Environmental Assessment of Veterinary Medicinal Products in Denmark
5. Legislation concerning environmental risk assessment of veterinary medicinal products
5.1 Introduction
This chapter includes a review of the current risk assessment of veterinary medicinal products and outlines the similarities and differences to the current risk assessment principle of new and existing chemicals in the EEC. Initially it should be stressed that this chapter only describes the risk assessment procedure for new veterinary medicinal products introduced to the EU market after January 1998. No legislation or risk assessment procedure for existing veterinary medicinal products exist. To the authors knowledge there is no initiatives for doing so in the future
Whereas legislation and regulation of domestic and industrial chemicals with respect to the protection of the environment have been implemented in Europe and North America for decades, such legislation of medicinal products was first initiated in the late 80´s in the USA (FDA 1985; 1987 and 1995) and in the beginning of the 90´s in Europe.
The European Commission first integrated a basic evaluation of the environmental impact of veterinary medicinal products in its Directive 81/852/EEC as amended by Directive 92/18/EEC. This Directive outlines basic requirements for conducting an environmental risk assessment of veterinary medicinal products. The 81/852/EEC and 92/18/EEC Directives was followed by a guidance document by the European Agency for the Evaluation of Medicinal Products (EMEA) - EMEA/CVMP/055/96, which provides a technical guidance for the assessment of environmental risk of new veterinary medicinal products. In addition, the European Commission has also laid down guidelines for an environmental risk assessment of animal food additives (94/40/EEC). These guidelines include a description of the basic requirements for assessing the fate and effects of the food additives in the environment, e.g. environmental fate and persistence; toxicity to algae, daphnia, fish in the aquatic environment and microbial processes and fauna in the terrestrial environment. At present, the Technical Directive for human medicinal products (75/318/EEC) does not include any reference to an assessment of their potential environmental risk or to ecotoxicology in general. However, a detailed technical draft guideline issued in 1994 (III/5504/94) indicates that the approach applicable for veterinary medicine also would apply for human medicinal products.
In the following sections a short review of the principles for assessment of environmental risk of new and existing chemicals in the EU is given (section 5.2) with the aim to compare these risk assessment methodologies with the risk assessment for veterinary medicinal products (section 5.3).
5.2 ERA for new and existing substances in the EEC region
The Council Directive 67/548/EEC (as amended by Directive 92/32/EEC) on the classification, packaging and labelling of dangerous substances requires the manufactures of new substances, if these are sold in quantities exceeding a certain limit (1 t per year per manufacturer or imposter , to notify relevant national authorities in the country of manufacture or import. A notification includes a package of test data concerning various information of the physical-chemical, toxicological and ecotoxicological properties of the notified substance. With the base set of data, the competent authorities are then required to carry out an assessment of the risk of the substances to man and the environment in accordance with the principles of Commission Directive 93/67/EEC. For existing substances the Council Regulation (EEC) no. 793/93 requires an assessment of risk to man and environment of priority substances given in the Commission Regulation no. 1488/94 on risk assessment on existing substances. For existing chemicals an informal priority setting (EURAM-IPS) method is used for selecting chemicals among the 100,000 listed in the EINECS database (The European Inventory of Existing Commercial Chemical Substances). The purpose of IPS is to select chemicals for a detailed risk assessment among the EEC high production volume compounds, i.e. >1000 t/y/manufacture (HVPC, a subset of the EINECS database containing approx. 2700 chemicals). Data necessary for the IPS and an initial hazard assessment is called the HEDSET (European Commission 1993) and covers issues as environmental exposure, environmental effects, exposure of man and human health effects. The data are contained in the IUCLID database. On the basis of national priority lists of chemicals and priority indicated by the employed EURAM a comprehensive risk assessment documents is produced by competent authorities in the member states in collaboration with industry.
