC and structurally similar UV filters in EU legislation    

3.1    Regulating 3-BC in cosmetics legislation    

3.2    Regulating 3-BC under REACH    

3.3    Status of structurally similar UV filters    

4    Comments on fitness check evaluation criteria    

5    References    

Annex 5B. Case Study: bis(2-ethylhexyl) phthalate (DEHP)    

1    Case study factsheet: overview of concerns for human health and environment and main regulatory intervention    

2    Substance profile    

2.1    Exposure pathways of concern    

2.2    Hazard endpoints of concern    

3    DEHP and other phthalates in EU legislation    

3.1    Regulating phthalates under REACH and preceding legislation    

3.2    Enforcement of regulatory measures under REACH    

3.3    Regulating phthalates under food contact materials (FCM)    

3.4    Regulating DEHP in cosmetics    

3.5    Regulating DEHP in medical devices    

3.6    Regulating DEHP at the interface of chemical and waste legislation    

3.7    Regulating DEHP under the Water Framework Directive    

4    Impact of regulation    

4.1    Decline in use and exposure    

4.2    Shift to alternatives    

5    Comments on fitness check evaluation criteria    

6    References    

Annex 5C. Case Study: nonylphenol    

1    Case study factsheet: overview of concerns for human health and environment and main regulatory intervention    

2    Substance profile    

2.1    Exposure pathways of concern    

2.2    Effects on human health and the environment    

3    Regulating nonylphenols in EU legislation    

3.1    Regulating nonylphenols in REACH    

3.2    Regulating nonylphenols under other chemical legislation    

3.3    Regulating nonylphenols under the Water Framework Directive    

4    Impact of regulation    

4.1    Reduced exposure    

4.2    Shift to alternatives    

5    Comments on fitness check evaluation criteria    

6    References    

ABBREVIATIONS

1 INTRODUCTION

1.1Background to the Fitness Check

Endocrine Disruptors (EDs) are chemical substances that can alter the functioning of the endocrine (hormonal) system and negatively affect the health of humans or animals. The EU response to EDs has developed over the past twenty years, in line with the actions outlined in the Community Strategy for Endocrine Disruptors 1 . Significant progress in understanding these substances has been made since then.

Overviews of the state of the science on EDs have been published by various bodies. Notably, in 2012, the United Nations Environment Programme (UNEP) and the World Health Organization (WHO) published an update (WHO/UNEP, 2012) of the 2002 International Programme on Chemical Safety (IPCS) report (IPCS, 2002). This report notes the increase in trends of many endocrine-related disorders in humans, and associations between exposure to EDs and some diseases. It also acknowledges the multi-causal nature of the diseases and the resulting difficulty in attributing a given disease to a single factor, such as chemical exposure. The report also emphasises the existence of critical windows of exposure (such as foetal development and puberty) during which exposure can lead to irreversible effects, some being detected only many years later. Furthermore, the report gives examples of wildlife populations recovering after decreased exposure to EDs. Finally, the need to develop appropriate testing methods is highlighted.

Endocrine disruption is a relatively new consideration in toxicological or eco-toxicological assessment, whereby the focus is on the way in which a chemical disrupts the development and functioning of humans or organisms (i.e. through disturbing hormonal balance) eventually leading to an adverse effect (or effects) on function that arises as a consequence of the disturbance, where the nature of the adverse effects may vary according to the timing of the exposure and the species exposed.

For some chemicals, adverse effects, such as cancers in hormone-sensitive organs, or reproductive and developmental disorders, may already be known, and the chemicals already flagged as substances of concern (e.g. through classification as hazardous under the EU Classification Labelling and Packaging (CLP) Regulation 2 ), have been banned or restricted on the basis of these adverse effects.

The significance of non-monotonic dose–response relationships and low-dose effects of endocrine disruptors has been widely debated and there is still no consensus in the scientific community on whether and how certain toxicological principles such as the ‘safe threshold’, (i.e. the dose below which no adverse effect is expected to occur) are applicable in assessing the safety of EDs. This lack of agreement presents particular difficulties at the policy level in deciding how risk assessments should be conducted for these chemicals and whether to follow a generic risk approach or a specific risk approach 3  (Solecki et al., 2017).

A variety of EU regulatory measures address the risks from exposure to hazardous chemicals. Collectively, these measures aim to ensure a high level of protection of human health and the environment, while ensuring the smooth functioning of the internal market. Some pieces of legislation have specific provisions for EDs (i.e. chemicals in general (REACH), plant protection products, biocidal products, water and medical devices). For other pieces of legislation, with no ED-specific provisions, there are general provisions for the risk management of hazardous substances which may be used to control the risks posed by EDs (once they are identified).

EU action on EDs relies largely on work undertaken by the EU scientific assessment bodies, such as the independent scientific committees (on Consumer Safety – SCCS – and on Health, Environmental and Emerging Risks – SCHEER), the European Chemicals Agency (ECHA) and the European Food Safety Authority (EFSA). Following the hazard or risk assessments conducted by these bodies, the Commission takes the final decision on risk management measures, together with the Member States and the other European Institutions (Council and Parliament) as appropriate. When scientific evaluation cannot conclude with sufficient certainty on a risk, the Commission is guided by the precautionary principle to take protective measures for its citizens and/or the environment.

For plant protection products and biocidal products, the co-legislators decided – as for other hazard properties of particular concern (such as substances classified as carcinogenic, mutagenic, or toxic to reproduction, categories 1A or 1B) that once it is proven that a substance is an endocrine disruptor, the substance in principle cannot be authorised for use; while there are very limited derogation possibilities. For biocides, the ban is only applicable in relation to ED properties for human health and not for the environment.

Criteria for identifying endocrine disruptors have been established under the legislation on plant protection products 4 and biocidal products 5 . These criteria were informed by various EU initiatives. These included the work of a Endocrine Disrupters Expert Advisory Group (Munn and Goumenou, 2013a, b) and of the EFSA Scientific Committee (EFSA Scientific Committee, 2013), as well as an impact assessment of different options for the criteria 6 . The criteria are based on the 2002 IPCS/WHO definition (IPCS, 2002), according to which an endocrine disruptor is “an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub)populations”. In practical terms, this means that the identification of a substance as an ED should be based on an adverse effect for which: a) there is convincing evidence of a biologically plausible causal link to an endocrine disrupting mode of action; and b) disruption of the endocrine system is not a secondary consequence of other non-endocrine-mediated systemic toxicity.

EU legislation in fields other than plant protection products and biocidal products does not contain such criteria. It has been argued by stakeholders that the same criteria should apply across the relevant pieces of EU chemicals legislation for reasons of legal certainty and in order to avoid the possibility that a substance is identified as an ED under one piece of legislation and not under another.

Different regulatory approaches exist in different pieces of legislation for substances that have endocrine disrupting properties. According to some pieces of legislation, these substances cannot in principle be authorised for use, unless limited derogation possibilities apply. Under other pieces of legislation, chemicals with endocrine disrupting properties may be authorised for a limited period or a limited range of uses, or under restricted conditions.

Other pieces of legislation do not mention EDs specifically, however, substances with such properties can be subject to case-by-case regulatory action based on the general requirements of the legislation.

Efforts to identify, assess and manage risk from EDs under REACH have been underway for some time and two EDs (ethoxylates of octylphenol and nonylphenol) have been placed on the list of substances (Annex XIV) requiring specific authorisations to be on the market. A further 15 substances have been identified as substances of very high concern (SVHCs) with endocrine disrupting properties and are included in the Candidate list of substances for possible inclusion in the authorisation list in the future; another 80 or more are currently being evaluated due to concern about their potential endocrine disrupting properties. Substances with ED properties can also be subject to restrictions, e.g. four phthalates at a sum level of 0.1% or higher in a variety of products 7 . Time has been relatively short since the criteria to assess active substances used in biocidal products and plant protection products came into force (2018). Nevertheless, three biocidal active substances and three active substances in plant protection products (PPP) have already been identified as EDs for both human health and the environment and a further four PPP active substances identified as EDs for human health 8 . Table 1.1 provides a list of the substances identified as SVHCs with endocrine disrupting properties under REACH or identified as EDs under the Biocidal Products Regulation as of September 2020.

Table 1.1 Substances identified as endocrine disruptors at EU level 9

Under the legislation on water, the Commission has included several endocrine disruptors in the list of “priority substances” of particular concern in surface waters, for which environmental quality standards and emission controls apply. Furthermore, the Commission has included several demonstrated or suspected endocrine disruptors in the "watch list" of substances for which Union-wide monitoring data should be gathered.

Due to possible concerns of endocrine disruption, the chemical bisphenol A was banned from use in baby bottles and other containers for foods for infants and young children on the basis of the precautionary principle, and very low migration limits are set for other food contact materials. Bisphenol A is also subject to limit values in toys for young children under 36 months or intended to be placed in the mouth and paper used in receipts. An occupational exposure limit value for bisphenol A has also been set to protect workers from exposure through inhalable dust.

On 7 November 2018, the Commission adopted its Communication “Towards a comprehensive EU framework on endocrine disruptors” 11 , updating the Strategy of 1999. The Communication confirms the Commission's commitment to protect EU citizens and the environment from EDs. It outlines actions to step up the EU approach in order to further progress and maintain the expected high level of protection. In particular, to address the concerns on the coherence of the EU legal framework, the Communication announced the launch of this cross-cutting Fitness Check on EDs. While evaluations or Fitness Checks of relevance for EDs had already been carried out or are under way, a systematic analysis of the coherence of relevant provisions on EDs across the EU regulatory measures had not yet been completed. This was therefore a particular focus of this Fitness Check.

1.2Objectives of the Fitness Check

EU regulatory measures that address the risks from exposure to endocrine disruptors (EDs) have been developed at different points in time and have, in certain cases, different specific objectives. This has resulted in different approaches for managing endocrine disruptors, depending on the sector being regulated. Collectively, these measures aim to ensure a high level of protection of human health and the environment, while ensuring the smooth functioning of the internal market. While regulatory measures have allowed protective action to be taken against these substances (e.g. by introducing bans or restrictions on their use), questions have been raised by stakeholders on the overall coherence of the EU legal framework in relation to EDs.

To address the concerns on the coherence of the EU legal framework, the Commission launched a cross-cutting Fitness Check on EDs 12 , with the objectives to assess:

1)the coherence of the relevant EU legislation, including an analysis of how different provisions in different legal instruments interact, identifying potential gaps or inconsistencies.

2)whether EU chemicals legislation is effective in delivering its objective to protect human health and the environment by minimising the overall exposure to endocrine disruptors.

3)to the extent possible, the efficiency, relevance and EU added-value of the relevant EU chemicals legislation.

In addressing these objectives, particular attention is paid to legislation that does not contain specific provisions for endocrine disruptors, such as the legislation on toys, cosmetics and food contact materials. Another focal point is on whether the different pieces of legislation take into account the protection of vulnerable population groups that are particularly sensitive to endocrine disruptors when assessing and regulating such substances. The international dimension is also assessed, taking into account the impact of EU provisions on trade flows within and outside the EU. The findings of this Fitness Check will feed into the forthcoming Chemicals Strategy on Sustainability.

2METHODOLOGY

2.1Scope of the Fitness Check

The horizontal nature of the Fitness Check requires a broad scope, covering many legal acts addressing chemicals that are relevant to human health and the environment. For much of the chemicals acquis, Fitness Checks and other 'Better Regulation' activities addressing specific pieces of legislation had been completed or were ongoing.

On the basis of the mandate of the Fitness Check described in the Commission Communication “Towards a comprehensive EU framework on endocrine disruptors”, in-scope legislation was defined as legislation including provisions for the identification, hazard assessment, risk assessment or risk management of chemicals with endocrine disrupting properties for human health and the environment. The legislation considered in scope is listed in Table 2.1.

Table 2.1 Legislation in scope for this Fitness Check

*EU Pharmaceuticals Strategy is under development

2.2Fitness Check questions and baseline

The evaluation questions placed emphasis on coherence issues across the different in-scope regulatory instruments in Table 2.1. To apply the Fitness Check questions, it was necessary to establish a baseline for comparison. When a temporal reference needs to be established, the baseline is the situation with no specific provisions in EU legislation, corresponding to the year of adoption of the first strategy on endocrine disruptors (1999). The evaluation questions relate to the specific policy interventions or non-interventions across the different regulatory sectors from 1999 to date (e.g. implementation of REACH or the establishment of ED criteria for plant protection products and biocidal products). Other points of reference, such as non-EU jurisdictions, were considered when appropriate.

2.3Mapping of of regulatory provisions and review of key documents

Annex 3 describes the approach taken to map regulatory provisions related to the identification or risk management of endocrine disruptors, and to review key policy and scientific documents.

2.4Case studies

Case studies were developed for specific chemicals that are regulated under different and interconnected legislative instruments, with the aim to provide representative examples illustrating how EDs are identified, assessed and managed across legislation for their effects on human health and the environment. Substances selected for the case studies:

·3-Benzylidene camphor (3-BC; under assessment under REACH and the Cosmetic Products Regulation; endocrine disruptor with human health and environmental concern)

·Diethylhexyl phthalate (DEHP; regulated under several pieces of legislation: REACH, cosmetic products, medical devices, food contact materials, toys, Water Framework Directive (WFD); endocrine disruptor with human health and environmental concern)

·Nonylphenol (NP; regulated under REACH (covering many sectors: detergents, biocides, plant protection products, cosmetic products, medical devices), WFD; endocrine disruptor with environmental concern)

Two economic case studies were also carried out to explore possible associations between the implementation of risk management measures on selected chemicals and trade flows, both within and between the EU and non-EU countries (Canzian et al., 2020).

The main criterion for selecting case studies was that the substances should be relevant to more than one policy area, thereby enabling an analysis of the coherence of the assessment and risk management approaches applied. DEHP, 3-BC and nonylphenols and their ethoxylates are regulated by various pieces of in-scope legislation, including the ones specifically mentioned in the Fitness Check mandate. DEHP and nonylphenols are representative of two main groups of substances which have been identified as SVHCs with endocrine disrupting properties (phthalates, alkylphenols) and for which risk management measures have been applied. Thus, the case studies were chosen to be as representative and informative as possible, from the pool of SVHCs with endocrine disrupting properties that were identified at the time of selection. (see Table 1.1).

2.5Consultation activities

2.5.1External consultation activities

Consultation activities were carried out to gather inputs from a broad range of stakeholder groups as well as citizens to ensure that views from all interested parties were considered in the evaluation. The consultation activities relied primarily on the results of three surveys aimed at the general public, stakeholders and SMEs. The results were used as an additional source of evidence, recognising that they are associated with an inevitable bias towards those who have an interest in responding. Further information on the approach and findings is given in Annex 2.

3STATE OF PLAY

Substances with endocrine disrupting properties can be present in a wide range of materials and products. Some are incorporated in formulations intended for direct applications to humans (e.g. cosmetic products) or in the environment (e.g. plant protection products). Some are directly released upon product application (cosmetic products, plant protection products). Others are incorporated in materials and may be released over longer time scales along the product life cycle, from manufacturing to waste and possibly from recycled material streams. The system of EU regulation dealing with chemicals aims at minimising exposure of humans and the environment to EDs from different sources and exposure pathways. Figure 3.1 illustrates at which stage of a product’s life cycle, from manufacturing to waste, specific legislation is intended to act. Overall, risk management provisions (intervention logic) aim to strike a balance between human health and environmental protection, the societal benefits of the use of chemicals and the smooth functioning of the EU single market.

Figure 3.1. EU legislation (not exhaustive) mapped against pathways of endocrine disruptors’ exposure to humans and the environment 13

Annex 4 provides a detailed analysis of regulatory provisions and processes in place relevant to the control of endocrine disruptors for each piece of EU legislation in the scope of this Fitness Check. It also provides an overview of international legal obligations (WTO). Many pieces of legislation regulate the placing on the market of chemicals, and materials and products containing them (horizontal list in the centre of Figure 3.1). Their scope is complementary with respect to human and environmental protection, with the REACH regulation setting the baseline for general chemicals (Figure 3.1). These regulations and directives form the core of EU chemical legislation, together with the Classification, Labelling and Packaging (CLP) Regulation. Table 3.1 summarises the intervention logic of each piece of ‘in scope’ legislation and highlights the different approaches implemented to manage EDs or which could be used to manage EDs (for those regulations that do not contain ED specific provisions). The Table also shows the main interlinkages between various pieces of legislation (more details are given in Annex 4).

Table 3.1 Overview of intervention logic and provisions for endocrine disruptors within in-scope legislation

4COHERENCE

Q1 – To what extent are the different provisions on and approaches to endocrine disruptors coherent across regulatory areas (e.g. horizontal legislation on chemicals, sector-specific and media-related legislation)?

4.1Data requirements 

Data requirements (sometimes known as information requirements) are legal obligations placed on manufacturers or importers to generate data, such as results from toxicity tests, and to provide them to authorities. Establishing data requirements in legislation requires balancing the need for information on the (eco-)toxicological properties of substances with the burden on economic operators and with animal welfare considerations. Provisions should also avoid multiple generation of data for the same substance subject to different pieces of legislation.

Some regulations such as the Biocidal Products Regulation and the Plant Protection Products Regulation, require comprehensive data sets, justified by substances used in those sectors purposely being designed to be biologically active and/or with potential high exposure to humans or the environment. The general legislation on chemicals (REACH) obliges the registrant to register substances and to provide data as set out in the REACH annexes on standard information requirements with increasing requirements according to the production volume for all substances on their own or in mixtures (and in some cases also for articles) manufactured or placed on the market unless there is an exemption. Inter alia, substances manufactured or imported in low quantities (< 1 tonne per year) or polymers do not require a registration. The obligation to register does not apply to active substances, safeners and synergists used in plant protection products, as active substances in biocidal products, in medicinal products for human or veterinary use, or substances intentionally added to food and animal feed. Chemical substances used in other sectors such as toys, food contact materials or cosmetic products are not exempted from the obligation to register and provide data under REACH. Other legislation such as the Cosmetic Products Regulation (CPR) and the Food Contact Material (FCM) Regulation require further information that are specific for the products or the sector they regulate.

In general, further provisions are in place (such as the substance evaluation procedure under REACH) to request further data if a concern arises that is not covered with the standard obligations.

Legislation with no or limited obligations to generate (eco-)toxicological data (such as toys, medical devices or environmental legislation) uses all available scientifically sound data for assessing risks, such as data generated under other regulations, data from academic literature or data supplied voluntarily at the initiative of the industry parties concerned.

The recent fitness check on the EU chemicals legislation concluded that differences in data requirements “can for the most part, be explained and justified on the grounds of differing likelihood of exposures (risks), of costs and proportionality and of laboratory animal welfare considerations. Differences are mostly justified and do not seem to lead to major incoherence issues” 17 . This overall conclusion can also be regarded as applicable to the data requirements for the evaluation of endocrine disrupting properties. However, it should be recognised that these differences are policy decisions driven by proportionality balancing costs and animal welfare against generic exposure considerations.

Provisions for the generation of toxicological data related to chemicals or products that can inform on endocrine-mediated effects on human health are included in the legislation on plant protection products, biocidal products, REACH, food contact materials, food additives, cosmetics and pharmaceuticals. Provisions for the generation of ecotoxicological data that can inform on endocrine-mediated effects on wildlife are included in the legislation on plant protection products, biocidal products, REACH and pharmaceuticals. At present, data requirements under these regulations focus primarily on in vivo repeated dose, reproductive or developmental toxicity testing in mammals, and tests with aquatic (or terrestrial) organisms, providing data on adverse effects. Further investigations to confirm an endocrine mode of action are triggered based on observations of potentially endocrine-mediated adverse effects from these tests. Specific in vitro or in vivo tests investigating endocrine modes of action (mechanistic tests) are not currently part of the standard data requirements. The ongoing revisions of standard data requirements for biocidal products and plant protection products consider the inclusion of additional tests for endocrine properties as part of the core data set. Under REACH, work on incorporating further tests for endocrine properties has started.

Legislation and/or regulatory guidance on data requirements and hazard assessment for plant protection products, biocidal products, REACH, food contact materials, food additives and cosmetics refer to accepted international guidance such as OECD Test Guidelines (TGs). The TGs relevant for testing of EDs are grouped in the OECD Conceptual Framework for Testing and Assessment of Endocrine Disrupters. The OECD Guidance Document 150 (which includes the Conceptual Framework) provides guidance on the interpretation of the tests and suggestion on possible next steps in testing (OECD, 2018). The Conceptual Framework comprises five evidence levels listing the OECD TGs and standardised tests methods available, under development or proposed, that can be used to evaluate chemicals for endocrine disruption. Many of the in vivo TGs (at levels 4 & 5 of the Conceptual Framework) are the standard toxicity TGs used for the evaluation of toxicity and ecotoxicity in general and are not specific to endocrine disruption. They are listed because they investigate toxic effects that may be endocrine-mediated. Levels 2 & 3 list the methods developed specifically to investigate endocrine modes of action and need to be used in combination with the methods at levels 4 &5. Level 1 refers to existing data and non-test information (e.g. computational predictions).

Data generated using OECD TGs are subject to Mutual Acceptance of Data (MAD), allowing data generated in one OECD Member Country to be accepted by all OECD Member Countries as well as other adherents to MAD, thus removing the need for repetition of tests when seeking regulatory approvals across different global regions. EU regulations generally require consideration of all available information, i.e. not only data from accepted international guidelines, but also data coming e.g. from peer-reviewed literature.

For active substances regulated under the Plant Protection Products Regulation (PPPR), EDs can be identified based on level 4 and 5 tests from the Conceptual Framework (i.e. in vivo assays providing data on adverse effects on endocrine relevant endpoints, and in vivo assays providing more comprehensive data on adverse effects on endocrine relevant endpoints over more extensive parts of the life cycle of the organism) which are required for human health assessments. Some mechanistic tests, informing on specific endocrine modes of action can be triggered, namely the uterotrophic assay (OECD TG 440), the Hershberger assay (OECD TG 441), or the in vitro assays for the detection of estrogenic activity and measurement of steroidogenesis (OECD TG 455 and 456). For non-target organisms, data required are mainly based on level 3 and level 4 fish and amphibian tests. The data requirements for active substances and plant protection products are described in Regulations (EU) No 283/2013 and (EU) No 284/2013, respectively, and the accompanying Commission Communications (2013/C 95/01 and 2013/C 95/02).

For biocidal products, tests required for human health safety assessment in Annexes II and III to the Biocidal Products Regulation (BPR) are similar to the ones required for plant protection products, with mechanistic studies being also only triggered on the basis of observations in in vivo tests. For the ecotoxicological assessment however, data requirements listed in the Annexes are less specific than for plant protection products. However, the BPR requires applicants to provide the relevant information necessary to carry out the evaluation.

For these two Regulations (PPPR and BPR), a joint guidance document ‘Guidance for the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009’ (ECHA et al., 2018) describes which information should be provided in order to conclude on the potential endocrine disrupting properties of a substance.

Under REACH, standard information requirements depend on the tonnage manufactured or imported (Annex VII to Annex X). Further tests can also be triggered by conditions listed in column 2 18 . However, the REACH substance evaluation process offers the possibility to require registrants to submit data derived with an accepted test method that is not listed in the standard information requirements, if there is a specific concern such as endocrine disruption. To date, “endocrine disruption” was a concern mentioned for launching this procedure in more than 80 cases. The need to align the framework with the newly established criteria under the PPP and BP Regulations and the latest methods available has been recognised. The process has been initiated within a REACH CARACAL subgroup, which met for the first time in February 2020.

Substances in food contact materials (FCM) for which an authorised list has been established in the framework of the Regulation (EC) no 1935/2004, such as plastics, are assessed by the European Food Safety Authority (EFSA) in accordance with its own Note for Guidance (EFSA Panel on Food Contact Materials et al., 2008). This EFSA Note sets out the data requirements for the safety assessment based on the migration of the substances from the plastic food contact material into the food. As a general principle, the greater the exposure through migration, the more toxicological information is required. Although there are no specific provisions for EDs in the FCM legislation, in the case of plastic FCM, where migration is relatively high (i.e. migration 5–60 mg/kg food) studies investigating reproductive and developmental toxicity are required (EFSA Panel on Food Contact Materials et al., 2008). Business operators using substances in the FCM or the applicant are also required to inform the Commission immediately should any new scientific information become available that may affect the authorisation. This could include changes in the migration or additional toxicological information e.g. from the scientific literature or new data available under REACH. If necessary, further assessment by EFSA should be carried out. Moreover the EFSA scientific opinion on recent developments in the safety assessment of chemicals in food (EFSA Panel on Food Contact Materials et al., 2016) supports consideration of additional studies on specific endpoints including those resulting from possible endocrine disruption (as from OECD GD150, 2012), as well as immunotoxicity and neurotoxicity, regardless of the level of exposure. Regarding food additives, there are no specific provisions for EDs, however the repeat dose and reproductive/developmental testing methods listed in the EFSA Guidance for submission of food additive evaluations (EFSA Panel on Food Additives and Nutrient Sources added to Food, 2012) do include endocrine-relevant endpoints. Moreover, the guidance refers to other studies (in addition to the core areas for evaluation) that may be relevant and useful for assessing the risk and establishing the safety of an additive include immunotoxicity, hypersensitivity and food intolerance, neurotoxicity, endocrine activity and mechanisms and modes of action.

Substances used in cosmetic products are also subject to the registration obligation under REACH, (based on the tonnage band), and thus the relevant REACH data requirements have to be fulfilled. There are no explicit data requirements specified in the Cosmetic Products Regulation (CPR), which instead refers to the Scientific Committee for Consumer Safety’s Notes of Guidance (SCCS, 2018). These describe the types of data that may be used or generated by the manufacturers to make an assessment for each toxicological endpoint relevant to human health, which the manufacturer takes into account when there is a need to submit a safety dossier under the CPR authorisation procedures. Data relevant to endocrine modes of action may be provided through the assessment of reproductive toxicity or through special investigations on suspected endocrine disruptors for human health.

The guidance refers to the OECD Conceptual Framework, however, due the ban on animal tests, only in silico and in vitro methods can be used to evaluate safety for human health, while data from in vivo studies may be used if they are already available or generated through other regulatory processes such as under the PPPR, the BPR and REACH. The relevant toxicological studies described in the guidance cover standard apical endpoints, including reproductive toxicity. For reproductive toxicity, three in vitro embryotoxicity screening tests have been validated by ECVAM to support assessment of reproductive toxicity. Overall, the complex endpoint of reproductive toxicity is not covered by the above systems and no alternative methods are currently available covering all aspects of reproductive and developmental toxicity. Many recent SCCS opinions have relied on in vivo reproductive/developmental toxicity studies (e.g. OECD TGs 416, 414, 421, 422 and 443) performed before the animal testing ban (e.g. Annex 5A - 3-BC case study). The guidance does not require tests addressing the four E-A-T-S modalities. Results from available in silico and in vitro tests for endocrine activity do not fit in the risk assessment framework followed by the SCCS for safety evaluations. In addition, according to the SCCS, such tests would only inform on endocrine activity but not adversity. According to the SCCS, the animal testing ban would not enable the identification of EDs according to the WHO definition or according to the criteria for biocides and pesticides, which require information on three elements, namely endocrine activity, adverse effect and plausible causal link.  However, the SCCS recognises the need to transition toward an animal-free risk assessment methodology, sometimes called Next Generation Risk Assessment (NGRA). However, this will require methodological developments and experience in the application and interpretation of New Approach Methodologies (NAMs) (Rogiers et al., 2020).

The combination of provisions for data requirements under REACH (tonnage-based) and the CPR together with the incremental restrictions on animal testing over time explain the current differences in data packages available for individual substances. For some cosmetic ingredients evaluated and approved in the past, including substances listed in the annexes to the CPR (positive list of approved colourants, preservatives and UV-filters), data for the evaluation of endocrine-mediated effects can be very limited if the substance is not registered under REACH or registered at the lowest tonnage level (1 to 10 tonnes per year) (see also Annex 5A, 3-BC case study). In case the substance is not registered in REACH (i.e. not on the market in a tonnage of at least 1 t/y or above), there are no mechanisms to know if those substances on the positive lists continue to be used at all.

In addition to the data prescribed to ensure therapeutic efficacy and safety (not in the scope of this Fitness Check), regulations on human and veterinary medicines include data requirements to evaluate potential unintended ecotoxicological effects. A draft guidance by the European Medicines Agency (EMA) on the environmental risk assessment (ERA) of human medicines was released for public consultation in 2019 19  and is currently under revision. According to this draft guidance, testing for endocrine active substances may be triggered as part of a tailored risk assessment for endocrine active substances (Annex 4). The draft Guidance published for public consultation lists recommended tests to address (anti)estrogenic, anti(androgenic) and thyroid effects (OECD TGs 234, 240, 241) Testing on thyroid agonist or antagonist effects is recommended, even though the definition of endocrine active substance given in the guidance does not cover the thyroid modality. Discussions on the finalisation of the Guidance are ongoing.

Guidelines on environmental safety assessments for non-target organisms of veterinary medicines refer to the VICH guidance (EMA, 2000, 2005). The tiered approach consists of a two-phase approach. Only when environmental exposure from the use of the veterinary medicine is considered to be significant (by exceeding established threshold values), a higher-tier assessment based on a risk ratio will be triggered. This second phase will be conducted starting from basic risk assessments using short term (acute) toxicity to chronic aquatic toxicity studies with standard aquatic species from three trophic levels. Required studies do not include endpoints that are relevant and can identify endocrine-mediated effects. The ability to identify potential EDs is limited to evidence from changes indicative of endocrine dysfunction reported in the mammalian repeated dose toxicity studies (performed to establish a maximum residue limit to ensure the safety of veterinary drug residues in human food) or to scientific literature. Such evidence may prompt the member state competent authority or EMA to request additional studies targeted to investigate ED properties.

To conclude, ongoing revisions of standard data requirements for biocidal products, plant protection products and for REACH consider the inclusion of additional tests for endocrine properties as part of the mandatory data set .Currently test guidelines are only available to cover the estrogen, androgen, thyroid and steroidogenesis modalities (EATS) but this is likely to increase over time, as more OECD Test Guidelines covering other endocrine modes of action are validated and adopted. Since many substances used in products subject to product-specific legislation in the scope of this fitness check are required to be registered under REACH, the information requirements in REACH can be used as a centralised way to gather the relevant data, thus avoiding repetition of studies on the same substance in other regulatory domains. In relation to arguments of proportionality of costs and considerations of animal welfare, when adopting REACH the co-legislators have chosen to use production volume of substances as a surrogate of exposure (and consequent risk) potential and thus substances produced at volumes below 10 tonnes/producer/year have a data set which does not give any indications of endocrine disrupting properties. However, all available information, including from in vitro tests, QSARs and read-across to other substances has to be taken into account by manufacturer or importers. The OECD GD 150 lists several in vitro tests that provide information on endocrine activity. Currently, these in vitro tests are not a mandatory part of data requirements of the relevant regulations. Substances produced at less than 1 tonne/producer/year are currently not registered under REACH and thus it is important to be aware that REACH will not provide any data in these cases to support ED identification for such substances used in other product-specific legislation. Legislation covering the food additives and food contact materials which has data requirements independent of, or additional to REACH, respectively, follows a similar approach whereby concerns for endocrine disrupting activity may be identified from the required toxicity tests. Coordination is required between REACH and these regulations to avoid the duplication of testing and ensure the best use of resources.

4.2Hazard identification

4.2.1Identification of endocrine disruptors

Three EU Regulations contain provisions to identify EDs, i.e. the Biocidal Products Regulation (BPR), the Plant Protection Products Regulation (PPPR) and REACH.

Identification of active substances as endocrine disruptors is mandatory for plant protection products and biocidal products (for the latter also non-active substances need to be assessed), following scientific criteria established in 2017 and 2018. It is pursued, to the extent possible, under REACH. It is not required in other sectors. The identification of EDs requires: i) a definition, ii) clear criteria to determine whether the definition is met, iii) data requirements to establish if criteria are met, and iv) guidance for implementation of the assessment process.

Across chemical legislation, there is wide acceptance of the WHO definition: “an endocrine disruptor is an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub) populations”. The definition is supported by most stakeholders ((European Commission, 2015). However, differences exist in the wording used across regulations. Although all relevant provisions refer to ED properties, regulations vary in the emphasis on the adversity and causality attributes, as analysed in the Fitness Check of EU chemicals legislation (excluding REACH) 20 . Differences in wording might create uncertainty as regards which chemicals are considered EDs by the legislative provisions and what level of evidence is required to identify such chemicals.

Criteria, data requirements and guidance to enable the identification of EDs exist for the Plant Protection Products and Biocidal Products Regulations. The European Chemical Agency (ECHA) and the European Food Safety Authority (EFSA) jointly published a guidance document (ECHA / EFSA 2018) based on these criteria. The guidance has been recognised as a major milestone providing clear and scientifically sound interpretation of the WHO definition (PETI Committee for the European Parliament 2019). It provides a starting point for any future cross-sectorial definition of EDs. ECHA, EFSA and Member State authorities are sharing first experiences to ensure consistent application of the guidance and to identify possible improvements 21 . The Agencies are also collaborating on ED assessments for specific substances of common interest.

Currently, REACH does not contain criteria for EDs, but bases the identification of Substances of Very High Concern (SVHCs) with ED properties on the IPCS/WHO definition. The identification of an ED as SVHC needs also to consider whether the effects give rise to an “equivalent level of concern” to those of other SVHC categories. This assessment is done on a case-by-case basis. In case of lack of consensus in the REACH Member State Committee, the Commission takes the final decision through the REACH comitology process.

The Cosmetic Products Regulation (CPR) does not require the identification of EDs. However, substances used in cosmetics, may be identified as EDs under REACH in relation to environmental concerns or, for substances that are registered for other uses, with respect to human health. According to the Scientific Committee on Consumer Safety (SCCS), it would not be feasible in the context of the animal testing ban to implement all the criteria established for plant protection products and biocidal products in the CPR, i.e. the requirement to demonstrate an adverse effect, which currently is mainly demonstrated through in vivo animal studies. The Commission can request the SCCS to evaluate the safety of substances identified as potential endocrine disruptors but the specific risk approach that underpins the SCCS safety evaluations does not depend on formal identification of EDs (SCCS 2018).Error! Bookmark not defined..  Chemicals used in medical devices, food contact materials, cosmetic products, detergents and toys as well as in other sectors can in principle be identified as SVHCs with endocrine disrupting properties via REACH and therefore the absence of specific provisions for ED identification in the sector-specific legislation does not necessarily represent a regulatory gap across legislation. 

Although the absence of horizontal criteria creates a potential for inconsistent identification across regulation, which is further compounded through different data requirements and different interpretations of the same data, no inconsistent identification has occurred so far. The targeted stakeholder survey provided insights into the different views about the lack of a horizontal approach to identification and classification of EDs (Box 4.1). There is wide agreement among stakeholder groups that the lack of a harmonised approach is a problem. However, there are diverging stakeholder views on the way forward for how to accomplish a harmonised approach, for instance via the SVHC-identification procedure under REACH, a dedicated hazard class under Classification, Labelling and Packaging (CLP) Regulation and/or the UN Globally Harmonized System of Classification and Labelling of Chemicals (GHS) with or without categories, or via a separate legislation. In a similar way, the implementation of a category of suspected endocrine disruptors would have the advantage of stimulating assessments and substitutions but could lead to the exclusion of substances that could be used safely. Another consideration is in the context of the PPPR and BPR, where the criteria for EDs do not include a suspected category. The implementation of an additional hazard class in the CLP regulation and possibly categories could help to define clear regulatory triggers as part of assessment workflows from screening assessments to risk management measures.

4.2.2Identification through Classification, Labelling and Packaging Regulation hazard classes

Currently there is no hazard class for endocrine disrupting properties under the Classification Labelling and Packaging (CLP) Regulation. Many known EDs are classified as carcinogenic or toxic to reproduction, or both. Along with mutagenic substances, these hazardous properties are considered to produce effects of a serious and irreversible nature and are collectively referred to as CMRs (Carcinogenic, Mutagenic or toxic for Reproduction). The CMR categories include important ED-mediated effects, such as cancers in hormone sensitive organs, and reproductive and developmental disorders, which might be ED-mediated. Six of the seven chemicals identified so far as Substances of very High Concern (SVHCs) for endocrine disrupting properties with respect to human health under REACH are also classified as reproductive toxicants (category 1A or 1B) under the CLP regulation.

However, the substance butyl 4-hydroxybenzoate (butyl paraben) was identified as ED for human health in REACH in June 2020, but is not classified as CMR, since effects related to fertility and development has not led to a harmonised classification as toxic for reproduction.  In addition, available toxicological data triggering CMR classification may not capture certain effects potentially mediated by endocrine disruption, such as certain neurodevelopmental and metabolic disorders. As methods and data on thyroid disruption, steroidogenesis and other less studied modalities of endocrine disruption become available, it is expected that more substances will be identified as EDs based on effects beyond those defining CMRs. Other hazard classes, under the CLP Regulation, relevant to human health, focus on specific target organ toxicity (STOT) and include potency considerations such that a substance toxic to e.g. the thyroid may not be classified if the effects are only seen above a certain exposure concentration. Such substances would remain unclassified in relation to the adverse effect, but may still fulfil the definition of an ED.

In relation to wildlife, CLP classification for ecotoxicity is generic and does not refer to specific modes of toxic action. Many known EDs happen to be also classified using the aquatic hazard classification scheme (chronic categories 1 to 4). However, data triggering this classification are the chronic no observed effect concentration (NOEC) if available, fish bioconcentration factor (BCF) and/or degradable properties of the chemical. This means that if the chronic NOEC is based on a test which does not have specific endocrine-mediated endpoints (e.g. OECD TG 210), an ED may or may not be classified as a chronic aquatic toxicant.

The importance of horizontal classification of EDs under the CLP Regulation is widely recognised 22 since a hazard classification may be used to trigger risk management measures in many so-called ‘downstream’ product and sector-specific regulations and directives. For example, the CLP hazard classification of substances is also the basis for the identification and classification of hazardous waste containing them.

CMRs of categories 1A or 1B can be identified as substances of very high concern (SVHC) under REACH and also fulfil the cut-off criteria for non-approval under the plant protection products and biocidal products regulations (unless the limited derogation possibilities in the Regulations apply). Identification of a CMR as an ED for human health will probably not lead to any additional risk management measures, where they are already the strictest (see 4.4 Risk Management) but may nevertheless contribute to the overall evaluation within a risk-benefit context and facilitate assessments of combined exposure/mixtures. In addition, as described above, not all EDs for human health are identified as CMRs and identification as an ED for the environment may also trigger additional risk management measures.

*Answers are given in percentage of respondents from each stakeholder group. For more details see Annex 2.

**Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Some of the stakeholders who disagree with the introduction of a hazard class covering endocrine disrupting properties in the CLP Regulation and/or GHS state that it would be redundant as adverse effects triggered by endocrine activity are already covered by existing GHS/CLP hazard classes. However, this would be true only for some endocrine–related effects, excluding for example effects of obesogens, as tests generally used for classification in the existing hazard classes cannot identify such EDs. Regarding the statement that classification is not necessary for risk management, this may be the case. However, the CLP Regulation may have advantages with respect to simplification since it is often the instrument currently used for hazard classification. It is also referred to by many of the current product and sector-specific regulations in relation to invoking specific risk management measures and thus an ED hazard class could be considered for use in a similar way. Communication is also an important aspect of classification through the CLP Regulation with respect to protection of workers and consumers (especially vulnerable groups) although this issue has been partially resolved for workers through the requirements to indicate endocrine disrupting properties on safety data sheets.

There is a commitment of the EU to explore the possibility to harmonise classification at the international level through the UN GHS. However, it could take many years to reach agreement at international level, if at all, and thus delay action at EU level with the consequent implications on effectiveness. Alternatively, as also stated by some stakeholders, REACH could also be a potentially suitable instrument that could be further reinforced to ensure consistent ED identification and appropriate risk management measures across sectors and products.

The views on the introduction of a suspected category of EDs are more polarised across the different stakeholder groups. Ideally, testing for endocrine disruption should proceed until a decision can be made that the substance is or is not an ED. In practice it may be difficult to generate sufficiently conclusive evidence one way or the other for all substances to inform on regulatory decision making in a reasonable timeframe. Hence the suggestion for a ‘suspected’ category. Proponents of a suspected category argue that it would be prudent to flag the possible hazardous properties of the substance, while the detractors consider that it would “stigmatise” a substance that may not have such properties and discourage users from its use. Important considerations in the discussion are deciding where on the spectrum of evidence the criteria for a ‘suspected ED’ should be set, and what risk management measures, if any, should be applied to such a category. In the context of setting the criteria for biocidal and plant protection products the Commission Communication 23 stated that establishing different categories of what may be an endocrine disruptor does not help to define what is an endocrine disruptor and would decrease legal certainty for regulators and stakeholders without established benefits in terms of protection of health and the environment. However, this approach with categories is, for example, already followed for CMRs.

4.3Risk assessment

4.3.1Feasibility of risk assessment

Risk assessment is the cornerstone of safety evaluation procedures across many pieces of chemicals legislation. Risk is quantified as the ratio of the chemical exposure in an organism or in an environmental media to an exposure level considered safe. The approach builds on the concept that, for a given substance, a biological threshold often exists for a given toxicological effect, below which exposure does not exert any adverse effect in organisms. Within a given regulatory framework, risk assessment is feasible, when such a threshold can be derived using the (eco)toxicological data available.

There is an ongoing debate in the scientific community about the viability of performing risk assessment for EDs. The debate is reflected in the views of members from the Endocrine Disrupters Expert Advisory Group (ED EAG), which was composed of toxicologists and ecotoxicologists with regulatory and/or endocrinology backgrounds, nominated by Member State Competent Authorities for REACH, the Biocidal Product Regulation (BPR), and the Plant Protection Products Regulation (PPPR), relevant industry associations and non-governmental consumer/environmental protection organisations (Munn and Goumenou, 2013b). According to this report, most experts considered that biological thresholds of adversity are likely to exist; several maintained that it is difficult to estimate them based on currently available test methods, while others considered it is possible to estimate them with appropriate testing. There was lack of consensus on the sensitivity of existing methods to detect responses at low doses.

The debate also exists in the context of EU regulatory assessments. For example, in 2013, in its restriction dossier on nonylphenol ethoxylates, the Swedish competent authority for REACH stated that “identifying safe concentration limits for all possible endpoints within the endocrine system that can be affected by EDs is not possible within the current test methods and would be inconsistent with the objective in REACH to reduce animal testing” (see Annex 5C, nonylphenol case study). The Risk Assessment Committee (RAC) abstained from any opinion about whether a threshold proposed during its discussions was sufficiently protective for all hazards of nonylphenol (NP), including those posed by endocrine disrupting properties, pointing to the then on-going discussion in the EU Commission on the possibility to derive a safe level of exposure 24 . Instead, it based its opinion on the identified risks for endpoints not related to endocrine disrupting properties.