In support of the legislation a detailed Technical Guidance Document (TGD) and a software program (EUSES) have been produced, including extensive technical details for conducting hazard identification, dose-effect assessment, exposure assessment and risk characterisation in relation to man and environment. The risk assessment of new notified substances is based on the data submitted by the notifiers in accordance with Directive 67/548/EEC. The directive provides a scheme of step-wise, tonnage-related data requirements (Figure 5.1). The tests have to be carried out in accordance of the EU methods (Annex V to Directive 67/548/EEC). If no EU method is available or scientifically pertinent other international test guidelines, preferably OECD guidelines may be used. As a minimum requirement, the full base-set (FBS) testing required for new substances (Annex V, Directive 67/548/EEC) must be available for a risk assessment. Among the existing high production volume chemicals (>1000 tons/year/manufactor), the minimum data set for the high priority substances is the HEDSET, which encompasses the full base-set for new substances and a screening for reproductive toxicity (OECD 421). This ensures that for both new notified and for chemicals of priority data from studies on short term toxicity on algae, daphnia and fish and key data on the environmental fate, e.g. log Kow, biodegradability, vapour pressure and water solubility, are available for the risk assessment. For new substances further data are foreseen at level 1 and 2 of the risk assessment procedure. For existing substances information beyond the base-set is some time very limited other times very large and comprehensive. On the basis of the toxicity data available, an assessment factor ranging from 10-1000 is applied to the EC/LC/NOEC values to derive the Predicted No Effect Concentration (PNEC). The choice of assessment factor depends on the quantity and quality of toxicity data (Table 5.1).
Figure 5.1.
The general principles of risk assessment of new and existing substances in the EU.
Figur 5.1.
De generelle principper for riskovurderingen af nye og eksisterende kemiske stoffer i EU.
Table 5.1.
Assessment factors to derive PNEC according to the Technical Guidance Document for new and
existing substances. See TGD for detailed notification.
Tabel 5.1.
Applikationsfaktorer for estimering af PNEC ifølge den tekniske vejledning for vurdering
af nye og eksisterende stoffer.
|
Assessment factor |
At least one short-term L(E)C50 from each of three throphic levels of the base set (fish, Daphnia and algae) |
1000 |
One long-term NOEC (either fish or Daphnia) |
100 |
Two long-term NOECs from species representing two throphic levels (fish and/or Daphnia and/or a algae |
50 |
Long-term NOECs from at least three species (normally fish, Daphnia and algae) representing three throphic levels |
10 |
Field data or model ecosystems |
case by case |
The exposure assessment should be based on representative monitoring data and/or model
calculations. However, for many chemicals it is very difficult to obtain representative
monitoring data, as detailed monitoring studies only exist for a few organic substances.
To calculate the exposure by e.g. models may be very complicated as detailed information
about distribution, specific use and application may not be available. The primary
manufactures of a chemical may not hold detailed information about the use of the chemical
in products, i.e. "downstreeam uses". Box 5.1 includes a brief discussion
about uncertainties and limitations in the prediction of the environmental concentrations
(PECs) .
An assessment of the risk of secondary poisoning to top predators has to be included in the risk assessment report if indicated by the available data, e.g. if the substance is bioaccumulative in fish and excert chronic or reproductive effects in mammals. At a base-set level the available physico-chemical and (eco)toxicological data can be used to decide whether or not there is any indications for a potential for bioaccumulation and/or indirect effects. The Technical Guidance Document provides detailed information on how to assess the risk of secondary poisoning
The risk assessment should be carried out for all relevant compartments (concern areas), i.e. for micro-organisms of sewage treatment plants, aquatic, sediment and soil dwelling species as well as for secondary poisoning of top predators of the food chains. A quantitative ERA is currently not possible for the air compartment. As the base set of effect data only include aquatic organisms and micro-organisms from sludge, and no base-set of tests currently exist for sediment and soil organisms, short-term toxicity tests on sediment and soil dwelling species are only required if a potential risk for the sediment or soil compartment have been identified on basis of a risk characterisation using the equilibrium partitioning method on the aquatic base-set, as outlined in the TGD. The OECD-tests on earthworm (207) and plants (208) are currently the only available internationally standardised tests for the soil environment. Standard tests on sediment species are currently being developed. Whereas, a quantitative evaluation of risk in the aquatic (especially the freshwater system) and the terrestrial environment should be possible for most substances, the risk assessment for air can only be carried out qualitatively, as no adequate biotic testing systems currently are available.
Box 5.1.