In the context of REACH authorisation, the Commission indicated in 2016 that EDs might or might not have a threshold and that “it remains the responsibility of applicants to demonstrate that a threshold exists and to determine that threshold. Even though this might be particularly difficult for EDs, it cannot be excluded on the basis of current knowledge that it will be possible. It is up to RAC to assess the validity of the assessment and ultimately decide on the possible existence or not of this threshold” 25 . In the authorisation procedure, the application of risk assessment to demonstrate safety has been termed the ‘adequate control route’. However, adequate control based on risk assessment is not an option if a threshold cannot be demonstrated. For example, NP ethoxylates (NPnEO) were added to the Authorisation list (Annex XIV) for their endocrine disrupting properties for the environment (Annex 5C, nonylphenol case study). In a note intended to provide general advice to companies intending to apply for authorisation of uses of octylphenol and nonylphenol ethoxylates, the RAC “did not consider that exposure values below [a conventionally derived] predicted no effect concentration were consistent with a ‘safe level’, specifically noting the uncertainties associated with the endocrine disrupting properties of NPnEO” (ECHA RAC 2017). To date no applicant tried to demonstrate a safe threshold and applied for the adequate control route. In this case authorisations are granted based on risk-benefit considerations, provided that exposure is minimised and alternative substances are not available. REACH restrictions require evidence of an unacceptable risk to human health or the environment, arising from the manufacture, use or placing on the market of the substance. In cases where it is difficult to derive safe thresholds, a qualitative assessment of the likelihood that effects are avoided can be used when implementing measures. In recent restrictions on chemicals with endocrine disrupting properties (e.g. DEHP, nonylphenol ethoxylates), the unacceptable risk was proven since risk ratios exceeded the value of 1. In both cases, the endpoint assessed was not explicitly related to endocrine properties.

In its 2013 opinion, EFSA’s Scientific Committee considered that biological thresholds of adversity do exist and considered human and environmental risk assessment (taking into account hazard and exposure data/predictions) the best approach to inform risk management decisions in regulations that base decisions on the risk and level of concern (EFSA Scientific Committee, 2013). According to EFSA’s Scientific Committee “EDs can therefore be treated like most other substances of concern for human health and the environment, i.e. be subject to risk assessment and not only to hazard assessment”. Risk assessment had been applied to assess the risk of bisphenol A (EFSA Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids, 2015) and phthalates (based on reprotoxic effects) in food contact materials (Annex 5B, DEHP cases study) before their identification as EDs under REACH. In these cases, the assessment factors applied to derive tolerable daily intakes can account for uncertainties related to endocrine-mediated effects. In the 2015 EFSA risk assessment of BPA, for example, uncertainty in mammary gland, reproductive, neurobehavioural, immune and metabolic system effects justified an additional assessment factor of 6.

The Scientific Committee on Consumer Safety (SCCS) endorsed the 2013 EFSA opinion on EDs (SCCS 2014, and COM (2018) 739) and evaluates cosmetic ingredients based on risk assessment, including the derivation of safe exposure thresholds. To date the SCCS has not completed an evaluation for substances after these have been identified as EDs under other legislation. However, the SCCS is currently evaluating five substances with potential endocrine disrupting properties (homosalate, octocrylene, benzophenone-3, resorcinol and propylparaben).

Rather than being a means of demonstrating safety, risk assessment is also applied in some sectors to assess the probability of harm and conclude on tolerable risk levels for risk-benefit evaluations or risk mitigation. This is also the case for biocidal products, medical devices (human health risks as part of the benefit-risk analysis), pharmaceuticals (for environmental risks), veterinary medicines (for environmental risks as part of the benefit-risk analysis).

4.3.2Differences in risk assessment

The subdivision of assessment activities across policy areas means that the same chemical (or group of chemicals) might need to be assessed for human health or environmental endpoints under multiple frameworks (Figure 3.1). Although agencies and Scientific Committees have increasingly sought alignment in recent years, stakeholders expressed the need for better coordination of agencies and committees 26 .

The three case studies performed for this Fitness Check (Annex 5) were chosen to evaluate the horizontal coherence of regulatory assessments for EDs across policy areas. Differences in risk assessments may emerge at different levels – chemical grouping, exposure pathways assessed, methodology or data used. Where EDs have been assessed, our analysis did not identify significant deviations from the standard methodology, described in the guidance documents, owing to ED-specific considerations. In one case, when conducting the risk assessment for nonylphenol in the REACH restriction report, the Swedish competent authority introduced an extra assessment factor of 10 in the calculation of safe environmental concentrations to account for the incomplete knowledge and the uncertainties related to its ED properties. The ECHA Risk Assessment Committee (RAC) opinion did not endorse the approach but referred to the risk ratios obtained with the standard approach, which already indicated a risk (Annex 5C, nonylphenol case study). The uncertainties associated with the possible peculiarities of endocrine disrupting effects (i.e. non-threshold, non-monotonic behaviour) are recognised and in some cases have been taken into account through the use of uncertainty factors, defined on a case by case basis.

Differences in risk assessments performed for the same substance(s) under different pieces of legislation reflect differences in the legislative scope (especially, in the cases identified, the scope of the exposure assessment and the chemical grouping). The case of low molecular weight phthalates recently assessed by ECHA and EFSA illustrates the challenge (Annex 5B, DEHP case study). Both agencies performed a risk assessment based on the same hazard (reproductive toxicity) and toxicity data. The two assessments differed in terms of the definition of chemical grouping (DIBP was not assessed by EFSA since it is not authorised as an additive in FCM) as well as the approach and data used to estimate exposure (e.g. exposure sources considered, population age grouping, food intake estimates). ECHA concluded that there is risk to human health from combined exposure and especially of significant exposure to DEHP from the diet based on modelling estimates. EFSA’s assessment concluded that dietary exposure did not result in group-based or individual tolerable daily intakes being exceeded. A risk is obviously more likely to be found if aggregate exposure from all possible sources is considered rather than only a single source. The two assessments are consistent within their respective mandates, but the different scopes and approaches to estimate exposure resulted in different regulatory interventions, reflecting the absence of a coherent cross-sector approach.

The case study on nonylphenol revealed an example where hazard datasets are not harmonised across chemical and water legislation, leading to differences in derived threshold values (Annex 5C, nonylphenol case study).

Evidence from the case studies suggests that where the same substance(s) have been assessed under different frameworks it is difficult to ensure coherence of assessments. The problem is generic and does not relate specifically to EDs. Regarding the exposure assessments, when differences in the exposure scenario(s) and exposure routes assessed are linked with the specific scope of sectorial legislation, the issue cannot be easily solved in the current framework. While most stakeholders largely agree on working towards a framework when one substance is associated with one assessment, based on a shared data set across all regulatory sectors, the current framework does not provide a systematic approach to integrate total aggregate exposure with sector-specific exposure scenarios. The Commission’s ambition to develop a one substance-one assessment approach may provide a means to address this issue.

4.3.3Consideration of vulnerable groups

(Sub)population groups may be at higher risk due to numerous factors, including those affecting external exposure, internal exposure and biological susceptibility (Table 4.1). Factors affecting external and internal exposure are relevant in general but are not specific to EDs.

Table 4.1. Factors contributing to higher risk of (sub)population groups and possible options to address them in risk assessment

Several pieces of legislation within the scope of this Fitness Check consider risks to vulnerable groups. Where such risks are taken into consideration, the definition of vulnerable population varies as there is no horizontally applicable definition of 'vulnerable group'. The Plant Protection Products Regulation and the Biocidal Products Regulation specifically define vulnerable groups in relation to chemical risks: both refer to ‘pregnant and nursing women, the unborn, infants and children, the elderly and workers and residents subject to high exposure to active ingredients over long term’. The rationale for explicitly mentioning specific vulnerable groups in legal provisions is not always clear 27 . For example, legislation on detergents and food contact materials does not include specific provisions, but cosmetic products and other food legislation (e.g. food additives) does. However, lack of reference or definition of vulnerable groups in legislation does not necessarily lead to lack of assessment or action, as exemplified by the risk assessment of and restrictions on substances in food contact materials.

Current risk assessment methodologies across chemical safety legislation aim to cover risks to the general population including the most sensitive sub-populations (i.e. vulnerable groups). In practice, the default safety/assessment factors applied when extrapolating from the animal data to humans may not be considered large enough and additional assessment factors are sometimes used to cover these vulnerable groups. Thus, risks to vulnerable groups are addressed on a case-by-case basis through triggers and guidance established under sector-specific legislation. Assessments for specific vulnerable groups are mostly triggered by factors affecting internal and external exposure and focussed mainly on neonates and young children. Some examples of how vulnerable groups are considered in regulatory assessments are reported in Box 4.2.

Under REACH, guidance exists to account for specific exposure situations and to derive specific no-effect levels (DNELs) for vulnerable groups (non-ED specific) with an emphasis on reproductive toxicity 28 . The REACH guidance on exposure refers to the OECD considerations when assessing children’s exposure to chemicals and a decision tree for deciding when to trigger tailored assessments for children (OECD, 2019). In this context, children are broadly defined as any human from birth to 21 years old, according to the WHO definition.

The Cosmetic Products Regulation requires a specific assessment for products intended for children under 3 years. The Scientific Committee on Consumer Safety (SCCS) Notes of Guidance mention certain circumstances, which might require specific margins of safety for certain subpopulations, such as childrenError! Bookmark not defined.. Correction factors are reported for different age groups to account for higher surface area per body weight. Other factors listed that can affect vulnerability of new-borns are toxicokinetic parameters (slower metabolism and elimination rates), and sensitivity to inflammation for specific uses (i.e. for products intended for use in the nappy area).

Considerations about the level of protection of vulnerable groups to EDs are presented in Section 5, including responses from stakeholders and citizens surveys (Box 5.9).

4.3.4Prioritisation and Environmental Quality Standards under the water legislation

The regulation of EDs under water-related legislation is largely dependent on data generation and evaluation carried out under other regulatory frameworks, such as REACH, or in scientific studies. In recent years, substance prioritisation and setting of Environmental Quality Standards (EQS) have taken advantage of the increased amount of ecotoxicological information, including information on ED-mediated effects, which has become available through chemical registration dossiers under REACH, the Plant Protection Products Regulation and the Biocidal Products Regulation 29 .

The Water Framework Directive (WFD) applies a combination of generic and specific risk considerations from both experimental and modelling data for the classification of chemicals as priority substances (PS) and priority hazardous substances (PHS). The first list of PS followed a scoring-based ranking approach, including elements of exposure and effects as well as biodegradability and bioaccumulation potential 30 . Nonylphenol and DEHP, for example, were recommended for inclusion in the list for their relatively high modelling-based ranking combined with additional evidence of environmental occurrence and endocrine disrupting effects (Fraunhofer-Institut, 1999).

The designation of PS or PHS considers both generic (e.g. for Persistent, Bioaccumulative and Toxic, PBT substances) and specific risk criteria 31 . For the final designation, "additional considerations", including suspected endocrine disrupting potential, can influence the designation. For nonylphenol, evidence of “widespread risk to or via the aquatic environment” 32 assessed under pre-REACH regulation 793/93 determined the classification 33 . DEHP was not originally designated as a priority hazardous substance, but it was designated as such in 2013.

Prioritisation approaches have evolved over the years, but have maintained a combination of generic (PBT, Substances of Very High Concern (SVHC), and Persistent Organic Pollutant (POP) criteria) and specific risk approaches from experimental and modelling data. An early step of the prioritisation carried out in 2011 comprised scoring against four criteria: persistence, bioaccumulation, toxicity and endocrine disrupting properties29.  

Environmental Quality Standards are established for both PS and PHS, even though the WFD aims to ensure the phase-out of PHS emissions to water, not only a reduction in those emissions as in the case of the other PS. The logic of setting EQS for PHS has been questioned by some stakeholders, but it is generally accepted that there is a need for a benchmark in the context of chemical water quality assessments. The persistent nature of many PHS makes it difficult to comply with the EQS, let alone a more ambitious target. EDs are treated like other chemicals when deriving EQS (e.g. Annex 5C, nonylphenol case study). However, if there are indications of adverse effects via endocrine activity (e.g. in vivo bioassays) an additional assessment factor may be considered to cover the anticipated effects 34 .

Recently, the Drinking Water Directive (DWD) has stepped up efforts to monitor risks from exposure to EDs. Under the Drinking Water Directive, three substances have been proposed as indicators of estrogenic contamination from sewage effluents, based on WHO recommendations: β-oestradiol, bisphenol A and nonylphenol 35 . Bisphenol A was added to Annex I with a health-based parametric value based on a 2015 EFSA opinion (EFSA Panel on Food Contact Materials et al., 2015), and nonylphenol and β-oestradiol will be added to a watchlist, for assessing their occurrence and treatment efficacy where necessary.

4.4Risk management

4.4.1Management approaches across legislation on chemicals

Generally, regulatory intervention on chemicals can be based directly on their hazardous properties (generic risk approach), on the result of risk assessments (specific risk approach), or on risk-benefit considerations 36 . Legislation often combines elements of the different approaches. For example, under REACH the listing as Substance of Very High Concern (SVHC) is hazard-based (generic risk approach); application for authorisations may be based on (adequately controlled) risk or on risk-benefit arguments; restrictions require that there is evidence of an unacceptable risk to human health or the environment. Restriction specifications are also substantiated by (socioeconomic) risk-benefit considerations. Product-level legislative instruments, such as the Cosmetic Products Regulation (CPR), Toy Safety Directive and the Food Contact Materials (FCM) Regulation apply generic risk approaches for Carcinogenic, Mutagenic and toxic to Reproduction (CMR) hazard classes, with specific risk approaches for the risk management of other types of hazardous substances. Exceptions exist where the specific risk approach can also be applied to CMRs. While based on assessments by EU Agencies and scientific committees, decisions can also be subjected to the application of the precautionary principle.

The Fitness Check on chemical legislation (excluding REACH) concluded: “In many instances, these differences reflect variations in legal scopes and objectives and thus different needs in terms of depth of analysis and evidence required to draw conclusions and decide upon any risk management measures that may be needed. Therefore, these differences do not necessarily imply incoherence. They illustrate the legislator’s intention to provide a framework that is tailored to the specific circumstances of the substances used and/or the likely hazards and exposure”; and further on risk management of known adverse effects: “The majority of currently known adverse effects on human health and the environment are covered [in legislation]”. However, “some inconsistencies occur regarding risk management decisions for EDs […]” 37 . This analysis did not go in depth, referring to the decision by the Commission to launch the present Fitness Check on EDs.

Substances with endocrine disrupting properties pose specific challenges for their assessment. The identification of EDs demands a significant amount of information to clarify potential modes of action and different effects, which often necessitates additional testing (possibly including further animal testing). Risk assessment approaches have limitations for ED-mediated effects in those cases for which it is difficult to derive a safe (or acceptable) threshold based on the available scientific evidence. The same limitations add to the uncertainty of risk-benefit analysis. However, socioeconomic considerations based on other approaches can be applied for non-threshold substances. The legislator has opted for different approaches to risk management, considering policy-specific objectives and constraints (See Table 3.1 and Annex 4 for a complete description). Under the Plant Protection Products Regulation (PPPR) and the Biocidal Products Regulation (BPR) the legislator chose to ban EDs but put in place specific derogation possibilities. The approach implemented under REACH is also a generic risk approach insofar as EDs can be identified and listed as SVHCs. The intended consequence of listing SVHC in Annex XIV of REACH is substitution as soon as technically and economically feasible. Authorisation may be granted if risks are adequately controlled or if benefits outweigh risks (e.g. authorisations of DEHP and nonylphenol ethoxylates). Restrictions under REACH follow a specific risk approach and take socioeconomic considerations into account. Regulations on medical devices and in vitro diagnostic medical devices require a risk-benefit justification for EDs (identified via REACH and, only for medical devices, via the BPR). Thus, in PPPR, BPR, and the medical devices regulation, the management approach for EDs is comparable to that applied to CMRs. In the case of REACH, however, the automatic ban applied to CMRs in mixtures for consumer use is not applicable to EDs.

Where legislation does not make specific reference to EDs (e.g. Cosmetic Products Regulation, Food Contact Materials Regulation) the specific risk approach is applied. Also the Toy Safety Directive does not refer to EDs but this is a particular case because risk management overlaps with REACH with regard to human health concerns. Classification as hazardous chemicals (including carcinogens and mutagens) is the main trigger to risk management provisions under Occupational Safety and Health (OSH) legislation through a hierarchy of measures from preventing exposure to reducing it to the lowest level technically possible, and in waste legislation through the classification of hazardous waste.

The legislator opted for different risk management approaches based on specific policy objectives and considerations, including:

- Application of the precautionary principle. When there are reasonable grounds for concern that potential hazards may affect the environment or human, animal or plant health, or when the available data do not allow to conclude on the absence of risk following a detailed risk evaluation, the precautionary principle has been politically accepted as a risk management strategy. The approach under both the PPPR and the BPR is “underpinned by the precautionary principle” taking into account the specific nature of the products in question and the scientific uncertainties regarding their assessment (for example as regards the existence of a safe limit of exposure) 38 . In other cases, the precautionary principle is invoked for substance-specific decisions. For instance, it served as a basis for the currently applicable ban of Bisphenol A (BPA) in baby bottles since 2011 (Commission Implementing Regulation (EU) 321/2011) and more recently for FCM aimed at infants and young children (Regulation (EU) 2018/213).

- Risk-benefit analysis. In combination with an evaluation of available alternatives, risk-benefit analysis is the principle of approval procedures for medical devices (e.g. DEHP). It is also applied under REACH for authorisations (e.g. socioeconomic route in REACH authorisations of nonylphenol ethoxylates) and restrictions, and under the Biocidal Products Regulation for derogations to the general non approval of active substances identified as EDs. For human medicines, therapeutic benefits have the priority. Unintended effects on the environment need to be assessed and to some extent managed (e.g. through waste management) but do not impact approval procedures.

- Trade-offs between policy objectives. Differences in the objectives between chemical, product and waste regulation (e.g. safety vs recyclability targets) unavoidably result in trade-offs. For example, decisions on authorisation for certain SVHCs in secondary materials is influenced by consideration of the benefits of recycling (Annex 5B, DEHP case study).

While policy-specific objectives and considerations, such as the ones highlighted above, generally explain the rationale behind the current differences in risk management approaches, the rationale is not always clear and explicit. For example, the co-legislator’s choice for plant protection products and biocidal products was justified by “the specific nature of the products”38, which can be interpreted as the likelihood of exposure and toxicity. This argument is not clear or always supported by scientific evidence of higher ED-related risks compared to other sectors. Another more specific example is the different derogation principles under the Plant Protection Products Regulation (negligible exposure or essential use) and the Biocidal Products Regulation (negligible risk, essential use, or risk-benefit considerations). Stakeholders often criticise situations where substances are restricted under some regulations but not under others and consider them not justified (Box 4.3).

Even where regulations apply the same risk management approach, differences in the scope of a risk assessment can also result in different risk management measures for the same substance(s). While not ED-specific, this issue poses a challenge to the coherence of interventions across sectors, as described in Chapter 4.3.2 and Annex 5B (DEHP case study).

Overall, it can be concluded that different approaches to risk management have arisen as solutions to specific policy considerations. Although based on a limited number of examples (Table 1.1), this Fitness Check found no cases of inconsistent risk management caused by ED specific issues, such as the lack of a horizontal approach to identification.

However, the difficulty in choosing a risk management approach is partly related to the fundamental scientific question of whether EDs are (all) threshold or non-threshold substances. There is no scientific consensus on the threshold question. In fact, since EDs cover a multitude of modes of action, in some cases it may be justified to assume that a threshold exists and to follow a specific risk approach. In the absence of scientific consensus, legislation can either opt for an approach that does not require an answer to the question (e.g. generic risk approach with (limited) derogations as used for plant protection products and biocidal products) or it can determine case by case whether or not a safe (or acceptable) threshold can be quantified and consequently apply an appropriate risk management approach, as done in REACH, where a risk/impact-benefit approach applies if it is not possible to determine a threshold 39 . Certain pieces of sectorial legislation (e.g. Cosmetic Products Regulation, Food Contact Materials Regulation) not only lack specific provisions for assessing EDs but also lack specific guidance on how to deal with EDs for which it is not possible to quantify a safe (or acceptable) threshold.  In practice, in cases where a threshold cannot be established the regulatory approach followed under EU legislation is to minimise exposure as far as possible, including the option to prohibit the use of a substance. Options for consolidation compatible with the Commission’s ambition of “one substance one assessment” should be explored, with the specific aim to minimise exposure to EDs. If the current situation persists, many stakeholders will continue criticising the status quo for its lack of coherence, effectiveness and efficiency (Box 4.3).

4.4.2Management of EDs through water legislation

Chemicals listed as Priority Substances (PS) under the Water Framework Directive (WFD) require EU-wide monitoring as well as measures to achieve the progressive reduction in the emissions of Priority Substances, the cessation of emissions of Priority Hazardous Substances (PHS) in a reasonable time frame, and compliance with the Environmental Quality Standards (EQS). Provisions in the water sector, however, are mostly limited to end-of-pipe interventions, such as increasing wastewater treatment standards. The identification per se of a substance as a priority substance under the WFD does not necessarily trigger any risk management process beyond the WFD, although several pieces of legislation require consideration of the need to meet the EQS set for them. Similarly, the identification of EDs through chemical legislation does not automatically have regulatory consequences under the WFD (see 4.3.4). For PS and PHS substances that are authorised under upstream legislation on chemicals for uses leading to environmental emissions, it is problematic to minimise or eliminate emissions to the aquatic environment. This is a challenge for PS in general, including EDs such as nonylphenol (Annex 5C, nonylphenol case study) and was mentioned by several stakeholders (see Annex 2). The issue reflects a generic lack of coordination between the WFD and risk management measures taken under other upstream legislation on chemicals, as pointed out by the recent fitness check of EU water legislation 40 , despite recent efforts to require better coordination 41 . For example, there is no link to the objectives of the WFD in legislation on human and veterinary pharmaceuticals.