Uncertainties and limitations in risk assessment
All kinds of risk assessment procedures are bound to have some degree of uncertainties and limitations. Many factors controlling exposure and toxicity has to be considered when performing a risk assessment, and as many of these vary from situation to situation, some margin of uncertainty has to be taken into account to reach the desired level of protection. This Box attempt to describe a few of the most important factors that have to be taken into account when performing a risk assessment. PEC estimation: Most of the variance between the ERA of drugs and chemicals as described above is due to the fact that veterinary substances to some extend is evaluated the same way as pesticides. This may be justified by the fact that exposure scenarios for veterinary medicines in the same manner as for pesticides often are far more reliable than for chemicals with a diverse use and environmental release. For new and existing chemicals PEC should be calculated on both a local and a regional spatial scale preferable from monitoring data If this information is not available estimates may be made from exposure models such as e.g. EUSES. One of the major challenges of estimating exposure of industrial chemicals is, however, to establish representative release estimates according to realistic worst case scenarios. The environmental release of an industrial compound may occur during the many steps from the initial producers via several consumers to its final dispose or incineration. Although the uses and recommended doses of veterinary medicines are relatively well known compared to industrial chemicals, less is known about metabolisms and excretion of the substances. Another major uncertainty in estimating the environmental concentration of veterinary drugs is a possible persistency in the excreta before and after the manure or slurry is applied as fertilisers to arable land. A uniform procedure to estimate PEC for veterinary medicinal products is suggested by Spaepen et al. (1997) PNEC estimations: Many uncertainties can be found in the estimations of predicted no environmental effect concentration (PNEC). The procedure of prediction a safe level for a complete ecosystem on the basis of laboratory short term tests on a few species is bound to include a high level of uncertainty. Of course this is the major reason for applying safety factors on the no effect concentration for single species. These assessment factors cover factors like inter- and intra species differences, acute to chronic exposure and extrapolation from laboratory to field, e.g. how to extrapolate effects observed in controlled single species tests to ecologically relevant effects in a unrestrained multispecies natural ecosystem. In addition to these a number of other uncertainties in PNEC calculations can be listed:
These parameters are not equally important for all chemicals or in all ecosystems, and do not necessarily draw in the same direction. However, they all represent a level of uncertainty in the present calculation procedure for PNEC and plead for the present use of assessment factors. |
The requirements for the dossier to be submitted for the authorisation of a plant
protection product is more extensive than the ones for new and existing substances (Annex
VI, 91/414/EEC). In addition to the acute toxicity test with algae, daphnia and fish,
chronic toxicity, bioaccumulation and sublethal studies on reproduction and growth is
required for fish and daphnia. The applicant must furthermore provide data from toxicity
studies on birds and other non- target organisms, e.g. honey bees and earthworms.
5.3 Comparison of ERA for veterinary medicinal products and food additives with ERA for new and existing substances in the EEC region
The European Commission has included the evaluation of the environmental impact of veterinary medicinal products in its Directive 81/852/EEC as amended by Directive 92/18/EEC. This Directive outlines in the basic requirements for conducting an environmental risk assessment of veterinary medicinal products. For details on the evaluation and a technical guidance, the document by the European Agency for the Evaluation of Medicinal Products (EMEA) - EMEA/CVMP/055/96, may be consulted.
The risk assessment as described in the EMEA guidance document is based on a tiered approach. Hence more detailed information is required as the risk assessment proceeds. The first level (phase 1) includes an identification of exposure routes and a first estimation of PECs. Phase 1 includes a number of cut-off values for further assessment. In the second step (phase 2 - Tier A) ecotoxicological effects are compared to the PECs, and a first risk assessment is made for the aquatic and the terrestrial environments. If further data are needed, Phase 2 - Tier B details additional toxicity tests and fate studies. The "Decision Trees" used for risk assessment in the EMEA-guideline is presented in Figure 5.2-5.4. In the following paragraphs this risk assessment will be compared to the one used for new and existing substances in the European Union. To facilitate the reading the term chemicals or industrial chemicals will be used as a collecting term for new and existing substances in this chapter, unless important differences in legislation exist for one of the two groups. The term drugs will regularly be used instead veterinary medicinal products.
Directive no. 94/40/EEC from the European Commission lays down a guideline for assessing the effects of food additives. According to the Directive applicants should review the physical-chemical, toxicological and ecotoxicological properties according to the Commission Directive 67/548 (latest amendment 93/105) or other international recognised methods. If other methods are used, such use must be justified. The guideline for assessing food additives, however, is currently being revised by the Commission. In addition the existing guideline is adopted from the risk assessment procedure for new chemicals, and no further discussion concerning food additives will therefore be presented in this report. Attention should however be drawn to the fact that many food additives, e.g. growth promoters, frequently are used as medicinal products as well.
Figure 5.2.
Phase I in the EMEA procedure for assessing risk of veterinary medicinal
products to the environment.
Figur 5.2.
Fase I i EMEA-proceduren for miljørisikovurdering af veterinære medicinale produkter.
Figure 5.3.
Phase II in the EMEA procedure for assessing risk of non fish medicines to the
environment.
Figur 5.3.
Fase II i EMEA-proceduren for miljørisikovurdering af de veterinære medicinale
produkter, som ikke er fiskemedicin.
Figure 5.4.
Phase II in the EMEA procedure for assessing risk of fish medicines to the environment.
Figur 5.4.
Fase II i EMEA-proceduren formiljørisikovurdering af medicinske produkter for fisk.
5.4 Phase 1
The first step in Phase I of the EMEA guideline, is to identify the uses and the active ingredients, which is not assumed to lead to any significant environmental exposure or pose significant risk, e.g. vitamins, electrolyte or natural amino acids. The identification of substances which are presumably non-toxic or very low-toxic, aims at avoiding unnecessary and unjustified testing and assessment efforts. Although e.g. substances used for treatment of companion animals (pets) and individual treatment of domestic animals may be produced and used in relatively large quantities, the treatment is not expected to lead to any significant exposure of the environment. Similar to the use of human medicine, medicinal products for companion animals may, however, if washed out to the drains by rainwater, be found in municipal waste water treatment plants.
This differs from the evaluation of new chemical substances, where a certain quantity marketed triggers the requirement of experimental investigations, no matter assumed toxicity. For existing substances another priority is valid. This is briefly based on a first IPS screening for risk and national priority among the high production volume chemicals (see 5.2) and uses already available data.
After the first elimination of substances, the EMEA guideline provides three different assessment schemes, depending on the application to the animals or exposure to environment: Direct entry into the water; internal application or external application. If applied directly to water, the phase II evaluation gives a procedure for prediction of the environmental concentration (Predicted Environmental Concentration - PEC). The same applies for externally applied medicine when substances are likely to enter the environment, e.g. the soil. The guidance document does not lay down criteria for estimating if and how the substances may enter the environment by external application. In the case of internally applied medicine, a cut-off concentration in manure and fresh dung regulate the need for further assessment. If the concentration of the active compound or major metabolite is found in concentrations below 100 µg/kg in manure and in concentrations below 10 µg/kg in fresh dung and no active ingredient or major metabolite is excreted by animal urine on pasture, no further assessment is considered necessary. If substances are degraded with a half-life faster than 30 days after spreading of manure/slurry on land, no further assessment is necessary, even though the concentration of the substance exceed a concentration of 100 µg/kg in manure. After the evaluation of the possible exposure routes, with the above mentioned possibilities to stop further assessment, the EMEA document prescribe that further assessment is needed, as described in a phase II decision tree, only if the PEC in groundwater and soil exceed 0.1 µg/l or 10 µg/kg, respectively.
The ERA of new and existing substances differs in several aspects from the initial risk assessment of veterinary medicinal substances. First of all, regardless of substance category the fundamental steps for risk assessment, i.e. hazard identification, exposure and effect assessment, risk characterisation and possible risk management, as laid down in the TGD for new and existing substances, is not followed in the risk assessment procedure found in the EMEA guideline. In several instances during initial exposure assessment it is possible to stop further assessment Furthermore, in cases where PEC for groundwater is < 0,1 µg/l or PEC for soil is <10 µg/kg no further assessment is required. In such cases no PNEC is elaborated and risk characterisation, by a comparison of PNEC and PEC, not given.
5.5 Phase II
Phase II deals separately with fish medicines and other veterinary medicinal substances. Major differences exist between the risk assessment of these two groups of medicines, but the rationale behind this is not properly explained in the guideline. Although variation in the exposure situation in the aquatic and terrestrial environment is large, the same principle in risk characterisation could be applicable with a few amendments.