Regulatory efforts to manage chemicals of concern including EDs have been stepped up recently also in other pieces of water legislation. The coherence of the Drinking Water Directive (DWD) with the WFD is especially important as the protection of drinking water resources is established as an indispensable part of the plans and measures under the WFD. The proposal for a new DWD to be adopted by the end of 2020 implements a specific risk approach, requiring further prevention and mitigation measures to protect drinking water sources. It requires a risk assessment of the catchment area(s) to be carried out aimed at reducing the level of treatment required for the production of water intended for human consumption, for instance by reducing the pressures causing the pollution of the water bodies. MS are asked to pay particular attention in their risk assessment to endocrine disrupting substances, and should, where necessary, require water suppliers to also monitor those substances and other parameters included in the DWD watch list if considered a potential risk to human health, and to treat the water accordingly.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

4.5Coherence with non-EU jurisdictions

Q2. To what extent is the regulation of EDs in the EU coherent with international legal obligations (e.g. WTO) and regulatory approaches in other jurisdictions? What are the impacts of incoherencies and overlaps?

4.5.1Differences across jurisdictions

The EU framework for assessing and managing chemicals is recognised globally as the most ambitious in its objectives to achieve a high level of protection for human health and environment 42 . Compared to other jurisdictions, the EU legal framework for testing and assessment of chemicals places a bigger burden on the chemicals industry than on governmental institutions.

Common to many jurisdictions is the fact that the risk posed by EDs is the subject of general chemicals legislation, as well as specific regulatory domains such as plant protection products, food legislation, pharmaceuticals etc. (Milieu Ltd et al., 2017; UNEP IPCP, 2017). Plant protection products are subject to regulations and restrictions in all major global jurisdictions. Assessment values (acceptable daily intakes) applied for pesticide residues are rather uniform across different jurisdictions, which is probably the result of the internationally harmonised procedures for the hazard characterisation of pesticides (Milieu Ltd et al., 2017). The Codex Committee on Pesticide Residues, for example, operating under the umbrella of FAO and WHO, is responsible for establishing Codex Maximum Residue Limits (CXLs) for pesticide residues in specific food items or in groups of food, as well as providing international standards for food safety and fostering international trade 43 . The US Endocrine Disruptors Screening Program uses information on endocrine activity or exposures to prioritise screening and testing programs (EC 2016b). A similar approach is used in the EU for general chemicals. ECHA screens chemical substances in their databases for available information on ED properties, combined with methods such as Quantitative Structure-Activity Relationships (QSARs) and read-across 44 . The screening serves for setting priorities for further regulatory action (for which tonnage information can be taken into account as well), such as setting up the REACH Community Rolling Action Plan for Substance Evaluation 45 or to develop the regulatory management options 46 for a (group) of substance(s).

4.5.2International legal obligations

International laws, such as the Stockholm Convention on Persistent Organic Pollutants (POPs), as well as internationally developed guidelines and standards on chemical safety (e.g. the Globally Harmonised System on Classification and Labelling, and Codex Maximum Residue Limits (MRLs)) do not contain specific provisions on EDs. By establishing specific approaches to assess and regulate EDs, the EU applies higher standards of human health and environmental protection, compared to those established by international standards. WTO rules allow higher standards, provided they are duly notified and scientifically justified. The EU has regularly notified new measures, including changes in legal provisions, restrictions and safety standards on EDs. Starting from 2014, a specific trade concern was raised by Argentina, China, Ecuador, Guatemala, India, Panama, Paraguay, and the United States of America with the support of another 38 members in relation to the EU’s intention to publish a road map outlining different options to assess, classify and regulate EDs in the context of the Plant Protection Products Regulation (PPPR) 47 . WTO members have since then repeatedly and frequently challenged the consistency of the generic risk approach in the PPPR with the obligation of the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS agreement) to base more restrictive measures on “an assessment of the risks to human, animal or plant life or health”. The EU maintained in the meeting of the WTO Committee on Sanitary and Phytosanitary Measures in November 201947 that the criteria for EDs are legitimate and motivated by scientific and societal concerns (low level of tolerance to risk from residues of pesticides in the EU). Moreover, criteria for ED identification are in line with the IPCS/WHO definition. The debate has continued ever since and remains unresolved, although it has not escalated to a formal dispute.

Several WTO members also requested the European Union to harmonise its import tolerances for active substances which were not re-approved with existing levels (MRLs with Codex standards) to allow for trade to continue in line with the SPS agreement. In response, the EU has clarified that whereas the use of substances falling under the exclusion criteria was not allowed in the European Union, import tolerance requests would be nevertheless processed under the MRL Regulation (EC) 396/2005, and a risk assessment would be carried out by an evaluating EU Member State and EFSA47. So far, no MRL has been reduced solely due to a substance meeting the exclusion criteria.

5EFFECTIVENESS

Q3 To what extent has EU legislation been effective in identifying endocrine disruptors and managing risks related to their exposure across different legal frameworks, ensuring the protection of human health and the environment?

It is important to bear in mind that many substances may have more than one hazardous property and indeed this is often the case with EDs since the adverse effect associated with the endocrine mode of action might also be reproductive toxicity or carcinogenicity, including by the same mode of action (e.g. DEHP and other phthalates have been identified as ED and for reproductive toxicity caused by the same ED mode of action). EDs might also be persistent, bioaccumulative and toxic (PBT) or very persistent, very bioaccumulative (vPvB). Risk management measures may already have been taken due to the substance’s other hazardous properties (e.g. phthalates). If these measures are the maximum that can be taken then there is no further impact of identifying the substance as an ED. However, in other cases, the identification of a substance as ED in addition to other toxicities (e.g. due to different effects or modes of action) may lead to a need to apply further risk management measures. In addition, not all EDs for human health will be identified as CMRs and identification as an ED for the environment may also trigger additional risk management measures. In the future, EDs with adverse effects other than reproductive toxicity or carcinogenicity may be identified once methods to detect different related adverse effects and their modes of action are available.

5.1Current process of ED identification

In relation to ED identification the critical point is the availability of suitable, relevant and validated methods to test substances for endocrine disrupting properties. The European Commission supports the development of methods at the international level so that the data generated with the methods will be accepted globally under the mutual acceptance of data (MAD) agreement which is both cost efficient for European businesses and prevents the unnecessary repetition of animal tests. The OECD test guidelines programme has been striving for many years (since the mid-1990s) to develop and validate appropriate tests which investigate possible ways in which the endocrine system can be perturbed and consequently cause an adverse effect. Test guidelines have been developed in laboratory animals (fish, amphibians and mammals) as well as in vitro assays in mammalian cells. The OECD Conceptual Framework for the testing and assessment of endocrine disruptors currently refers to over 30 test guidelines which can be used (together) to investigate endocrine disrupting properties of a substance (OECD, 2018). The methods focus on interference with estrogen (E) and androgen (A) pathways including the process of steroidogenesis (S) (production of steroid hormones such as estrogen and testosterone) and some tests can also identify thyroid (T) hormone disruption, together often referred to as the EATS pathways or modes of action. It is recognised that the toolbox of available tests is not yet complete and intensive work is ongoing through EU-funded research programmes such as the current 5-year EURION project 48 . EURION is a cluster of eight research projects funded with €50 million by the European Commission’s Horizon 2020 Research and Innovation Programme, the largest public funding of this type of research in Europe. Each project in the cluster is focusing on a different aspect of new testing and screening methods identifying EDs. The projects focus on improving methods for the identification of thyroid hormone disruptors, as well as developmental, metabolic disorders and female reproductive disorders caused by endocrine disruption. There is also a project specifically exploring cross-species extrapolation of thyroid effects from fish to mammals.

The Commission is also promoting the adoption of the methods (once validated) as OECD Test Guidelines and has contributed to the development of standardised and internationally agreed test methods through grants to the OECD test guideline programme. For the periods 2015/2016 and 2017/2018 a contribution of around €1 million was given twice towards the development of guidelines for the testing of endocrine disruptors either through grants to the OECD or through framework contracts. The contribution also benefitted the work on Integrated Approaches to Testing and Assessment (IATA) and Adverse Outcome Pathways (AOP), which are expected to play an important role in defining future alternative testing approaches 49 . In addition, the JRC’s EURL ECVAM in collaboration with the network of validation laboratories in the EU Member States (EU NETVAL) 50 , is carrying out the validation of 17 in vitro methods drawn from the OECD scoping document 207 (OECD, 2017) which can then be used to identify substances that interfere with thyroid hormone production and action.

In summary, some endocrine disruptors can be identified using regulatory test guidelines whilst for others, methods are still under development and require validation. These tests are the basis for ED identification (where required) under any existing EU regulatory instruments related to chemical safety assessment. In practice, to date, identification of endocrine disruptors has also relied on information derived from non-regulatory studies in the published scientific literature. In 2016, the European Commission commissioned a study on commonalities and differences in approaches for testing and assessment of endocrine disruptors within the EU and among relevant international trading partners (Brunel University London and DTU Food National Food Institute Denmark, 2016). The study compared the approaches for the regulatory screening, testing and assessments of substances between the legal frameworks, and used cases studies to demonstrate commonalities/differences. The analysis confirmed that the current approach of considering data from scientific literature alongside those obtained by regulatory testing was needed (5 out of the 8 EDs used as case-studies were discovered through research activities, and not regulatory assessment).

Many of the current activities at OECD level and EC level are addressing the concerns of the stakeholders who replied to the survey (Box 5.1).

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

As described in the state of play (Annex 4), there are currently five regulatory instruments that contain provisions related to endocrine disruptors. These are REACH, the BPR, the PPPR, the WFD and the medical devices Regulation.

Criteria for ED identification are laid down for Biocidal Products and Plant Protection Products but not under REACH, which relies directly on the IPCS/WHO definition. The ability to detect endocrine disrupting properties under the three pieces of legislation is limited by the requirements for data generation (see 4.1), which are in the process of being updated for ED endpoints, in particular to include the mechanistic assays (addressing stakeholders concerns (Box 5.2).

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Moreover, the adaptive nature of the endocrine system requires a combination of tests for ED identification, which currently mostly rely on animal testing and require considerable cost and time to produce the necessary data (Browne et al., 2020). Most of the available in vivo toxicity test guidelines currently required, although not specifically designed for the purpose of investigating endocrine disrupting properties, do include some endocrine-related parameters. These tests have to be carried out for biocides and pesticides and higher tonnage REACH substances to investigate toxicity in general, independent of endocrine disruption. However, to reach a conclusion on endocrine disruption, additional tests will likely be required, covering additional evidence of specific interactions with the endocrine system (so-called mechanistic studies). Sometimes this may also include the need to repeat an old study in accordance with updated test guidelines incorporating more sensitive markers of endocrine disruption. The amount of additional testing to be carried out thus depends on the amount of reliable data available for the substance and the data gaps that need to be filled to draw a conclusion.

An example of a potential increased cost due to ED testing would be the difference between the cost of the basic design of an Extended One Generation Reproductive Toxicity Study (OECD TG 443) compared with the additional cost of extending the study to two generations and/or including additional groups of animals to investigate effects of the substance on neurological development. The basic design of the EOGRTS was included in the REACH data requirements in 2015. Extensions are triggered in case of a concern, e.g. for endocrine disruption. In a study commissioned by ECHA on the analysis of capacities and capabilities of laboratories to conduct the OECD TG 443, contract research organisations stated that the cost of the basic study design in 2015 to be on average around 430,000€ (Risk & Policy Analysts Limited, 2015). Extension of cohort 1B (to test for reproduction/development) increases the cost by 20% (median value) compared to the basic study design, the developmental neurotoxicity cohorts 2A & 2B increase the cost by 30%, the developmental immunotoxicity cohort 3 by 15%, and a full study design by 60%. Examples of costs of mechanistic methods specifically designed for endocrine testing, are in 2020, about 170,000 € for the uterotrophic and Hershberger assays (combined). For the Amphibian Metamorphosis Assay an estimation of 75,000 USD is given in OECD GD 150 (version 2012). Regarding in vitro testing (OECD TG 455, 456, 458, OPPTS 890.1200), the average cost in 2020 of each test goes from 2000 to 20,000€ depending on the test or service provider. In relation to the time taken to conduct the test, contract research organisations have recently indicated durations of around 9 months to conduct the mechanistic in vitro tests and around 20 to 24 months for the in vivo assays such as the uterotrophic and extended one generation study in rodents and up to 28 months for an extended one generation reproductive toxicity test in fish (medaka) (MEOGRTS) or a larval amphibian growth and development study (LAGDA) 51 . However, the estimated time periods do not consider possible limitations of testing capacity of laboratories, which is an important factor to take into account when updating data requirements.

Regarding the REACH data requirements for environmental assessment, chronic toxicity data on fish are required for tonnages above 100 tonnes, but will often be based on a fish early life-stage (FELS) test which does not allow for ED identification, as it does not include any ED-related endpoints. Therefore, additional testing would be required for ED identification. In addition, for thyroid-specific effects, an amphibian test would be needed as there is no thyroid-related endpoint in the fish test guidelines. However, this should change as several research projects funded by the European Commission are investigating the possibility to add thyroid-related endpoints to the fish tests.

Although data requirements under the PPPR, BPR and REACH are still in the process of being updated, available or newly generated data sets, including studies reported in the published scientific literature, have allowed some substances to be identified as EDs under each of these regulations as explained further below. 

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Regarding the stakeholder statement that “the best way to minimise the use of animals would be to apply the one substance–one assessment approach, allowing data to be used across regulatory sectors”, it should be highlighted that under REACH and the Biocidal Products and Plant Protection Products Regulations, the sharing of data from vertebrate animal tests is an obligation.

5.2Biocidal Products and Plant Protection Products Regulations 

The PPPR and BPR introduced, in 2009 and 2012 respectively, an approach that active substances identified as EDs cannot be approved – unless very limited derogation possibilities are met. Such an approach is already applied to substances with CMR properties (the so-called exclusion or cut-off criteria; see Section 3 and Annex 4).        

According to EFSA 52 , seven of the 57 ED assessments of active substances conducted so far under the PPPR resulted in the active substances being identified as EDs with respect to human health (Table 5.1). Of the remaining 50, there was no evidence of endocrine disruption for 17, further information was requested for 20 substances, the ED assessment was waived for 12 substances for different reasons and one assessment is still in progress. The 55 assessments conducted with respect to the environment (non-target organisms) concluded that three active substances had endocrine disrupting properties whereas five did not, eight assessments were waived and 39 required further testing.

According to ECHA 53 , the Biocidal Products Committee (BPC) has delivered 17 opinions on biocidal active substances since the criteria came into action (June 2018). In these opinions, three substances were concluded to meet the ED criteria for human health and/or non-target organisms, and three substances were concluded not to meet the ED criteria. Of the remaining 11 substances, the BPC could not conclude due to the absence of sufficient data. However, for assessment reports submitted before 1 September 2013, the evaluating Competent Authority has to conclude based on the already available data and/or the data provided by the applicant. Where this data is insufficient to reach a conclusion, the BPC may conclude in its opinion that no conclusion could be drawn 54 . In addition, when the substance is already non-approved (due to meeting other exclusion criteria) no additional data are requested for the ED assessment.

An additional 12 substances have already been discussed by the BPC Working Groups without having progressed to the BPC. In the Human Health working group, 11 substances have been discussed: 4 substances were concluded to not meet the criteria. For the remaining 7 substances no conclusion could be drawn, although in some cases there could be technical limitations to providing additional data. In the Environment Working Group, 10 substances have been discussed: for 1 substance the ED assessment was waived, while for the remaining 9 substances no conclusion could be drawn due to the absence of sufficient data. In most cases, additional data will need to be generated in order to conclude on the ED properties for non-target organisms, particularly for non-mammals.

Table 5.1: Outcomes of active substance assessments for endocrine disrupting properties under the PPPR (as of August 2020)

Table 5.2: Outcomes of active substance assessments for endocrine disrupting properties under the BPR (as of September 2020)

The experience gained so far demonstrates that for some substances the existing data were sufficient to reach a decision but for many the existing data were insufficient. This confirms that the data requirements for the Regulations need to be updated in order to increase the likelihood that a decision on endocrine disrupting properties can be made on the basis of the data package presented as part of the approval or renewal applications. This should significantly reduce the frequency of the necessity to ‘stop the clock’ to request further information related to ED properties which slows down the regulatory process. An update of the data requirements has been initiated under the BPR and the PPPR as described in the state of play (Section 3 and Annex 4). Since the criteria only came into effect in 2018 it may be considered too early to assess the effectiveness of the ED criteria for the BP and PPP Regulations (a view supported by many of the respondents to the stakeholder survey) – in fact, the Regulations establishing the criteria foresee a review by 2025 (i.e. 7 years after they start to apply). However, 44 PPP active substances considered as EDs according to the WHO-UNEP Report of 2012 have been already withdrawn from the market due to other hazardous properties leading to unacceptable risks to human health and the environment 55 . In addition, many active substance renewal/non-renewal regulations already include consideration of the ED aspect in light of the submitted dossier and criteria.

In summary, since the criteria came in force in 2018 any pending or new request for approval of a new active substance under the BP and PPP Regulations require an evaluation of endocrine disrupting properties. In addition, for active substances currently on the market, renewal of approvals is required on a regular basis (every 7 to 15 years), ensuring that all substances will be re-evaluated in light of new scientific information. This also covers the assessment of endocrine disrupting properties, according to the new criteria, and renewal processes which were already ongoing when the new criteria entered into force but were still pending. Furthermore, re-assessment is also required when new evidence emerges 56 . The data requirements under both regulations are currently being updated to enhance the possibility to identify EDs based on relevant OECD test guidelines within the OECD Conceptual Framework. A (systematic) review of the literature is also required to identify relevant studies from the published scientific literature. Evaluations conducted since 2018 have demonstrated that it is possible to identify EDs, or the absence of ED properties, for a number of substances based on data already available within the dossiers supported by the published scientific literature, which is always considered since evaluations must be based on all available data (see Table 5.1).

5.3REACH

Under REACH there are provisions for restricting or authorising the use of substances with endocrine disrupting properties. The REACH registration database contains 22,877 substances (as of 11 July 2020) 57 . One of the principles of REACH is an increasing requirement for information on hazardous properties in accordance with potential for exposure, based on uses and production levels. The process of substance evaluation, however, allows for the evaluation of any substance based on concerns from existing data.

The ED-related provision is introduced through Article 57(f), which allows endocrine disruptors to be identified as substances of very high concern (SVHCs) if they exhibit serious effects that are of equivalent concern to other substances of very high concern such as carcinogens, mutagens and reproductive toxicants (CMRs) (Category 1A/1B) or persistent, bioaccumulative and toxic substances (PBTs/vPvBs).

Although criteria for the identification of EDs are not defined in the legal text of REACH, this has not prevented substances from being identified as EDs. This has been achieved through agreements by Member State competent authorities to follow the IPCS/WHO definition and use the OECD guidance document 150 on the testing and assessment of endocrine disruptors (OECD, 2018) with their own expertise to evaluate existing data (including non-regulatory studies in the scientific published literature). Through the substance evaluation procedure of REACH, it is also possible to request the registrant to conduct additional studies listed in the OECD Conceptual Framework in order to clarify concerns on toxicity, including endocrine disrupting properties.

As an action under the Community Strategy for EDs from 1999 a study was contracted by DG ENV in which an extensive literature search was carried out to identify substances that had been reported in the literature to be associated with endocrine activity either from in vitro or in vivo studies. Around 500 substances were identified and a process of verification of the validity of the studies was undertaken through expert workshops. Substances were categorised according to the presence of endocrine activity in vitro or in vivo in at least one positive test. Further analysis revealed that many of the uses of these substances were already restricted under EU legislation in relation to their other hazardous properties (e.g. CMRs or PBTs/vPvBs).

This list of substances along with other information (e.g. from the ChemSec SIN list 58 ) fed into the ECHA screening approach 59 , 60 for REACH registered substances, which serves as a starting point for further investigation of ED properties under REACH substance evaluation or for regulatory action such as a restriction or SVHC-identification/authorisation. The starting pool of substances for all ECHA’s screening activities comprises approximately 130 000 substances in ECHA’s database. The screening approach also uses computational tools to identify substances with chemical structures or sub-structures/functional groups related to the structures of known endocrine disruptors as well as searches in registration dossiers of endocrine-related terms including potentially endocrine-related (adverse) effects.

In this manner, substances in ECHA’s database have been screened for potential endocrine activity based on existing data in REACH registration dossiers but also, in some cases, the published scientific literature and those with indications of activity have been taken up by REACH competent authorities for further regulatory action.

Screening information on potential ED properties of pre-registered substances in the tonnage range of 1-10 tonnes per year was also made available by ECHA on their dissemination portal as part of the REACH Annex III inventory. Registrants of low tonnage substances could check prior to registration whether their substances were considered likely to have hazardous properties. Lower numbers of low tonnage substances were registered at the last registration deadline (end of May 2018) than originally expected. While reasons for a lower number of registrations are unknown, it could be hypothesised that the screening information provided by ECHA may have discouraged industry from registering such potentially hazardous substances.

To support the effectiveness of the process of ED assessment an ED Expert Group has been set up to provide an informal consultation body for the Member States, the Commission and ECHA on all regulatory assessments related to EDs under REACH and the BPR. The ED Expert Group helps define the best testing strategy and data to be requested in a substance evaluation decision before a substance enters the formal substance evaluation decision-making process. This happens during the 12-month period given to Member States to evaluate the substance, and therefore does not affect the duration of the overall process. Besides the MS experts, the ED Expert Group includes stakeholder organisations as observers (representing industry, consumer and environmental protection groups, trade unions and animal welfare organisations). According to the endocrine disruptor assessment list on ECHA’s website 61 , the ED Expert Group has discussed 92 substances or group of substances under REACH or the BPR. Many substances are brought to the expert group before being moved to SVHC identification, which has resulted in improved dossier quality and less need for discussion at Committee level. Member State Competent Authorities responsible for drafting the Annex XV dossier can decide to focus on either ED for the environment or ED for human health (or both) since often different agencies with different expertise carry out each type of assessment. This allows Member States some flexibility in deploying their resources. In principle, this could lead to a lack of coherence on the data sets used and reduce the possibility for effective read-across between the human health and environmental data, which should be used together in a weight of evidence approach to reach decisions. However, in most cases Member State authorities consider human health data when assessing environmental properties and vice versa, and there is no evidence showing that separating the assessments is affecting the assessments.