5.5.1 Tier A
The Phase I of the ERA of veterinary substances does not require a collection of toxicity data. Only in cases where the cut off values for PEC in dung (10 m g/kg), manure (100 m g/kg), soil (10 m g/kg) or groundwater (0.1 m g/L) are exceeded, do the assessor has continue to Tier A of phase II. . Here data on biological effects for the first time has to be conducted for both medicines applied on soil and medicines applied directly in the water.
Veterinary medicinal substances other than fish medicines
In phase II, a minimum of information about physical/chemical and ecotoxicological properties must be obtained for all drugs. For example, e.g. Koc and degradation rate in 3 soils, toxicity test using plants, earthworms and if groundwater is subjected to risk also Daphnia. As indicated in the explanatory notes, but not described in the Decision Tree (Figure 5.3), the phase II -Tier A should include more detailed evaluation of the possible fate and effects of the drug and/or its major metabolites. Major metabolites are defined as metabolites with a PEC higher than 20% of the applied dose. For comparison, major metabolites of pesticides are defined as metabolites with a PEC higher than 10% of the applied dose . In phase II -Tier A, it should be considered whether there is indication of significant air exposure or exposure of freshwater systems by run-off. This is different from the ERA of chemicals, where all relevant compartments (areas of concern) in principle should be evaluated during the initial phase. Areas of concern for chemicals are: water, sediment, soil, air, secondary poisoning and the function of waste water treatment plants. Of these it is only the last that may be unimportant for veterinary drugs.
In cases where PEC in dung or soil exceed 10 µg/kg, Koc and degradation rates in three different soil types have to be determined. Furthermore, a phytotoxicity tests and an acute toxicity tests with earthworm is required. For veterinary drugs with insecticidal properties additional toxicity tests on fauna are required. If the PEC in groundwater is estimated to be higher than 0.1 µg/l the acute toxicity test with Daphnia has to be included in the basic investigations in Tier A. The determination of Koc and degradation rates may lead to a demand of further studies. PEC for surface water and acute toxicity tests with algae, Daphnia and fish have to be include if the Koc < 500. Data on the effects of the drug to soil micro-organisms is required if half-life of the parent compound is longer than 60 days. The recommendation of tests with plants and earthworms both refer to EU and OECD test methods, whereas an US-FDA microbial growth test is recommended.
The risk characterisation used in the EMEA guideline for veterinary medicinal substances differs from the ERA used for new and existing chemicals on the following issues.
- The risk characterisation of veterinary drugs does not consist of a direct PEC and PNEC comparison, which is an essential principle of the risk characterisation for new and existing chemicals. Instead effect values are compared directly with the PEC.
- In cases were half-life has been estimated to be shorter than 60 days in three soils, the basic requirements for test species only includes two throphic levels (earthworms and plants), compared to three level in the risk assessment of new and existing chemicals. If DT50 > 60 days, a MIC for microorganisms is required.
The risk characterisation of veterinary drugs does not consist of a direct PEC and PNEC comparison. Instead are the effect values compared directly with the PEC. This is mainly an academic difference if the actual application of safety factor is the same in the risk estimation. However, this is not the case. For compounds with a DT50 < 60 days, the exposure-effect ratio (e.g. PEC/EC50) has to exceed 0.1 to initiate an appropriate risk management strategy or a collection of further data on the fate effect (Tier B). Since PNECs normally are derived by the application of a assessment (safety) factor on the EC50-value (or NOEC), this correspond in reality to an application of an assessment factor of only 10. In cases of DT50 > 60 days the risk quotient for earthworms has to exceed 0.01 to initiate further measures, which correspond to a safety factor of 100.