Substances with endocrine disrupting properties are identified as SVHC if the substance is of an equivalent level of concern (ELoC) to CMRs Cat. 1A/1B or PBT/vPvBs as specified in Article 57 (f) of the REACH Regulation. Lack of unanimous agreement in the REACH Member State Committee (MSC) of ECHA on identifying a substance as an SVHC under REACH and the need to refer such cases to the Commission can be seen to have slowed down the process (as highlighted by some stakeholders). However, through the decisions taken by the Commission (after a vote in the REACH Committee) on these disputed cases and by judgements of the Courts in Luxembourg, a common interpretation has started to emerge, which can be applied by the MSC in future cases and should increase the efficiency of the overall process. As of July 2020, 17 substances have been identified as endocrine disruptors of equivalent concern to SVHCs for either human health (5 substances) or the environment (10 substances), plus 2 identified as ED for both human health and the environment. All the substances that have been agreed to be endocrine disruptors by ECHA’s Member State Committee have also been identified as SVHCs (i.e. ELoC) except for resorcinol. In the case of resorcinol, the MSC opinion that the substance is an ED of ELoC has recently been referred to the European Commission’s REACH Committee, along with the minority position disagreeing with the opinion. Some of these substances were already listed as SVHCs and undergoing authorisation due to their other hazardous properties. Of those EDs placed on the candidate list, 6 have been transferred to the authorisation list (Annex XIV). Applications for authorisations are undergoing or have undergone evaluations by ECHA’s Committee for Risk Assessment (RAC) and Committee for Socioeconomic Analysis (SEAC) leading to the granting of authorisations for certain specific uses while other uses have been restricted following the restriction procedure (see Annex 5 – Case studies).

Substances identified as endocrine disrupting substances of an equivalent level of concern to CMRs or PBT/vPvBs include the phthalate esters DEHP, DIBP, DBP and BBP (see Annex 5B, DEHP case study). In fact, regulatory action on DEHP and other phthalates started in relation to the reproductive toxicity of the substances before ED-specific provisions were in place. The additional inclusion of a concern for endocrine disrupting effects on human health in the SVHC list did not directly affect REACH authorisation and restriction procedures since they are based on the same effects that led to the inclusion due to toxicity for reproduction. However, it triggered additional risk provisions for risk assessment (Chemical Safety Report) and risk communication (Safety Data Sheets) and required specific consideration of endocrine effects in reviews of authorisation. DEHP was also identified as an SVHC for endocrine effects on the environment and was added to the Candidate list for authorisation under REACH which also leads to an obligation to obtain authorisations in other sectors (food contact materials, medical devices). It also prompted the re-designation of DEHP from a Priority Substance to a Priority Hazardous Substance under the Water Framework Directive which requires that emissions be phased out.

Other substances identified as SVHCs for the environment due to endocrine disrupting properties and placed on the REACH Candidate list include a number of alkylphenols and alkyl phenol ethoxylates. One of these, nonylphenol (NP), is an intermediate in the production of various derivatives including nonylphenol ethoxylates (NPnEOs) which were used as surfactants and emulsifiers in many different applications including detergents with consequent widespread emissions to the aquatic environment (see Annex 5C, NP case study). They have been progressively restricted over many years due to risks to the aquatic environment, firstly via the 1992 PARCOM 62 convention and then through REACH restrictions together with setting of Environmental Quality Standards (EQSs) and actions taken under the WFD. NP and NPnEOs were identified as SVHCs of equivalent concern under REACH due to endocrine disrupting effects on the environment in 2012 and 2013 respectively and further restrictions were introduced in 2016. They are widely restricted in the EU, with a few speciality biomedical applications currently subject to the authorisation requirements after NPnEOs were placed on the Authorisation List (Annex XIV) of REACH. In recent years increasing scientific knowledge has led to updates of predicted no effect concentrations (PNEC) and EQS values for the aquatic environment. With respect to human health concerns, NP was classified as a category 2 reproductive toxicant in 2001 and consequent restrictions included the addition to the list of prohibited substances in cosmetic products. The closely related octylphenol and their ethoxylates were also identified as SVHCs due to endocrine disrupting effects on the environment in 2011 and 2012, respectively, and are similarly subject to authorisation requirements.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Apart from SVHC identification and authorisation, REACH contains the restriction procedure as a powerful tool for imposing risk management measures including the complete ban of a substance.

A restriction can be put in place if there is an unacceptable risk to human health or the environment that arises from the manufacturing, use or placing on the market of a substance. Examples of restricted substances with confirmed or suspected endocrine disrupting properties are multiple organo-tin compounds, the mentioned octylphenol, nonylphenol as well as their ethoxylates, phthalates, perfluorooctanoic acid and related substances, or substances that are now regulated under the Regulation (EU) 2019/1021 (e.g. decabromo diphenylether). However, only octylphenol, nonylphenol and their ethoxylates are restricted due to their ED properties.

In the targeted stakeholder survey comparing BPR, PPPR and REACH, the large majority of respondents with an opinion stated that REACH is the regulation with the lowest likelihood of identifying a substance as an endocrine disruptor. This view was mainly coming from respondents from public authorities, as well as respondents from academia and civil society organisations. The main two reasons for this ranking were: 1) the few data requirements for low tonnage substances, and 2) the lack of specific information requirements for EDs. Testing for endocrine properties is triggered based on observations coming from repeat dose and reproductive/developmental testing, and chronic tests in aquatic organisms, however, these tests may not be available since REACH applies a tiered approach for information requirements, based on production tonnage. Thus such tests would not be required for low tonnage substances, limiting the possibility to detect toxicological effects (this aspect is not ED specific). In contrast, investigation of endocrine disrupting properties is mandatory under the BPR and PPPR, irrespective of tonnage level.

Some respondents also mentioned that the REACH process could be strengthened by making the BP and PPP ED-identification criteria applicable and formally used by the REACH-SVHC process. This option would be in line with the Commission’s aim to develop a horizontal approach to ED identification. However, it should be noted that the BP and PPP criteria were developed specifically for these two regulations and it needs to be examined whether they can be applied directly for REACH processes. It was also mentioned that the additional requirement within REACH to demonstrate an equivalent level of concern as for CMRs or PBTs/vPvBs when identifying an ED as a substance of very high concern (SVHC) also decreases the likelihood of an identification.

A point was also made in favour of the creation of a European mechanism to enable the European and national agencies, to carry out independent studies on potential EDs. It was suggested that the work should be financed by an increase in the fees charged by the agencies to companies.

5.4Other sector-specific and product-specific legislation

Many pieces of sector-specific and product-specific legislation have provisions to address risks posed by hazardous substances, but do not require the necessary toxicity data to be generated to allow the identification of the hazard. Substances used in some sectorial legislation, such as the Toy Safety Directive, Detergents Regulation, Fertilisers Regulation, Food Contact Materials Regulation, the Cosmetic Products Regulation and the Medical Devices Regulation, are required to be registered under REACH (unless produced/imported at <1 tonne/annum). Some pieces of sectorial legislation rely primarily on REACH for ED identification, others such as food contact materials, cosmetic products and food additives have further requirements for products or sectors they regulate but these do not, as yet, specifically address endocrine disruption.

It is important to bear in mind that many substances may have more than one hazardous property and indeed this is often the case with EDs since the adverse effect associated to the endocrine mode of action might also be classified as e.g. a carcinogen, mutagen or reproductive toxicant (CMR). EDs might also be persistent, bioaccumulative and toxic or very persistent and very bioaccumulative (PBTs/vPvBs). For substances where the adverse effect caused through disruption of the endocrine system has already been identified and appropriate risk management measures have been applied, the effective measures to minimise exposure may already be in place and identification of a substance as an ED would not bring any additional measures. However, this may not always be the case and knowledge that a substance is an ED can increase the level of concern, which may lead to stricter risk management measures being applied.

In product or sector specific legislation with no ED-specific provisions, there are possibilities to evaluate the risks and apply appropriate risk management measures on a case-by-case basis. This may occur when endocrine disrupting properties of a substance become known through other regulatory processes e.g. under REACH or through publication of scientific papers.

When asked for which sectors they think ED identification should be more specifically introduced some stakeholders referred to the WFD (any Priority Substance with known ED properties should be identified as a Priority Hazardous Substance), the Detergents Regulation (identification of EDs especially for the environment should be emphasized), and the toy safety directive.

Several respondents also proposed to adopt a “one substance-one assessment” approach, which would, in their opinion, increase the effectiveness of regulatory procedures while avoiding discrepancies in ED identification. Under this scenario, there would be no need for specific provisions in each of the legislative pieces, but rather a reference could be made to an overarching regulation like REACH which already has provisions allowing for ED identification.

Pieces of legislation where some stakeholders consider that risk management measures for EDs should be more explicitly described include occupational safety and health Regulations, indoor air pollution, industrial emissions, waste and water Regulations. Several stakeholders suggested that ED identification under one regulation should automatically trigger risk management measures for the same substance under other regulations.

When asked about the impact on human and environmental health of either hazard-based or risk-based approaches 63 to decision making, both were viewed positively by the majority of respondents in the stakeholders survey with slightly more support for risk-based approaches. However, some differences emerge when looking at specific stakeholder groups. Representatives from industry (business association and company/business organisation) have a marked preference for risk-based decision making while civil society organisations favour a hazard-based approach. Similar responses were received with respect to both environment and human health.

Water Framework Directive and Medical Devices Regulation

The provisions on EDs in the Water Framework Directive and the Medical Devices Regulation refer to the steps to follow once a substance is identified as an ED but there are no mechanisms within either piece of legislation for ED identification. The Medical Devices Regulation relies on REACH and on the BPR for ED identification. However, there is no provision that requires all substances used in medical devices to be evaluated under REACH for ED properties. Only if the medical device manufacturer is also considered a manufacturer, an importer or a downstream user under REACH, would the related REACH obligations apply (i.e. a chemical safety assessment). The Water Framework Directive (WFD) has also relied largely on REACH for ED identification.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Cosmetic Products Regulation (CPR)

In the case of the Cosmetic Products Regulation (CPR), which does not have specific provisions for EDs, the environmental risks of endocrine disruptors are explicitly covered by REACH but the human health assessments are dealt with under the CPR. Under the CPR, substances classified as carcinogens, mutagens and reproductive toxicants (CMRs) category 1A/1B or 2 are prohibited unless specific derogations are applied (see Annex 4). EDs that are also CMRs would fall into the same category. The SCCS also proposes to follow a specific risk approach for identified or potential endocrine disruptors which might cause other types of adverse effects. The question still arises how to ensure the systematic consideration of endocrine disrupting properties in cosmetic ingredient assessments. Following the review of the CPR 64  the Commission committed to establish a priority list of potential EDs not already covered by bans or restrictions in the CPR for their subsequent safety assessment. A priority list of 28 potential EDs in cosmetic products was consolidated in early 2019 based on input provided through a stakeholder consultation. The list was prioritised and 5 substances are currently under risk assessment by the SCCS. Due to the animal testing ban in the cosmetics sector, a specific issue arises concerning the identification of EDs (see Annex 4) and in particular the need to demonstrate the adverse effect for an ED, which currently requires in vivo testing. This raises a fundamental question about the interpretation of the IPCS/WHO definition, namely whether it is strictly necessary to observe adverse effects in animals, or whether the likelihood of adversity can be extrapolated from effects in non-animal methods. However, the difficulty to conclude on adverse effects due to the animal testing ban is not restricted to endocrine disruption but is also relevant to other endpoints (such as reproductive toxicity). Respondents to the stakeholder survey propose some possible solutions (see Box 5.7).

*Answers are given as a percentage of respondents from each stakeholder group. For more details see Annex 2.

**Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Stakeholder views regarding the effectiveness of the Cosmetic Products Regulation reflect the differences in risk assessment and management where a specific risk approach is advocated by the SCCS compared with a hazard-based approach under the Biocidal Products and Plant Protection Products Regulations, or REACH. This aspect is further analysed in Sections 4.3 and 4.4.

Viewpoints of SMEs and citizens have also been gathered regarding the effectiveness of EU legislation.

In general, respondents to the SME survey consider the regulatory process to identify and control chemicals with endocrine disrupting properties to be effective in protecting people (40-45 agree, 9 disagree, 6-9 neither agree nor disagree) and wildlife (38 agree, 9 disagree, 11 neither agree nor disagree), in improving the functioning of the internal market, and enhancing competitiveness and innovation (30 agree, 10 disagree, 10 neither agree nor disagree).

With regard to the views from the citizens, the majority of respondents (54% to 74%) consider that EU laws do not protect them at all or only to a small extent from exposure to endocrine disruptors across all of the potential exposure sources listed in the survey. The four exposure sources where respondents consider that EU laws protect them the least are personal care products, food contact materials, clothing, and home or office furnishings, floors and paints. Moreover, when respondents were asked which other sources of exposure to endocrine disruptors were of particular concern, the most often mentioned were pesticides, toys, as well as effects due to combined exposure to EDs.

5.5Communication 

Current mechanisms are in place to inform consumers and recipients of articles in general about the presence of substances of very high concern (SVHCs) with ED properties (REACH Article 33). Furthermore, producers and importers of articles containing SVHCs must notify ECHA under certain conditions (e.g. if the SVHC is intended to be released from the product (REACH Article 7). Safety Data Sheets (REACH Annex II) are required to include information on whether a substance has been identified as an endocrine disruptor or whether such a substance is present in a mixture. An update of the provisions for Safety Data Sheets that includes further obligations to report on EDs was recently agreed in the Commission’s REACH Committee and will apply from 1 January 2021. Similarly, under the Biocidal Products Regulation, safety data sheets for active substances and biocidal products shall be prepared and made available in accordance with REACH (Art 70). Moreover, articles treated with a biocidal product shall display a list of the active substances on the label (Art 58).

The respondents to the citizens’ survey feel they lack information on the presence of EDs in products, or that the information is difficult to understand. Moreover, the respondents called for more and better information on EU activities regarding EDs. The European Commission has recently launched a web portal 65 providing a single point of access to all EC activities related to EDs.

5.6Monitoring data

Environmental and human monitoring data are useful for providing evidence on the effectiveness of a risk-reduction measure, as well as providing alerts to rising levels of substances of potential concern and in understanding actual human and environmental exposures including combined exposures to mixtures of substances. EU Member States monitor several EDs in surface waters under the WFD, either because Environmental Quality Standards already exist or to determine whether the substances pose a risk and should be regulated. The European Human Biomonitoring Initiative ( HBM4EU 66 ), is a 5-year project which was launched in 2017 to respond to the need for harmonised information at European level concerning human exposure to chemicals in order to support reliable risk assessment and management. HBM4EU is a joint effort of 30 countries, the European Environment Agency and the European Commission, co-funded under Horizon 2020. The initiative is coordinating and advancing human biomonitoring in Europe. HBM4EU is generating evidence of the actual exposure of citizens to chemicals and the possible health effects in order to support policymaking. A number of EDs such as bisphenols, phthalates and flame retardants have been prioritised under the programme, and data are being made available via the Information Platform for Chemical Monitoring 67 . A similar initiative, albeit smaller in scale, for terrestrial and aquatic wildlife is the LIFE APEX project 68 which focuses on chemical monitoring in apex predators and their prey. A more regular screening of 'unknowns' (i.e. sampling and testing designed to detect unsuspected hazardous chemicals) in humans and the environment is also missing.

Such monitoring was done in the US after the ban on PBDEs at the beginning of 2004. Blood measurements in pregnant women (111) showed a decline in the PBDE concentrations between 2008/09 and 2011/12, but appeared to plateau between 2011/12 and 2014 (possibly due to persistence and bioaccumulation, from dust and food) (Parry 2018). Another study on 334 children, with repeated sample collection from birth to 9 years of age over a 15-year period, found a significant decrease of PBDEs between 1998 and 2013 (Cowell, 2019).

However, an aspect which is not adequately considered with regard to the regulation of EDs is combined exposure, although this issue is not unique to EDs. Provisions for the assessment of cumulative exposure/effects exist in the Maximum Residue Levels of Pesticides, the Biocidal Products and Plant Protection Products Regulations. Two retrospective cumulative risk assessments of dietary exposure to pesticide residues have been recently carried out by EFSA: one considering two chronic effects on the thyroid system and another looking at two acute effects on the nervous system (EFSA, 2020a, b). In both cases, it was concluded that cumulative exposure to pesticide residues did not exceed the threshold for regulatory consideration established by risk managers on the basis of individual substance assessments. Under REACH, four phthalates (DEHP, DBP, BBP and DIBP) have recently been further restricted based on combined exposure estimates, with diet indicated as a signficant source of exposure to DEHP. While EFSA concluded that there is no risk from the three of these phthalates that are authorised in Food Contact Materials, consequential risk management takes into account both combined exposure to the phthalates with the same mode of action and cumulative exposure from other sources. In water legislation, the application of effect-based methods (European Commission, 2014) is a promising development.

Considering the possible non-monotonic dose-response of EDs, it is difficult to predict combined effects from exposure to multiple EDs using standard additivity-based approaches (which assume monotonic behaviour) using information on individual chemicals, and the potential for combined effects adds to the overall uncertainties of a risk assessment (Bopp et al., 2019; Munn and Goumenou, 2013b).

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Although only a few substances have been formally identified and controlled due to their endocrine disrupting activity, many substances with endocrine disrupting properties have already been identified due to other hazardous properties which may or may not be related to endocrine disruption (e.g. reproductive toxicity or persistent, bioaccumulative and toxic (PBT)/very persistent, very bioaccumulative (vPvB) properties). Consequently, regulatory measures to reduce exposures to some of these substances have already been taken and environmental monitoring and human biomonitoring have been used to follow the trends in exposure to such substances over time.

The case study on DEHP (Annex 5B, DEHP case study) illustrates through human biomonitoring data (levels of phthalate metabolites in the urine) how the successively increased restrictions on DEHP and other low molecular weight phthalates over time have been successful in reducing exposure of the human population in the EU. With respect to nonylphenol (Annex 5C, NP case study) there is evidence of substantial decreases in aquatic concentrations in EU wastewater, surface waters and biota in the last two decades. For instance, concentrations in fish (bream, muscle tissue) caught in Germany diminished by 67% between 1995 and 2001. However, reduced concentrations of NP in the aquatic environment cannot be linked to changes in estrogenic activity of water bodies because this is mainly driven by more potent natural and synthetic estrogens.

One concern is that of ‘regrettable substitution’ where the restricted substance is replaced by another substance able to provide a comparable technological function but which might also be similarly hazardous, although the substance is not so well tested so these properties are not known at the time of the replacement. This could reduce the effectiveness of the measures. To reduce the possibilities of this occurring there is a move towards group assessments, where substances with similar molecular structure, with similar properties or with similar functions are assessed together. This requires either the generation of comparable data for each substance in the group or the use of ‘read across’ approaches. Indeed, regulatory action on phthalates did follow a group approach whereby a group of low molecular weight phthalates (DEHP, DBP, DIBP and BBP) were assessed together.

A number of Persistent Organic Pollutants (POPs) have endocrine disrupting properties and inclusion in the POPs Regulation defines requirements for reporting environmental release inventories (Article 6), environmental monitoring (Article 10) and production volumes (Article 12). Release should be minimised or eliminated from all sources (Article 6) including waste (Article 7). Waste containing or contaminated with POPs should be handled to ensure no release or recirculation in recycled/reused material flows. Recent reviews describing spatial and temporal trends of some of these substances in the environment provide an opportunity to evaluate the effectiveness of the regulatory measures in relation to exposure reductions. One example looking at levels of flame retardants (PBDEs) in human breast milk, cord blood and placentas found levels of PBDEs in North America were substantially higher than those in many regions of Europe, Asia, Oceania or Africa. One possible explanation given for this was the higher use of flame retardants in furniture in North America (Tang and Zhai, 2017).

5.7Human health and environmental impact evidence and indicators

The trends in the main health and environmental impact parameters that are known, or suspected, to be associated with exposures to endocrine disrupting substances are important considerations when examining the effectiveness of EU chemicals policy. These trends include upward trends in the incidence rates of certain hormone-related cancers and reproductive diseases, decreasing fertility rates in females and males and decreasing sperm counts, and on the environmental side reduction in species diversity and eco-system health/resilience. However, caution needs to be exercised when using human health and environmental adverse effects as direct and reliable indicators of chemicals policy performance. This is because of the attribution challenge: many of the observed health and environmental adverse effects may derive from multiple causes (life-style, genetics, habitat destruction/degradation, etc.) and it is difficult to determine to what extent exposure to endocrine disrupting substances contributes to the observed adverse effects 69 . Statistical approaches exist and have been applied in (eco)epidemiology with the aim of separating the contribution of (mixtures of) chemicals from other contributing or confounding factors (Gennings et al., 2018; Slama et al., 2017). Differences in study design and low reproducibility of findings, however, remain a major challenge for use in a regulatory context.

Complicating things further is the fact that observable adverse effects in human health and the environment often do not materialise immediately after exposure. For example, the effects of early life exposure to endocrine disruptors may not materialise until many years later. Moreover, the involvement of the endocrine system in the control of processes such as embryonic development and puberty means that young people are at higher risk regarding endocrine disruption.

Indeed, developing organisms are extremely sensitive to perturbation by chemicals with hormone-like activity, and adverse effects may be most pronounced in the developing organism and occur at concentrations of the chemical that are far below levels that would be considered harmful in the adult. The continued maturation of key endocrine systems during childhood and adolescence, make these life periods also particularly sensitive to endocrine disruption, beyond foetal development (Schug et al., 2011). Many of the adverse effects on development in late pregnancy and early childhood are expressed as functional deficits in organs or systems, instead of overt malformations or growth retardation (Schug et al., 2011). Evidence from many studies support the hypothesis of a relationship between exposure to EDs and early maturation of secondary sexual characteristics (e.g. precocious puberty) (Lee et al., 2019; Lucaccioni et al., 2020) or predisposition to breast cancer (occurring later in life after exposure during puberty) (Lucaccioni et al., 2020). Exposure to EDs during early development have also been associated with non-reproductive effects such as obesity and related diseases (Petrakis et al., 2017). It is important when considering these vulnerable groups to ensure that the sensitive life stages are covered by the methods used to investigate toxicity and endocrine modes of action.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

A few attempts have been made to estimate the costs of endocrine-related diseases/disorders in humans and the proportion that might be attributable to exposure to endocrine disrupting substances.

Four studies published in 2015 (Bellanger et al., 2015; Hauser et al., 2015; Legler et al., 2015; Trasande et al., 2015) estimated the costs associated to the use of several endocrine disruptors for a subset of health effects to be hundreds of billion euros per year. Those studies subsequently received criticism from other scientists, claiming their aim would be to shape policy (Middelbeek and Veuger, 2015). Further criticism related to alleged methodological flaws and the hypothesis on which the studies were based (Bolt, 2017; Bond and Dietrich, 2017a, b; Middelbeek and Veuger, 2015). The main problem with such studies is the attribution challenge (i.e. that EDs are responsible for causing several diseases for a certain minimal percent factor of probability). The Commission reviewed the four studies when carrying out an impact assessment in the context of setting criteria for endocrine disruptors under the BPR and PPPR 70 . Other studies have tried to estimate costs linked to effects of endocrine disruptors on male reproductive health (Olsson and al., 2014).

Nevertheless, there are examples of non-approval of substances with endocrine disrupting properties resulting in significant benefits to human health and wildlife. For instance, the restriction of the use of the endocrine disruptor tributyltin (TBT, REACH Annex XVII, entry 20) as an antifoulant in marine paints has resulted in the recovery of mollusc populations in many ports and coastal areas in Europe 71 .

5.8Market surveillance

An aspect which could weaken the effectiveness of EU legislation is differences compared with laws in non-EU countries. The potential import of substances banned in the EU from non-EU countries raises the issue of EU market surveillance.

The respondents to the stakeholder survey mentioned that some substances banned in the EU but still allowed in other countries can still be found in products imported and marketed in the EU. Such statements are supported by an enforcement project conducted by ECHA’s Forum, which is a network of national authorities responsible for the enforcement of REACH as well as the CLP, Biocidal Products and PIC Regulation. The project investigating compliance with the REACH restriction on phthalates in toys and child-care articles found that 19.7 % of the inspected toys and 3.6% of childcare articles contained one of the phthalates above the permitted level. Most products found to be non-compliant came from countries outside of the European Economic Area or were of unknown origin (Annex 5B, DEHP case study).

Moreover, these substances can be found in the environment (e.g. water bodies, including by transboundary flow). Regarding the residues of pesticides found in imported products, some stakeholders suggested to set the maximum residue levels (as a general principle) at the level of quantification for substances falling under the cut-off (exclusion) criteria.

Respondents to the stakeholders’ survey consider that all sectors would benefit from an increased market surveillance. In particular, the fields of: feed and food (environmental contaminants, natural toxins, flavouring agents); water, including drinking water; industrial emissions; the General Product Safety Directive (GPSD); products in contact with vulnerable groups, and online sales were listed by stakeholders as those where market surveillance authorities should focus their activities. However, this aspect is not specific to endocrine disruptors and it has been recognised that its implementation depends on the resources available in Member States 72 .

5.9Functioning of the internal market, competitiveness and innovation

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

6EFFICIENCY

Q4. To what extent has EU legislation been efficient in identifying EDs and managing risks related to their exposure across different legal frameworks, ensuring the protection of human health and the environment?

There is evidence that a lack of coherence across regulatory sectors has had an impact on efficiency. This is particularly the case when multiple regulatory assessment and management procedures have focused on the same substances. Examples are given in the case studies (Annex 5).

In other cases, a lack of efficiency can be attributed to differences in the interpretation of terms in legislation. For example, under REACH, different views of Member States in ECHA’s Member State Committee (MSC) on the interpretation of 'equivalent level of concern' (Art. 57(f)) has resulted in a number of cases 73 where the MSC could not reach unanimity on the identification of a substances as substance of very high concern, meaning that the decision had to be referred to the Commission via its REACH Committee (EC 2018a, part 1/7).

The targeted stakeholder survey revealed a range of opinions on the efficiency of current regulatory procedures for assessing and managing EDs (see Box 6.1). A slight majority of respondents (53%) thought that the regulatory framework is not flexible enough to take into account new scientific information and methods in the assessment of endocrine disrupting properties. Some commented on the need to give more weight in the evaluation process to non-standard sources of information (e.g. non-guideline methods, scientific literature findings). Others argued that the substance-by-substance approach is hampering the assessment process, which would be more efficient if grouping approaches were more widely implemented. Several respondents referred to the time lag between hazard identification and risk management, as well as perceived inefficiencies resulting from the time taken to officially adopt test methods, to update standard information requirements, or to update guidance documents.

The costs and benefits for human health and the environment resulting from regulatory requirements to assess and risk manage chemicals, including endocrine disruptors, are difficult to quantify and compare. Regulatory decision making requires that besides impacts on human health and environment, economic and social impacts are assessed and considered. This is to provide regulation at minimum cost but with maximum benefits to citizens, businesses and workers, as outlined in the Commission Better Regulation guidelines 74 . Costs of regulatory testing are discussed in Section 5.1.

REACH incorporates these principles by requiring a socioeconomic analysis for applications for authorisations as well as for restriction proposals. Socioeconomic analysis is a tool to provide decision-makers with the necessary evidence and analysis on the socioeconomic impacts of different options for how to regulate chemicals. The REACH authorisation procedure requires applicants to submit a socio-economic analysis (ECHA, 2011). The ECHA Committee for Socio-economic Analysis (SEAC), which consists of experts nominated by Member States, scrutinises the socioeconomic analysis carried out in the authorisation and restriction processes. Stakeholder organisations participate in SEAC meetings as observers. Some stakeholders have claimed that the socioeconomic analysis for authorisations does not yield satisfying results and has methodological limitations, such as the choice of discounting rate (Chemsec, 2019) 75 . However, the socioeconomic analyses applied in regulatory processes use sensitivity analyses to assess the impact of different discount rates. These sensitivity analyses have shown that use of lower discount rates would not have resulted in a different policy choice for any of the REACH restrictions or authorisations reviewed.

In view of the challenge of conducting cost-benefit analyses, one recommendation voiced in the stakeholder survey was that the EU should establish a population-based monitoring scheme to assess the long-term benefits of regulatory action.

In the targeted stakeholder survey, 88% of respondents (53 out of 60) representing companies reported an increase in costs related to compliance with regulatory requirements for EDs (Targeted Stakeholder Survey). The additional costs were most often (21 out of 68 respondents) attributed to the provision of test data on endocrine disrupting properties. Costs related to substitution efforts were reported by 19 out of 65 respondents. According to some stakeholders, such efforts have been hampered by a lack of voluntary (non-regulatory) incentives, as well as lack of expertise, tools and guidance for avoiding regrettable substitutions (Ministère de la Transition Ecilogique et Solidaire, 2019). Other costs reported by stakeholders were related to the development of new testing methodologies and the regulatory reporting of ED properties. According to one industry stakeholder, the lack of a globally consistent approach to assessment can lead to additional costs exceeding 10%. In general, however, most respondents did not specify the origin or magnitude of additional costs.

In the SME survey, an increase in total operating costs was reported by 25 out of 70 respondents, whereas 12 respondents reported no effect on operating costs. For the remaining 33 respondents, this question was either not applicable (30) or no answer was provided (3). Costs were attributed to: a) the replacement of substances (21 respondents); b) the preparation of registration or authorisation dossiers (14 respondents); c) the provision of test data (14 respondents); and d) the development of new testing methodologies (13 respondents).

Stakeholder opinions on whether the additional costs are justified varied. Governmental and civil society organisations typically considered the costs justified, in the overall interest of public health and environmental protection. Conversely, industry organisations and associations expressed concerns that the hazard-based management of chemicals results in a competitive disadvantage for EU operators, which is unjustified in exposure scenarios where the substance poses a low risk.

In the SME survey, the additional costs were considered justified and proportionate for the benefits obtained by 23 respondents, not at all justified or proportionate by 8 respondents, while 38 did not know.

Overall, stakeholders who expressed an opinion considered the impacts of regulatory provisions for EDs on their sectors to be negative (Targeted Stakeholder Survey). Among the regulators, 80% of respondents (51 out of 64) reported that assessing substances for endocrine disrupting properties resulted in a delay in their assessment work in other areas of human health or environmental protection.

In the SME survey, impact of the provisions for endocrine disruptors on innovation, productivity, profitability and international trade within their sectors was perceived as negative by a minority of respondents (between 1 to 6) while another minority (7 to 10 respondents) considered the impact as positive.

In the public survey, 26% of the 474 respondents considered that the costs of EU laws on endocrine disruptors are fully proportionate for the benefits accrued, 18% to a moderate extent, 15% to a small extent only, and 11% not at all. Thirty percent replied that they do not know. In addition, there was a general perception that slightly higher costs are borne by the agricultural and industrial sectors compared with costs to citizens or ethical costs.

In principle, differences in regulatory assessment and management procedures between the EU and non-EU countries could have (positive or negative) impacts on trade. However, it is very difficult to predict such impacts. Retrospective assessments using economic methodology can analyse trade flows before and after regulatory interventions, but the results must be interpreted with caution, as illustrated by two case studies carried out by the JRC (see Box 6.2 and Canzian et al., 2020).

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

7RELEVANCE

Q5. To what extent do the EU legal provisions on and approaches to EDs take into account health and environmental concerns, and social and economic consequences that are relevant to citizens and stakeholders?

7.1Relevance to citizens and stakeholders

The overall objectives of human health and environmental protection enshrined in EU chemical legislation remains relevant to EU citizens. According to a Eurobarometer survey conducted in December 2019 76 , nine in ten Europeans are worried about the impact of chemicals present in everyday products on the environment; a slightly smaller proportion (85%) is worried about impacts on their health (the survey question concerned the overall impact and did not query ED effects specifically). Societal concerns on EDs have received attention and political support in the recent resolution by the European Parliament 77 and the opinion of the European Economic and Social Committee 78 and the European Committee of the Regions 79 .

Perspectives on the importance of regulating EDs are polarised between stakeholders representing citizens and civil society organisations and those representing business organisations (see Box 7.1). The discrepancy between the views of citizens engaged on the topic and business organisations appear to reflect different opinions about how to apply the precautionary principle and how to deal with the uncertainties in quantifying risks associated with exposure to EDs (such as the issue of whether EDs have a threshold effect or not). All parties, however, agree on the need to enhance the exchange on the latest science (e.g. ED stakeholders forum) and the collaboration among authorities and between authorities and stakeholders.

The Commission and the European Agencies regularly involve citizens and other stakeholders through public consultations at different steps of the policy cycle from new legislative proposals, to scientific evaluation and decisions, including EFSA opinions, biocide evaluations, REACH SVHC proposals, restriction regulations, and authorisations decisions.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

Q6. To what extent do the EU legal provisions on and approaches to EDs take into account evolving needs (e.g. transition to circular economy) and the latest scientific findings (e.g. developments in chemical mixtures and grouping assessments, biomonitoring, effect based methods, etc.)?

7.2Ability to adapt to the evolving societal and economic context

Societal concerns on chemicals in general and on EDs reflect the high expectations of Europeans for ambitious health and environmental policies. The EU is a global leader in environmental and consumer protection policy. The political agenda set by the Von der Leyen Commission aims at firmly setting the EU on a new path of sustainability, inclusive growth and healthy lifestyles. The action plan outlined in the European Green Deal 80 comes with significant economic and societal transformations. Many core elements of the Green Deal, including the zero-pollution ambition for a toxic-free environment, the farm to fork strategy, clean and circular economy 81 , and the protection of ecosystems and biodiversity are interlinked with the chemical policy agenda in general and with EDs in particular. Another societal trend relevant to ED management is the decline in population growth due to low fertility rates across much of the EU 82 . In parallel to these trends, changes in global patterns of production and consumption have accelerated in recent years. Global chemical production has grown in recent years and is expected to almost double from 2017 to 2030 (UNEP, 2019). Projected growth will be highest in Asia, with China estimated to account for almost 50 percent of global sales by 2030. The imports of consumer goods and other articles into the EU have tripled between 2000 and 2015 creating additional challenges for managing the risks associated with the presence of hazardous substances in articles 83 .

Many functional chemicals that have raised concerns for their negative impact on the circular economy, including plasticisers (e.g. DEHP and other low molecular weight phthalates), flame retardants (e.g. polybrominated diphenyl ethers), water repellents (e.g. perfluoroalkyl substances), preservatives and other biocides, have been assessed for their endocrine disrupting properties (Milieu Ltd et al., 2017) (sub-study b). The intentions to strengthen the single market for secondary raw materials and in general the inclusion of circular economy considerations into chemicals risk management will require a transformation of the life cycle stages and timescales considered in risk assessment83. Exposure data in REACH dossiers fail in many cases to cover all relevant exposure situations along the supply chain 84 . Exposure information on substances in waste and in recycled material streams is even more limited. Both information gaps are evidently at odds with the circular economy ambitions: it is difficult to conclude on long-term safety and sustainability of substances of concern (e.g. SVHC) without the ability to track their mass flows in material cycles and in the environment.

Hence, efforts to control hazardous chemicals (e.g. SVHC) in materials and products and recycled material streams need to step up to achieve the objectives of the circular economy 85 . The scale of the problem is significant and regards both legacy chemicals present in recycled material streams (Milieu Ltd et al., 2017) (sub-study b) and chemicals produced and used today, specifically those authorised for specific uses, if they are not managed in closed and controlled material cycles. The problem is also likely to increase, considering that the amount of recycled materials used in the EU needs to increase substantially from the current 12%Error! Bookmark not defined.. The EU Circular Economy Action Plan further stresses the need for action to ensure that the EU does not export its waste challenges to third countries and to facilitate preparing for re-use and recycling of waste in the EU 86 . SVHCs in recycled materials have been regulated through REACH authorisation and restrictions, as was the case for authorised uses of recycled PVC containing DEHP and for the exemption of recycled textiles from the restriction of nonylphenol ethoxylates (Annex 5B and 5C, DEHP and NP case studies). In other cases, however, and especially for heterogeneous waste, information on the presence of substances of concern is lost or not available to waste operators by the time products become waste. The establishment of the SCIP database on “Substances of Concern In articles as such or in complex objects (Products)” under the waste framework directive is an important first milestone in this direction as it intends to help waste operators in waste separation 87 .

7.3Ability to adapt to scientific progress

Recent policy evaluations concluded that the framework has generally been able to adapt to the evolving scientific progress 88   89 . The question is particularly important in the context of EDs, considering the increasing resources dedicated to scientific research on EDs over the last two decades 90 . All three case studies (Annex 5) provide examples of recent regulatory assessments including reviews of all available evidence in the open scientific literature. Academic studies, however, often miss to include critical parameters as described in standard guidelines and are consequently given a lower consideration in weight of evidence assessments (e.g. Annex 5A, 3-BC case study).

The consideration of new scientific evidence is usually more problematic for academic studies on human biomonitoring, (eco)epidemiological studies, studies addressing exposure to mixtures, biomarkers of effects, or effect-based methods in ecotoxicology. Current standard test and assessment methods typically on single substance assessments which do not fully exploit the value of these types of studies 91 .

Long-term low dose exposure to ED mixtures is a well-recognised challenge in both regulatory toxicology and ecotoxicology 92 . The scope of mixture assessment and management within the existing framework is mostly limited to intentional mixtures within single policy domains, even though the potential for aggregate and combined (mixture) exposures across policy domains is well known (Figure 3.1).

Unintentional mixtures are considered in some legislation, including for pesticide residues (MRLs) plant protection products and biocidal products. However, practical examples are mostly limited to the legislation on MRLs, where methodology is so far established for two cumulative assessment groups (effects on nervous system and on thyroid) and some recent examples of grouping assessments and management processes (e.g. phthalates regulated under REACH and FCMs, PFAS, etc.) (Annex 5B, DEHP case study). An increasing number of academic studies explore how to incorporate assessments of complex and realistic ED mixture exposure scenarios in regulatory assessments (Bopp et al., 2019; Ganzleben et al., 2017; Gennings et al., 2018).

Translating new scientific findings about the effects of combined exposure to EDs into regulatory action conflicts with a framework designed to assess one substance at a time and to implement risk management measures across multiple regulated sectors. Addressing this challenge might require a complete rethink of the way in which risk assessment and management of chemicals mixtures is currently performed (Bopp et al., 2019). One promising advancement in the monitoring field is the development of effect-based methods, which might be used in combination with effect-directed analysis in water quality assessment under the WFD (Brack et al., 2017; Hamers et al., 2018). The oestrogen receptor transactivation assay has been recommended by the JRC as a suitable in vitro method to determine estrogenic activity of water samples (European Commission, 2014). Other in vitro assays to determine (anti)androgenic and thyroid hormone disruption have been also proposed for both environmental (Hamers et al., 2018) and human health applications (Hamers et al., 2020). The method opens potential synergies between the water and chemicals legislation in the assessment of environmental effects of EDs. Similarly, in human biomonitoring some integrative biomarkers of effect have been proposed as an early marker of endocrine-mediated health effects (Baken et al., 2019).

A major scientific challenge in estimating the impact of chemical stress is the multi-factorial nature of negative trends of concern such as the increase of non-communicable diseases (Slama et al., 2017) and the decline in biodiversity (Posthuma et al., 2016). To disentangle the contribution of multiple stress factors, chemical mixture exposure must be assessed in combination with other factors, referred to as confounding factors in (eco)epidemiological research. Statistical methods to account for confounding factors have been developed and applied in recent (eco)epidemiological research addressing EDs exposure (Demeneix and Slama, 2019; Gennings et al., 2018; Slama et al., 2017).

The next generation of EU funded research is taking a more holistic consideration of environmental stressors on disease burdens. Examples include the Health Environment Research Agenda for Europe (HERA 93 ) and the candidate European Partnership for Chemicals Risk Assessment 94 . The Commission is exploring new mechanisms to distil emerging streams of scientific evidence into the policy cycle (e.g. HBM4EU 95 and the candidate European Partnership for Chemicals Risk Assessment). The issue goes beyond the well-known obstacles encountered in the development and validation of novel test methods.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

8EU ADDED VALUE

Q7. What is the added value of regulating EDs at EU rather than Member State level?

Regulating EDs at EU rather than Member State level is adding value in various ways. The Commission evaluations, carried out recently across various pieces of legislation for different sectors (e.g. REACH, non-REACH chemical legislation, plant protection products, etc.), have all concluded that a unique chemicals legislation centralised at the EU level has in general proven largely successful in terms of the protection of human health and the environment as well as the functioning of the internal market 96 .

8.1Joining forces and efficient use of resources

The assessment of EDs requires a high level of scientific expertise and experience. The sharing and pooling of knowledge and resources and the application of common rules and standards across the EU has resulted in significant positive economic, health and environmental impacts that would not have been possible to achieve on the basis of legislation at the Member State level alone, as concluded for chemicals in general in the Fitness Check of non-REACH Chemicals legislation 97 . Thus, regulating the risk assessment and management of EDs at EU level also increases efficiency. It helps to avoid duplication of efforts at Member State level and improve the state of knowledge, quality and availability of data needed for risk management decision making. It also provides a clearer overarching framework for companies operating across different Member States, so that they do not need to adapt to multiple potentially diverging national rules. An EU level approach offers advantages in terms of effectiveness by avoiding a fragmented approach in a market where firms are increasingly cross-border in their outlook 98 . The costs of achieving the objectives to protect human health and the environment are considered to be lower than in a system where each Member State conducts the risk assessments on its own (Ecorys, 2018). A uniform EU approach also strengthens the EU’s position in chemicals regulation worldwide, which also has benefits for the EU chemical industry.

Reaching a consensus at EU level in regulating EDs also plays a role in preventing unfair competition between Member States and leads to an equal level of protection for citizens and the environment throughout all Member States. Since the adaptation of sector-specific legislation to new scientific evidence regarding risks of endocrine disruptors requires a certain amount of time, measures for protecting health and the environment can be taken by more general provisions in EU legislation, namely under the General Product Safety Directive in the case of products, and under the General Food Law in the cases of food and feed.

8.2Preserving the functioning of the single market

There have been cases of unilateral measures at Member State level for a few ED substances. Bisphenol A is the most relevant example; its use has been banned in specific consumer products aimed at babies or young children in France, Denmark, Austria, Belgium and Sweden – and for all age groups in France. Some Member States, such as Italy and Denmark, have also taken ‘soft’ measures at the national level, such as providing advice to the public, creating incentives for industry for the development of safer alternatives or the promotion of voluntary agreements (Brunel University London and DTU Food National Food Institute Denmark, 2016; UNEP IPCP, 2017). In 2011, Denmark proposed to restrict the use of four phthalates (DEHP, BBP, DBP, DIBP) in consumer articles in the EU. However, ECHA's Risk Assessment Committee concluded that the "available data does not indicate that there is currently a risk from combined exposure to the four phthalates" and did not support the restriction 99 . In the following, Denmark subsequently planned to ban the four phthalates unilaterally. However, it dropped the plan in 2015 since it was not in line with the rules set by REACH (ECPI, 2014). The safeguard clause of REACH stipulates that a Member State can take unilateral measures if it has grounds for believing that urgent action is needed to protect human health or the environment. This triggers a decision taken via Comitology procedure on whether the unilateral measure should be revoked or authorised for a certain time period. This decision has to be taken within 60 days. If the measure is a restriction and if it is authorised, the Member States which took the unilateral decision should initiate the procedure for an EU-wide restriction under REACH. Similarly, unilateral action by MSs that contradict EU rules must be followed up at EU level for FCMs, for example in the case of bisphenol A.