These assessment factors are not in general accordance with the ERA procedure for new and existing chemicals (c.f. Technical Guidance Document to Directive 93/67/EEC and Commission Regulation No. 1488/94), where a use of an assessment factor of 1000 is prescribed as a starting point on the lowest short-term effect data from studies of at least three throphic levels, e.g. algae, daphnia and fish or plants, earthworms and micro-organisms. An assessment factor of 10 should normally, according to the TDG, only be applied when long-term NOECs are available from at least three throphic levels. The same assessment factors are also used for the high priority existing chemicals that annually are evaluated in a detailed risk assessment. The number of existing chemicals evaluated in EU each year is presently around 20. As a consequence of this, the ERA procedure of veterinary drugs, not used as fish medicine, generally uses a 10-100 times lower margin of safety in the effect assessment than the procedure used for industrial chemicals. The absence of PNEC calculation for terrestrial organisms and the indirectly choice of lower safety factors is not discussed in details anywhere in the EMEA guideline. However, the principles for evaluation of the undesired effects of veterinary drugs do to some extent resemble the approach and principles that are used to assess the risk on non-target soil fauna by using pesticides on agricultural land, as laid down in Directive 91/414/EEC (Uniform Principles). Some countries, like Denmark, have expressed concern about whether the margin of safety applied in the Uniform Principles is sufficient to effectively protect non-target organisms towards unintended adverse effects of pesticides, and have therefore approved the use of higher assessment factors (Box 5.2 for a discussion of risk management versus risk acceptance).
For veterinary substances, which have insecticidal properties, additional information on their toxicity to dung or grassland invertebrates is required if PEC exceed 10 µg/kg in dung or soil, respectively. Examples on evidence for insecticidal activities are given in Annex III of the EMEA guideline. The recommended test strategy for veterinary drugs with insecticidal properties follows guidelines given by the International Organisation of Biological Control (IOBC), which do not require a dose-respond calculation but rather gives cut-off values based on the effect level at relevant test concentrations. Concern for the environment is recognised in cases where more than 50 or 79% of the fauna is affected at PEC in dung and soil, respectively. If this is the case an appropriate risk management strategy or collecting of additional information on fate and effects has to made in a Tier B.
For veterinary drugs a risk characterisation for the groundwater compartment has to be made if PEC > 0.1 µg/l. The PNEC is derived by an application of an assessment factor of 100 based on the result from the acute test with Daphnia magna. In comparison the groundwater scenario in the TGD for chemicals is solely based on a risk assessment of human health a result of exposure through groundwater. The PNEC for human health is based on long term mammalian toxicity tests.
For very persistent veterinary drugs with a DT90 exceeding one year an appropriate risk management strategy has to applied or the applicant has to deliver more detailed information about fate and effects of the substance as prescribed in Tier B.
Fish Medicine
The ERA of fish medicine or veterinary medicinal substances released directly into water systems, also include drugs used in rearing and production of crustacean etc. Basic requirements includes Kow, an estimation of hydrolysis and/or photolysis, and toxicity studies with species covering three throphic levels, e.g. algae, daphnia (or other crustacean) and fish. Depending on the use of the medicine an evaluation of the fresh-water and/or marine system has to be made. On basis on the collected information a risk characterisation is performed (PEC/PNEC). A factor of 100 is used on the basic short term toxicity data set, based on the recommendation from a OECD workshop on extrapolation of laboratory toxicity data to the environment (OECD Monograph no. 33). Also in this respect the choice of assessment factor (100) is a factor of 10 below the one required for industrial chemicals on similar data.
Concern for the environment is recognised in cases where PEC/PNEC > 1 or Kow > 1000 or the half-life in water exceed four days. In such cases further assessment in a Tier B is required.
5.5.2 Tier B
If any of the trigger values are exceeded further assessment according to Tier B is required. At this stage the guidelines suggested in the EMEA document should only be considered as recommendations to the applicant, as many of the suggested tests, especially for the terrestrial environment is not standardised or even well developed. Therefore it is recommended to include the regulatory authorities already at the planning stage of tests required in a Tier B approach.
Veterinary medicinal substances other than fish medicines
As it was the case in Tier A, the ERA for the terrestrial environment (not presented in a decision tree) is mainly based on the ERA for pesticides, and includes e.g. identification of metabolites from the transformation pathway; effect on the microbial activity in the carbon and nitrogen mineralisation processes; sublethal effects on earthworms; further laboratory studies on the effects on plants and invertebrates found in dung and grassland or finally field studies on possible adverse effects caused by the substance on relevant plant and animal species.