Moreover, after considering results from FCMs inspection and control project, on September 2019, the Danish Ministry of Environment and Food announced that the Danish government will ban the use of any per- and polyfluoroalkyl substances (PFAS) in paper and cardboard used in food contact materials by July 2020 100 due to inter alia the concerns on the suspected endocrine disrupting properties for human health and the environment and evidence of bioaccumulation. The European Commission published a recommendation on a coordinated control plan to determine the prevalence of certain substances migrating from or present in FCMs—including the fluorinated compounds per- and polyfluoroalkyl substances (PFAS)—into food, in May 2019 101 . Notably, PFOA has been restricted for use in the production and placing on the market of articles, including FCMs from July 4, 2020 under REACH. Regulatory action on multiple other PFAS has been launched or already been concluded (REACH Regulation, Regulation (EU) 2019/1021).

More recently, France has also launched, based on its Second National Strategy on Endocrine Disruptors key measures ( Ministère de la Transition Ecologique et Solidaire et Ministère des Solidarités et de la Santé", 2019) (Annex 1), a project to draw up a list of chemicals that may present ED-properties to better manage risks depending on the degree of proof and the level of uncertainties (to be ready by 2021). This is intended to inform recommendations for appropriate management (for example, by considering a recognition in European regulation, by reducing use as a precaution, or by pursuing more in-depth assessment) of suspected or presumed EDs. This includes also a recommendation to inform citizens about the chemicals, notably endocrine disruptors, to be found in everyday consumer products.

Unilateral measures are triggered by provisions (safeguard clauses) that are included on purpose in legislation to allow Member States to act first. There is insufficient evidence to estimate the overall impact of the unilateral measures described above. However, such unilateral measures should be followed up at EU-level as fast as possible to minimise any disruptive effects on the single market.

The stakeholders answering the survey expressed a strong wish to avoid unilateral measures where possible. If they are taken by individual Member States as a precautionary measure, e.g. to avoid delays in finding agreement at EU level, such measures should be followed up at EU level. Even soft measures such as publishing advice on the use of products containing certain chemicals can have a strong influence on consumer behaviour. However, a Eurobarometer survey (Nr. 456) on Chemical Safety of 2019 102 showed that EU citizens consider products manufactured in the EU to contain safer chemicals than those imported from outside the EU. This indicates a higher level of confidence in the EU regulatory framework for manufactured products compared to regulatory regimes abroad 103 (EC 2019).

Also in the public consultation related with this Fitness Check, 6 out of 474 respondents 104 provided free text comments related to the EU added value. Even if they are mostly in favour of regulating EDs at EU level in a uniform way across Member States, they raised two issues. Firstly, they state that there is no full implementation of EU rules in each Member State and putting EU legislation into practice need to be accelerated. Secondly, citizens seem to be worried about imported products from non-EU countries with less stringent rules and ask for more controls and enforcement to ensure imported products comply with EU requirements.

8.3Legislative measures alignment between Europe and global regimen

The EU chemicals legislation has become a reference point for international standards in several areas, which helps to reduce potential trade frictions as well as address transboundary chemical related issues. REACH for example has influenced legislation in a few third countries (e.g. South Korea), although significant differences exist and there is room to further exploit the potential of REACH to serve as a global model for chemicals legislation 105 .

European companies also benefit from the perceived quality of EU products in non-EU country markets, which has brought important advantages in terms of international trade. The EU chemicals legislation has also helped to decrease the barriers to, and costs, of intra-EU trade by limiting the application of multiple and potentially diverging national rules with limited territorial coverage and existing only in the applicable national language(s) 106 .

However, there are also concrete examples of problems at international level, e.g. in the area of Plant Protection Products and Maximum Residue Levels (MRLs). Non-EU countries often remind the Commission that non-approval decisions and setting of MRLs need to respect WTO principles. The issue of cut-off criteria, including the criteria to identify endocrine disruptors, has been the subject of acrimonious discussion in the WTO-TBT and WTO-SPS Committees since 2013. Between 2015 and 2017, in the WTO Committees, non-EU countries raised specific trade concerns on pesticides 208 times against the EU. This should be compared with two specific trade concerns on pesticides raised by the EU against other countries. Although decisions under the MRL Regulation are based on assessments of risk only, the effects of the cut-off criteria applied under the PPP Regulation with subsequent lowering of MRLs, is perceived by third countries to result in inconsistency between the EU and WTO. However, the extent of this inconsistency is currently unproven as so far no active substance has not been approved (or approval not renewed) based solely on the cut-off criteria 107 .

*There were no clear trends according to stakeholder group. This is a summary of the most frequently mentioned free text comments.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

*Respondents to all three surveys (citizens, stakeholder and SME surveys) were self-selected and thus not necessarily representative of the full range of views of these groups.

9CONCLUSIONS

Endocrine disruptors (EDs) are chemical substances of synthetic or natural origin that adversely affect the health of humans and animals by altering the functioning of the endocrine system. Exposure to EDs can occur from different sources, such as pesticide residues in food or everyday consumer products. Different regulatory approaches for managing the risks posed by EDs exist because sector-specific regulations have been developed at different points in time and in some cases incorporate different specific considerations. This raises questions about the overall coherence of the EU’s legal framework in relation to EDs.

In this Fitness Check, the current ED-relevant provisions have been mapped across the EU Chemicals Acquis, and the overall regulatory framework has been analysed in terms of its coherence, effectiveness, efficiency, relevance and EU-added value.

Knowledge about endocrine disruptors

Significant progress has been made in understanding EDs since the adoption of the 1999 Community Strategy for Endocrine Disruptors. Much of this knowledge relates to chemicals that affect the estrogen, androgen or thyroid axes or that interfere with steroidogenesis. The need to elucidate additional mechanisms of endocrine disruption is being addressed via EU-funded research projects, such as those in the EURION cluster.

The next generation of EU funded research will aim to address realistic mixture (combined) exposure scenarios and to embrace the multi-factorial nature of certain indicators of concern, such as the increase in certain non-communicable diseases and the decline in biodiversity. Building on the currently available methods developed in key research areas (e.g. use and interpretation of human biomonitoring data and effect-based methods in water quality assessments) the Commission is exploring new mechanisms to use emerging knowledge in the policy cycle (e.g. the candidate European Partnership for Chemicals Risk Assessment).

There is also a need to share information on chemicals with all stakeholders, preferably via an open platform.

Coherence

The coherence of the regulatory framework was assessed by focusing on the definition for EDs, information requirements, hazard and risk assessment practices, and consequences for risk management.

The World Health Organization definition of endocrine disruptors is broadly accepted in EU legislation. The criteria for the identification of EDs under the Plant Protection Products Regulation and Biocidal Products Regulation, and as substances of very high concern under REACH, are based on this definition. EU Agencies (EFSA, ECHA) and the Commission’s Scientific Committee on Consumer Safety (SCCS) have endorsed the definition and apply it in their assessments. Cross-sectorial (horizontal) criteria for EDs in EU legislation would facilitate consistent identification across sectors.

Based on the limited number of chemicals examined for endocrine disrupting properties (mainly under the BPR or PPPR, and REACH), this Fitness Check did not identify any cases of inconsistent identification.

However, the current data requirements limit the opportunity to identify EDs. The update of data requirements to include further tests relevant for the detection of ED properties has started under the Biocidal and Plant Protection Products Regulations, as well as REACH.

EU agencies have been stepping up their efforts to coordinate assessment work, with a view to ensuring consistency as well as increased efficiency in the assessment process. These efforts will continue, as the European Green Deal commits the Commission to “review how to use better the EU’s agencies and scientific bodies to move towards a process of ‘one substance – one assessment’ and to provide greater transparency when prioritising action to deal with chemicals.” 108  

There are only a few examples of risk assessments based on endocrine disrupting properties. In this limited number of cases, risks were identified following the standard risk assessment approach based on exceedance of a safe threshold (derived no-effect levels) for endpoints which were not necessarily ED-specific. The uncertainties associated with the possible peculiarities of endocrine disrupting effects (i.e. non-threshold, non-monotonic behaviour) are recognised and in some cases have been taken into account through the use of uncertainty factors defined on a case by case basis.

Not all pieces of legislation have information requirements for the toxicological properties of chemicals, and nor do they need to, provided that mechanisms are in place to exploit data generated under other pieces of legislation. Differences in data requirements for the identification of EDs exist across different sectors, reflecting differing intended uses and exposure scenarios contributing to the risk, as well as the need to consider proportionality with respect to socioeconomic and laboratory animal welfare considerations. Where information requirements do exist, they need to be updated to include the latest validated methods relevant to ED identification.

Information requirements also need to address the effects of chemicals on vulnerable groups. In this respect, there is an opportunity to introduce and improve the consistency of definitions of vulnerable groups across legislation, and to clarify the scientific rationale for triggering specific provisions for vulnerable groups (such as the unborn, children and the elderly).

Across the chemicals acquis, the co-legislators have opted for different approaches to risk management (namely, generic risk approaches vs specific risk approaches vs risk/impact-benefit-based) according to specific policy considerations. This situation has been criticised by many stakeholders, who expressed concerns that differences in risk management measures may not be justified. Indeed, the rationale for some of the differences should be made more transparent. A case in point concerns the possibilities for derogation from the exclusion or cut-off criteria for biocidal and plant protection products, respectively. Under BPR, a derogation can be granted based on negligible risk, essentiality or risk-benefit considerations, while under the PPPR, it can only be granted when exposure is negligible or for essential uses.

Despite differences in risk management approaches, this Fitness Check found no cases of inconsistent risk management for specific substances based on the lack of a horizontal approach to ED identification or any other ED-specific considerations. This finding has to be qualified, however, by the limited number of EDs risk managed due to their ED properties as examined in this Fitness Check.

The difficulty in deciding between the generic risk approach and the specific risk approach is partly related to a fundamental scientific uncertainty – whether the effects of an ED are considered to have a threshold or not. In the absence of scientific consensus, legislation can either opt for an approach that does not require an answer to the question (e.g. generic risk approach with derogations as done for plant protection products and biocidal products) or it can determine case by case whether or not a safe (or acceptable) threshold can be quantified and consequently apply an appropriate risk management approach, as done in REACH. In certain pieces of sectorial legislation (e.g. Cosmetic Products Regulation, Food Contact Materials Regulation), existing guidance and examples do not as yet provide a clear indication about how to deal with EDs, for which it is not possible to quantify a safe (or acceptable) threshold. In practice, in cases where a threshold cannot be established, the regulatory approach followed under EU legislation is to minimise exposure as far as possible including the option to prohibit the use of a substance. Application of the generic risk approach is combined with derogations allowing essential uses or risk-benefit options tailored to specific policy contexts.

In view of the Commission’s ambition to develop a process of “one substance-one assessment”, including a horizontal approach to EDs, consolidation and simplification options should be explored, as well as better communication of the approach to citizens and stakeholders. This ambition would also provide an opportunity to systematically assess and manage the risks resulting from the aggregate exposure to the same ED across sectors and from the combined exposures to different EDs, which is a concern expressed by stakeholders. A “one substance-one assessment” approach was seen by many stakeholders as a way of avoiding discrepancies in ED identification while also increasing the effectiveness of regulatory procedures and improving communication with citizens.

The principle of a horizontal approach to ED identification and assessment is broadly supported by stakeholders. However, it needs to be better defined what this means in practice. Elements to consider are: a) the common criteria for ED identification, based on the International Programme on Chemical Safety/World Health Organization definition, which may be implemented via the CLP Regulation (and GHS at the international level); b) a common basic set of information requirements, including screening level information based on new approach methodologies, that provide the basis for applying the criteria; c) a common toolbox to fulfil the information requirements; d) guidance on the application of the toolbox and interpretation of the data generated; and e) a coordinated approach to ED identification and risk assessment, including whether an existing piece of legislation, such as REACH, should serve as the basis for filling data gaps, where necessary.

The International Programme on Chemical Safety/World Health Organization definition of endocrine disruptors is broadly accepted in the implementation of EU legislation. However, in view of the Commission’s ambition to develop a horizontal approach to ED identification, there will be a need to provide practical guidance on how this definition should be interpreted. A particular question will be whether evidence of an adverse effect can only be based on the observation of effects in living animals, or whether adversity can be concluded on the basis of scientific reasoning and extrapolation from the results of new approach methodologies that avoid the need for animal testing. This has implications not only for effectiveness (most available animal tests listed under legislative data requirements have not been specifically designed to detect endocrine effects), but also coherence (animal testing for all endpoints, including endocrine-related ones, is banned for cosmetic products, but not under other legislation) and efficiency (animal testing is costly, time-consuming and ethically questionable).

Effectiveness

Increasing trends in some adverse health and environmental impacts associated with EDs are a matter of societal concern. In this Fitness Check, however, it was not possible to draw firm conclusions on the effectiveness of EU legislation in reducing the potential impact of EDs on these trends.

The assessment of effectiveness is very challenging, mainly because of the attribution challenge. Since there are multiple causes of these effects, it is difficult to determine to what extent exposure to EDs contributes to the observed adverse effects. A further complication is that observable adverse effects on human health and the environment often do not materialise immediately after exposure. In particular, the effects of early life exposure to EDs may not materialise until many years later.

Nevertheless, for some EDs, environmental monitoring data and/or human biomonitoring data have provided evidence that restriction measures (although taken in relation to other hazardous properties) have been successful in reducing exposure levels. There are examples of non-approval of substances with endocrine disrupting properties resulting in significant benefits to wildlife. For instance, international bans on the use of the endocrine disruptor tributyltin as an antifoulant in marine paints has resulted in the recovery of mollusc populations in European coastal waters.

The methodological challenges in assessing effectiveness point to a need to develop and apply a more extensive suite of indicators.

It is too early to assess the effectiveness of the BPR and PPPR, since only a limited number of substances have been identified as EDs or as not being EDs, according to the criteria adopted in 2018. Reflecting the need to gain experience, the Regulations setting out the criteria foresee a review after 7 years, i.e. by 2025.

Under REACH, 17 substances have been identified as Substances of Very High Concern (SVHC) due to endocrine disrupting properties with respect to human health and/or the environment. However, only octylphenol, nonylphenol and their ethoxylates are subject to authorisation specifically due to their endocrine disrupting properties. The inclusion of a substance in the Candidate List for authorisation due to concerns for endocrine disrupting effects triggers additional provisions for risk assessment (Chemical Safety Report) and risk communication (Safety Data Sheet).

CMRs cat. 1A/1B can be identified as SVHCs under REACH and fulfil the cut-off/exclusion criteria for non-approval under the plant protection products and biocidal products regulations. Identification of a CMR as an ED for human health will probably not lead to any additional risk management measures, where they are already the strictest, but may nevertheless contribute to the overall evaluation within a risk-benefit context.

CMR classification is not, however, sufficient to capture all EDs for human health, since other effects are potentially mediated by endocrine disruption, such as certain neurodevelopmental and metabolic disorders. As methods for less studied modalities of endocrine disruption become available, it is expected that more substances will be identified as EDs beyond CMRs.

Data requirements for the identification of EDs are being updated under the PPPR, BPR and REACH to include already existing OECD test guidelines relevant to ED assessment, in particular to include the so-called ‘mechanistic’ tests, which can pinpoint a specific endocrine modality.

Some pieces of sectorial legislation rely primarily on REACH for ED identification. Others such as food contact materials, food additives and cosmetic products have further requirements for products or sectors they regulate but these do not, as yet, specifically address endocrine disruption. Due to the animal testing ban in the cosmetics sector, a specific issue arises concerning the identification of EDs given the need to demonstrate an adverse effect in intact organisms, which currently requires in vivo testing.

There may be a need to strengthen the interconnections between legislation, such as REACH, that have provisions for data generation with sector and product-specific legislation that rely on such data for assessment and management purposes. This may require that ED risk management provisions are introduced into certain pieces of legislation, which currently do not contain any reference to EDs, respecting the special case of cosmetics where in vivo tests should not be generated solely for the purposes of cosmetic ingredient evaluation. Effective sharing of data across legislation should ensure that the information eventually generated is made available for assessing potential EDs falling under sectorial legislation where information is not specifically required.

Irrespective of regulatory provisions for identifying EDs, there is a technical challenge concerning the ability to identify EDs. While progress has been made in developing standardised test methods for identifying EDs and characterising their hazards, the current assessment toolbox is still mainly based on animal tests. The toolbox is incomplete in terms of its coverage of all adverse outcomes and is also limited to a few modes of endocrine action. Another limitation is the ability to detect possible effects of EDs in sensitive life stages. The involvement of the endocrine system in the control of processes such as embryonic development and puberty means that foetuses, infants and adolescents are at higher risk regarding endocrine disruption.

Given the limitations of the current assessment toolbox, new approach methodologies, not involving the use of animals, need to be developed and integrated into the regulatory assessment process. The Commission is addressing this challenge on multiple fronts, from funding research into the development of new methods, to assessing their scientific validity, to promoting their adoption (once validated) as OECD Test Guidelines.

Communication on the presence of EDs in products (articles and mixtures) is required under REACH (when identified as SVHCs) and the BPR. However, there is scope to improve the availability of information on EDs across sectors.

Efficiency

Inefficiencies in the assessment and management of endocrine disruptors can result from situations where a given substance falls within the mandate of multiple agencies and scientific committees, but also from certain regulatory provisions, such as “equivalent level of concern” under REACH. EU agencies have been stepping up their efforts to coordinate assessment work. However, efficiencies could be gained by developing a horizontal approach to EDs, including an increased use of grouping approaches and new approach methodologies that avoid the use of animal tests.

The need to carry out testing and assessment for additional ED-related effects for a larger number of chemicals will inevitably lead to increased workload and costs. Some industry stakeholders are concerned that this will put economic operators in the EU at a competitive disadvantage, while some public authority stakeholders are concerned that this will result in delays to other areas of regulatory assessment work.

The costs and benefits for human health and the environment resulting from regulatory requirements to assess and manage risks of chemicals, including endocrine disruptors, are difficult to quantify and compare. There is a need to further develop and harmonise risk-benefit methodologies, such as socioeconomic analysis. In this Fitness Check, two case studies illustrated the methodological challenges in performing conclusive macroeconomic assessments on the possible impacts of regulatory interventions on trade. This is partly because there is a lack of trade flow data at the level of specific chemicals and the commodities containing these chemicals, as well as data on the market share of a chemical in a given commodity.

EU-added value

Regulating EDs at EU level contributes to ensuring an equal level of protection across the EU Member States regarding human and environmental health. The need for the EU to act in this area is broadly supported by the public and stakeholders, and at the political level by EU institutions such as European Parliament and the EU Committee of the Regions.

There is an opportunity for the Member States to join forces in testing capacities, experience and knowledge sharing. This should lead to greater efficiency and scientific consistency in the assessment of EDs. A uniform EU approach also strengthens the EU’s position in chemicals regulation worldwide, which also has benefits for the EU chemical industry.

Safeguard clauses in EU legislation allow unilateral risk management measures to be taken at Member State level, in cases where an individual Member State has concerns regarding human health or the environment. Such measures have been taken for a few ED substances, such as Bisphenol A and DEHP, and are usually followed by EU-level intervention or withdrawal of unilateral action, to ensure an equal level of protection throughout the EU.

Relevance

Societal concerns about the health and environmental impacts of EDs are high, consistent with concerns about chemicals in general. Perspectives on the importance of regulating EDs are polarised between different stakeholder groups. Citizens engaged in the topic generally lack trust in the ability of EU legislation to achieve its objectives of human health and environmental protection, while economic stakeholders are concerned that decisions may ultimately be based on societal concerns rather than science-based evidence.

A particular challenge over the coming decades is the expected global trend towards increased chemical manufacturing outside the EU, primarily in Asia. The successful implementation of the EU Circular Economy Action Plan will promote innovation and give proper incentives to chemical suppliers, wherever they are located, as regards the safe use and life cycle management of chemicals.

Overall conclusions

The need to address the impacts of EDs on human health and the environment is highly relevant to EU citizens, who also favour regulatory action at the EU level.

Based on the limited number of chemicals examined for ED properties under the BPR or PPPR, or as SVHCs with ED properties under REACH, this Fitness Check did not identify any cases of inconsistent identification across sectors. However, the need for a horizontal approach for identification of EDs has been recognised that is based on the definition of the World Health Organization and builds on the criteria developed for plant protection products and biocidal products while ensuring that it is fit for purpose for other relevant legislation, particularly for REACH but also legislation on cosmetic products, toys and food contact materials. This is also an opportunity to simplify and rationalise current parallel identification procedures.

A horizontal approach for the identification of EDs would also feed into the development of a ‘one substance-one assessment’ approach across legislation and facilitate the management of risks to humans and wildlife resulting from combined exposures to different endocrine disruptors regulated under different pieces of legislation. Current data requirements limit the opportunity to identify EDs. The strengthening of information requirements in the relevant legislation is key to improving the identification of EDs.

This Fitness Check could not conclude on the effectiveness of the chemicals acquis in protecting health and the environment from EDs. However, there is a need to explore options to strengthen the legislative framework to further minimise exposure to EDs, both for consumers and for the environment

Overall, there are opportunities for simplification, consolidation, burden reduction and better communication of the principles guiding risk management of EDs. The findings of this Fitness Check will provide an input to the Chemicals Strategy for Sustainability being developed in the context of the European Green Deal.

10REFERENCES

Baken, K.A., Lambrechts, N., Remy, S., Mustieles, V., Rodríguez-Carrillo, A., Neophytou, C.M., Olea, N., and Schoeters, G. (2019). A strategy to validate a selection of human effect biomarkers using adverse outcome pathways: Proof of concept for phthalates and reproductive effects. Environ Res 175, 235-256.

Bellanger, M., Demeneix, B., Grandjean, P., Zoeller, R.T., and Trasande, L. (2015). Neurobehavioral deficits, diseases, and associated costs of exposure to endocrine-disrupting chemicals in the European Union. J Clin Endocrinol Metab 100, 1256-1266.

Bolt, H.M. (2017). The current debate on cost burden by human exposure to endocrine disrupting chemicals. Arch Toxicol 91, 2965-2966.

Bond, G.G., and Dietrich, D.R. (2017a). Further thoughts on limitations, uncertainties and competing interpretations regarding chemical exposures and diabetes. J Epidemiol Community Health 71, 943.

Bond, G.G., and Dietrich, D.R. (2017b). Human cost burden of exposure to endocrine disrupting chemicals. A critical review. Arch Toxicol 91, 2745-2762.

Bopp, S.K., Kienzler, A., Richarz, A.N., van der Linden, S.C., Paini, A., Parissis, N., and Worth, A.P. (2019). Regulatory assessment and risk management of chemical mixtures: challenges and ways forward. Crit Rev Toxicol 49, 174-189.

Brack, W., Dulio, V., Ågerstrand, M., Allan, I., Altenburger, R., Brinkmann, M., Bunke, D., Burgess, R.M., Cousins, I., Escher, B.I., et al. (2017). Towards the review of the European Union Water Framework Directive: Recommendations for more efficient assessment and management of chemical contamination in European surface water resources. Sci Total Environ 576, 720-737.

Browne, P., Van Der Wal, L., and Gourmelon, A. (2020). OECD approaches and considerations for regulatory evaluation of endocrine disruptors. Mol Cell Endocrinol 504, 110675.

Brunel University London, and DTU Food National Food Institute Denmark (2016). Mapping commonalities and differences in approaches for testing and assessment of endocrine disruptors within the EU and among relevant international trading partners ( https://op.europa.eu/en/publication-detail/-/publication/0abd23e0-ec38-11e6-ad7c-01aa75ed71a1 ).

Canzian, G., Giua, L., and Verzillo, S. (2020). Impact of interventions regulating Endocrine Disruptors on trade dynamics for EU Member States. JRC120335. https://publications.jrc.ec.europa.eu/repository/handle/JRC120335

Chemsec (2019). Lost at SEA-The information policymakers actually need from applicants and SEAC opinions ( https://chemsec.org/publication/authorisation-process,reach/lost-at-sea/ ).

Demeneix, B., and Slama, R. (2019). Endocrine Disruptors: from Scientific Evidence to Human HealthProtection (doi: 10.2861/802173 | QA-01-19-272-EN-N ).

ECHA (2011). Guidance on the preparation of socio-economic analysis as part of an application for authorisation ( https://echa.europa.eu/documents/10162/23036412/sea_authorisation_en.pdf/aadf96ec-fbfa-4bc7-9740-a3f6ceb68e6e ).

ECHA RAC 2017. Risk-related considerations in applications for authorisation for endocrine disrupting substances for the environment, specifically OPnEO and NPnEO. https://echa.europa.eu/documents/10162/13637/npneo_and_opneo_for_agreement_final_en.pdf/

ECHA (2020). The use of bisphenol A and its alternatives in thermal paper in the EUduring 2014 -2022 ( https://echa.europa.eu/documents/10162/23294236/bpa_thermal_paper_report_2020_en.pdf/59eca269-c788-7942-5c17-3bd822d9cba0 ).

ECHA, EFSA, and JRC (2018). Guidance for the identification of endocrine disruptors in the context of Regulations (EU) No 528/2012 and (EC) No 1107/2009. EFSA Journal, 16, 5311.

Ecorys (2018). Study supporting the REFIT Evaluation of the EU legislation on plant protection products and pesticides residues (Regulation (EC) No 1107/2009 and Regulation (EC) No 396/2005) (doi: 10.2875/88433).

ECPI (2014). Denmark scraps plan to ban four phthalate plasticizers. Additives for Polymers 9, 11.

EFSA (2020a). Cumulative dietary risk characterisation of pesticides that have acute effects on the nervous system. EFSA Journal, 18, 6087.

EFSA (2020b). Cumulative dietary risk characterisation of pesticides that have chronic effects on the thyroid. EFSA Journal, 18, 6088.

EFSA Panel on Food Additives and Nutrient Sources added to Food (2012). Guidance for submission for food additive evaluations. EFSA Journal, 10, 2760.

EFSA Panel on Food Contact Materials, Enzymes, and Flavourings and Processing Aids (2008). Note for Guidance for the Preparation of an Application for the Safety Assessment of a Substance to be used in Plastic Food Contact Materials. EFSA Journal, 6, 21r.

EFSA Panel on Food Contact Materials, Enzymes, and Flavourings and Processing Aids (2015). Scientific Opinion on the risks to public health related to the presence of bisphenol A (BPA) in foodstuffs. EFSA Journal, 13, 3978.

EFSA Panel on Food Contact Materials, Enzymes, and Flavourings and Processing Aids (2016). Recent developments in the risk assessment of chemicals in food and their potential impact on the safety assessment of substances used in food contact materials. EFSA Journal, 14, 4357.

EFSA Scientific Committee (2013). Scientific Opinion on the hazard assessment of endocrine disruptors: Scientific criteria for identification of endocrine disruptors and appropriateness of existing test methods for assessing effects mediated by these substances on human health and the environment. EFSA Journal, 11, 3132.

EMA (2000). GL6 Environmental impact assessment (EIAS) for veterinary medicinal products - Phase I ( https://www.ema.europa.eu/en/vich-gl6-environmental-impact-assessment-eias-veterinary-medicinal-products-phase-i#current-effective-version-section ).

EMA (2005). GL38 Environmental impact assessments for veterinary medicinal products - Phase II (GL38 Environmental impact assessments for veterinary medicinal products - Phase II (CVMP/VICH/790/2003)).

European Commission (2014). Technical Report on Aquatic Effect-Based Monitoring Tools (doi: 10.2779/7260).

European Commission (2015). Report on Public consultation on defining criteria for identifying endocrine disruptors in the context of the implementation of the Plant Protection Product Regulation and Biocidal Products Regulation ( https://ec.europa.eu/health/sites/health/files/endocrine_disruptors/docs/2015_public_consultation_report_en.pdf ).

Fraunhofer-Institut (1999). Revised Proposal for a List of Priority Substances in the Context of the Water Framework Directive ( https://ec.europa.eu/environment/water/water-dangersub/pdf/commps_report.pdf ).

Ganzleben, C., Antignac, J.P., Barouki, R., Castaño, A., Fiddicke, U., Klánová, J., Lebret, E., Olea, N., Sarigiannis, D., Schoeters, G.R., et al. (2017). Human biomonitoring as a tool to support chemicals regulation in the European Union. Int J Hyg Environ Health 220, 94-97.

Gennings, C., Shu, H., Rudén, C., Öberg, M., Lindh, C., Kiviranta, H., and Bornehag, C.G. (2018). Incorporating regulatory guideline values in analysis of epidemiology data. Environ Int 120, 535-543.

Gore, A.C., Chappell, V.A., Fenton, S.E., Flaws, J.A., Nadal, A., Prins, G.S., Toppari, J., and Zoeller, R.T. (2015). EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocr Rev 36, E1-e150.

Hamers, T., Kortenkamp, A., Scholze, M., Molenaar, D., Cenijn, P.H., and Weiss, J.M. (2020). Transthyretin-Binding Activity of Complex Mixtures Representing the Composition of Thyroid-Hormone Disrupting Contaminants in House Dust and Human Serum. Environ Health Perspect 128, 17015.

Hamers, T., Legradi, J., Zwart, N., Smedes, F., de Weert, J., van den Brandhof, E.J., van de Meent, D., and de Zwart, D. (2018). Time-Integrative Passive sampling combined with TOxicity Profiling (TIPTOP): an effect-based strategy for cost-effective chemical water quality assessment. Environ Toxicol Pharmacol 64, 48-59.

Hauser, R., Skakkebaek, N.E., Hass, U., Toppari, J., Juul, A., Andersson, A.M., Kortenkamp, A., Heindel, J.J., and Trasande, L. (2015). Male reproductive disorders, diseases, and costs of exposure to endocrine-disrupting chemicals in the European Union. J Clin Endocrinol Metab 100, 1267-1277.

IPCS (2002). Global Assessment of the State-of_the-Science of Endocrine Disruptors, T. Damstra, S. Barlow, A. Bergman, R. Kavlock, and G. Van Der Kraak, eds. ( https://www.who.int/ipcs/publications/new_issues/endocrine_disruptors/en/ ).

Lee, J.E., Jung, H.W., Lee, Y.J., and Lee, Y.A. (2019). Early-life exposure to endocrine-disrupting chemicals and pubertal development in girls. Ann Pediatr Endocrinol Metab 24, 78-91.

Legler, J., Fletcher, T., Govarts, E., Porta, M., Blumberg, B., Heindel, J.J., and Trasande, L. (2015). Obesity, diabetes, and associated costs of exposure to endocrine-disrupting chemicals in the European Union. J Clin Endocrinol Metab 100, 1278-1288.

Lucaccioni, L., Trevisani, V., Marrozzini, L., Bertoncelli, N., Predieri, B., Lugli, L., Berardi, A., and Iughetti, L. (2020). Endocrine-Disrupting Chemicals and Their Effects during Female Puberty: A Review of Current Evidence. Int J Mol Sci 21.

Middelbeek, R.J., and Veuger, S.A. (2015). Letter to the Editor: Re: Neurobehavioral Deficits, Diseases, and Associated Costs of Exposure to Endocrine-Disrupting Chemicals in the European Union. J Clin Endocrinol Metab 100, L52-53.

Milieu Ltd, Ökopol, Risk & Policy Analysts, and RIVM (2017). Study for the strategy for a non-toxic environment of the 7th Environment Action Programme ( https://ec.europa.eu/environment/chemicals/non-toxic/pdf/NTE%20main%20report%20final.pdf ).

Ministère de la Transition Ecilogique et Solidaire, M.d.S.e.d.l.S. (2019). Second National strategy on endocrine disruptors ( https://www.ecologique-solidaire.gouv.fr/sites/default/files/SNPE%202%20english%20-%20Action%20plan.pdf ).

Munn, S., and Goumenou, M. (2013a). Key scientific issues relevant to the identification and characterisation of endocrine disrupting substances.

Munn, S., and Goumenou, M. (2013b). Thresholds for Endocrine Disrupters and Related Uncertainties.

OECD (2017). New Scoping Document on in vitro and ex vivo Assays for the Identification of Modulators of Thyroid Hormone Signalling ( https://doi.org/10.1787/9789264274716-en ).

OECD (2018). Revised Guidance Document 150 on Standardised Test Guidelines for Evaluating Chemicals for Endocrine Disruption ( https://doi.org/10.1787/9789264304741-en ).

OECD (2019). Considerations when Assessing Children's Exposure to Chemicals from Products ( http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ENV/JM/MONO(2019)29&docLanguage=en ).

Olsson, I.-M., and al., e. (2014). The cost of inaction - A Socioeconomic analysis of costs linked to effects of endocrine disrupting substances on male reproductive health ( http://dx.doi.org/10.6027/TN2014-557 ).

Petrakis, D., Vassilopoulou, L., Mamoulakis, C., Psycharakis, C., Anifantaki, A., Sifakis, S., Docea, A.O., Tsiaoussis, J., Makrigiannakis, A., and Tsatsakis, A.M. (2017). Endocrine Disruptors Leading to Obesity and Related Diseases. Int J Environ Res Public Health 14.

Posthuma, L., Dyer, S.D., de Zwart, D., Kapo, K., Holmes, C.M., and Burton, G.A., Jr. (2016). Eco-epidemiology of aquatic ecosystems: Separating chemicals from multiple stressors. Sci Total Environ 573, 1303-1319.

Risk & Policy Analysts Limited (2015). Analysis of capacities and capabilities of laboratories toconduct OECD TG 443 extended one-generationreproductive toxicity study (OECD TG 443) ( https://echa.europa.eu/documents/10162/13630/echa_sr26_eogrts_en.pdf ).

Rogiers et al (2020). The way forward for assessing the human health safety of cosmetics in the EU - Workshop proceedings, Toxicology 436. https://doi.org/10.1016/j.tox.2020.152421

SCCS (2018). SCCS Notes of Guidance on for the testing of cosmetic ingredients and their safety evaluation (10th revision) ( https://op.europa.eu/en/publication-detail/-/publication/1d5f1653-38ce-11e9-8d04-01aa75ed71a1/language-en/format-PDF ).

Schug, T.T., Janesick, A., Blumberg, B., and Heindel, J.J. (2011). Endocrine disrupting chemicals and disease susceptibility. J Steroid Biochem Mol Biol 127, 204-215.

Slama, R., Vernet, C., Nassan, F.L., Hauser, R., and Philippat, C. (2017). Characterizing the effect of endocrine disruptors on human health: The role of epidemiological cohorts. C R Biol 340, 421-431.

Solecki, R., Kortenkamp, A., Bergman, Å., Chahoud, I., Degen, G.H., Dietrich, D., Greim, H., Håkansson, H., Hass, U., Husoy, T., et al. (2017). Scientific principles for the identification of endocrine-disrupting chemicals: a consensus statement. Arch Toxicol 91, 1001-1006.

Tang, J., and Zhai, J.X. (2017). Distribution of polybrominated diphenyl ethers in breast milk, cord blood and placentas: a systematic review. Environ Sci Pollut Res Int 24, 21548-21573.

Trasande, L., Zoeller, R.T., Hass, U., Kortenkamp, A., Grandjean, P., Myers, J.P., DiGangi, J., Bellanger, M., Hauser, R., Legler, J., et al. (2015). Estimating burden and disease costs of exposure to endocrine-disrupting chemicals in the European union. J Clin Endocrinol Metab 100, 1245-1255.

UNEP (2019). Global Chemicals Outlook IIFrom Legacies to Innovative Solutions: Implementing the 2030 Agenda for Sustainable Development – Synthesis Report ( https://wedocs.unep.org/bitstream/handle/20.500.11822/27651/GCOII_synth.pdf?sequence=1&isAllowed=y ).

UNEP IPCP (2017). Worldwide initiatives to identify endocrine disrupting chemicals (EDCs) and potential EDCs ( http://wedocs.unep.org/bitstream/handle/20.500.11822/25633/EDC_report1.pdf?sequence=1&isAllowed=y ).

WHO/UNEP (2012). State of the science of endocrine disrupting chemicals - 2012 ( https://www.who.int/ceh/publications/endocrine/en/ ).

11GLOSSARY

Annex 1. Procedural information

1Lead DG and internal references

The "Fitness Check on the relevant EU legislation on Endocrine Disruptors" was led by DG Joint Research Centre (JRC). The mandate was established in the Communication “Towards a comprehensive European Union framework on endocrine disruptors” 109 .

Issues relevant to endocrine disruptors (EDs) were included in recent Fitness Checks and evaluations of chemical legislation, including the  REACH REFIT evaluation , the  REACH Review on the authorisation route of substances with endocrine disrupting properties according to REACH Art. 138(7) , the  Fitness Check of the chemicals legislation other than REACH, the review of the legislation on cosmetics with regard to endocrine disrupting substances , and the  evaluation of the 7th Environment Action Programme . Among legislation regulating downstream uses of chemicals in products, the evaluation of the legislation on food contact materials and the evaluation of the legislation on toy safety  were also considered relevant to this FC. Among pieces of environmental legislation, the  Fitness Check of the water legislation  was considered for potentially ED-relevant aspects.

Also relevant was Annex 9 of the impact assessment on the definition of criteria for identifying EDs in the context of the Plant Protection Products Regulation and Biocidal Products Regulation . This provides a multi criteria analysis of impacts (on health, environment etc.) based on a scientific review of the evidence linking endocrine disrupting chemicals to hormone-related diseases and the estimated related costs to society attributed to exposure to EDs.

The findings of this Fitness Check formed an important contribution to the Chemicals Strategy for Sustainability under the European Green Deal 110 . This strategy was part of the Commission’s Work Programme for 2020 111 .

2Organisation and timing

The pre-existing Inter-service Steering Group (ISG) on endocrine disruptors was used to steer the process and provide input. The ISG comprised representatives from the Directorate Generals for Agriculture and Rural Development (AGRI), Environment (ENV), Internal Market, Industry, Entrepreneurship and SMEs (GROW); Health and Food Safety (SANTE), Employment, Social Affairs and Inclusion (EMPL), Maritime Affairs and Fisheries (MARE), Research and Innovation (RTD), Justice and Consumers (JUST), Trade (TRADE), Legal Service (SJ), Joint Research Centre (JRC) and the Secretariat General (SG).

A meeting with DGs directly responsible for the implementation of the Commission communication (DGs ENV, GROW, SANTE, EMPL, JRC and SG) was held to kick-off the Fitness Check. The ISG met twice during the evaluation process and provided feedback through two written consultations (Table 12.1).

Table 12.1 Meetings of the Endocrine Disruptor Inter-service Steering Group and informal task force

3Exceptions to the better regulation guidelines

No exceptions were made to the Better Regulation Guidelines 112 during this Fitness Check.

4Consultation of the Regulatory Scrutiny Board

An upstream consultation of the Regulatory Scrutiny Board (RSB) of the European Commission on the planned methodology for the Fitness Check was held on 3 October 2019. A draft version of the evaluation report (Staff Working Document) was submitted to the RSB on 26 May 2020. The RSB issued its positive opinion on 24 June 2020. The Board made a number of recommendations to further improve the report. These were addressed in the revised report as follows:

5Evidence, sources and quality

The sources of evidence and their quality are described in Annex 3. These included the findings of previous fitness checks and evaluations, a series of case studies, and the results of public, stakeholder and SME consultations. No external expertise was used, apart from ad hoc consultations with EU agency staff related to EU chemical safety assessments. All of the work was conducted by Commission staff, with no external studies being performed.

Annex 2. Synopsis report: stakeholder consultation activities

1Summary

An important element of the Fitness Check is a consultation strategy aimed at citizens, stakeholders and companies.

In the public consultation, 474 responses were received over a twelve-week period from 16/12/2019 to 09/03/2020. The survey, which was translated into all official EU languages, was designed to gain the perspective of the citizen, rather than stakeholders or companies, who were targeted via separate surveys. In total, 431 of the 474 responses (91%) came from citizens. A majority of citizens felt informed about endocrine disruptors and considered that their effects on health and wildlife, while poorly understood, do represent a concern. Similarly, a majority felt informed about the regulation of endocrine disruptors, but do not feel that the current regulatory framework is sufficiently protective. A majority of respondents also expressed the view that the EU should have the same approach or more consistent approach for both identifying and managing the risks of exposure to endocrine disruptors. Opinions were divided on whether animal testing is minimised in the assessment process. Similarly, there were mixed views on whether the cost of regulatory action is proportionate to the benefits accrued for human health, wildlife and the economy. In general, the respondents considered that EU laws on endocrine disruptors are relevant to societal concerns, and that regulatory action should be taken at the EU level.

In the stakeholder consultation, 183 responses were received over an eight-week period from 06/12/19 to 31/01/20. The respondents were businesses, public authorities, academics, research organisations, and civil society organisations. A large majority (93%) of the respondents considered that the absence of harmonised criteria for identifying endocrine disruptors across sectors is problematic. A smaller majority (53%) considered that this harmonisation should be achieved by introducing an endocrine disruptor category under the classification and labelling legislation. Around three quarters said they were aware of inconsistencies, gaps or overlaps in the regulatory framework, and a similar proportion were aware of inadequacies in the availability of regulatory test methods. A large majority (88%) reported an increase in costs to comply with EU laws on endocrine disruptors.

The results of the stakeholder consultation also give a detailed breakdown of stakeholder views on the effectiveness of legislation in specific sectors (REACH, cosmetics, biocides, plant protection products). This includes the perceived effectiveness in protecting people and wildlife, improving the functioning of the Single Market, enhancing competitiveness and innovation, and promoting alternatives to animal testing. Opinions were typically divided on these issues. However, in relation to exposure to chemical mixtures, there was a tendency to consider the regulatory framework as insufficiently protective.

The answers to additional questions confirmed that a majority of stakeholders consider the objectives of the regulatory framework to be relevant in addressing health and environmental concerns. Moreover, the prevailing view was that regulatory actions should be taken at the EU rather than Member State level.

An additional survey, targeting micro, small and medium-sized enterprises (SMEs), was conducted through the Enterprise Europe Network and was open from 01/02/2019 to 09/03/2020. Responses were received from 70 SMEs. In general, opinion was divided, or not expressed, across a range of questions. However, many respondents considered the regulatory process to identify and control chemicals with endocrine disrupting properties to be effective in protecting people and wildlife, in improving the functioning of the internal market, and enhancing competitiveness and innovation. A majority considered that the lack of a hazard category for the classification and labelling of endocrine disruptors poses a problem for their consistent identification and risk management. On the question of EU-added value, most respondents reported that unilateral Member State actions had not affected their company.

2Introduction and approach

Stakeholder consultation was a key component of this Fitness Check to collect views and factual information in response to the evaluation questions and to ensure a balanced and comprehensive assessment of the legislative framework.

The objectives of the consultation activities were to:

·Assess public concerns and needs with respect to endocrine disruptors in the EU.

·Evaluate to what extent current EU legislation meets the concerns and needs of citizens.

·Collect views on possible lack of legislative coherence of EU legislation with respect to EDs and possible impacts on stakeholders.

·Identify any inconsistencies in the legal framework for endocrine disruptors and their consequences for small companies

·Collect information on the effectiveness of the current EU legislation for the identification and risk management of endocrine disruptors.

·Collect information on the efficiency of the current procedures for the identification and risk management of endocrine disruptors.

·Identify opportunities for improving the way endocrine disruptors are assessed, managed and potential risks communicated

The consultation strategy developed for the purpose of this Fitness Check 113 comprised:

·presentation of the methodological approach at the First Annual Forum on Endocrine Disruptors on 8 November 2019;