An assessment of poisoning of terrestrial vertebrate wildlife following use or disposal of veterinary medicinal products has to be made for the first time in Tier B. This includes both direct effects and biomagnification effects. An arbitrary classification system of toxicity to wildlife as recommended by the US-EPA is used to indicate whether there is need for further assessment. The classification is based on avian single-dose oral LD50, and an avian dietary LC50 test which group the substances into 5 different classes ranging from practically non-toxic to very highly toxic. If the substance is found highly or very highly toxic further assessment is needed, whereas for "practically non-toxic" substances no further assessment is needed. For "slightly" and "moderately" toxic substances the EMEA guideline states that the need of assessment depends on whether the level of exposure is sufficiently high (not specified). The risk assessment procedure for veterinary substances on vertebrate wildlife differs on a few minor issues compared with the ERA for chemicals.
Whether there is a need for assessing secondary poisoning is made at a later stage than for chemicals, as the consideration of whether the substance has any indication of bioaccumulation potential has to be made in the initial phase for chemicals. If the compound has a potential for bioaccumulation in fish and belongs to a certain classification (T+/T/Xn and R48/R60-R64) a risk characterisation has to be made. The EMEA document is less specific on the criteria for bioaccumulation, but refer to information collected as part of the residue package, i.e. information concerning the use of substances in domestic animals and pharmacokinetic studies in fish. For toxic compounds used for dips a warning printed on the package warning users of the need to keep wildlife away from the dip. For toxic compounds with a likely exposure of wildlife through feeding on terrestrial invertebrates or aquatic species such as fish, a procedure for risk assessment is outlined in the EMEA guideline. A collection of sufficient toxicity data is recommended if available in the literature. The EMEA guideline indicates that an assessment factor of 10 should be applied to toxicity data on species likely to be exposed or to a wider range of toxicity data.
Fish medicine
In cases where PEC/PNEC > 1 or DT50 in water is longer than 4 days, further aquatic studies have to be included in the final ERA of fish medicines. These may include additional acute toxicity studies on aquatic species; long-term toxicity studies to aquatic species; toxicity studies on sediment species and bioaccumulation studies, depending on estimations on the environmental fate of the substance. If degradation studies indicate a short exposure period, the EMEA guideline prescribes that the PEC should be considered as a short-term PEC, to which it is only necessary to compare short-term effects on aquatic species. This leads to a recommendation of adjusting the assessment factor used on the basic acute toxicity set down to only 10, as it is no longer considered necessary to take long-term effects into account. It is not explicit in the guideline when exposure is considered to be acute or long term. This procedure of veterinary medicinal products is somehow different from the procedure used for chemicals, which do not take biodegradation into account when recommending assessment factors. For chemicals only a change in the set of toxicity data, i.e. changes in the number of test species or by going from acute to chronic exposure, may alter the use of assessment factors, whereas the biodegradation data is used when estimating PECs..
In cases where application is repeated, which is not unusual in many fish farms, the exposure may be more or less chronic although the substance is broken down relatively fast. Therefore, reduction in the assessment factor due to degradation properties should only be accepted if repeated exposure does not occurs or occurs at very long intervals like it is discussed in the section concerning intermittent releases of industrial chemicals in the TGD.
In cases where chronic exposure is likely, long-term studies on the most sensitive group of the acute toxicity base-set should be conducted. In deriving the PNEC on the basis of this test, the guideline prescribe the use of an assessment factor of 10. For chemicals a safety factor of 10 is only prescribed in cases of long-term studies of at least three species or on data from long-term studies on the most sensitive species in the short term tests, supplemented by at least one other NOEC.
In Tier B of Phase II, a guideline for assessing the fate and effects of veterinary substances in the sediment compartment is given for the first time. This is at a later stage than for chemicals, where a risk characterisation for all relevant media is an essential part of the basic RA. The EMEA guideline suggests an initial assessment of PEC based on an estimate of the partitioning between sediment and water as extrapolated on basis of the physical/chemical properties of the test substances. The estimated interstitial water concentration should be compared to the PNEC derived for water column species as a first screening for risk. If the PEC/PNEC ratio is of concern or if the compounds, due to e.g. very lipophilic properties, is unsuited for such extrapolation, test using spiked sediment must be carried out. The EMEA guideline recommend the same assessment factors for deriving PNEC for the sediment fauna as for the assessment of the pelagic PNEC. These are again one order of magnitude lower than in the ERA of chemicals.
As studies using spiked sediment is time consuming and expensive it makes sense to recommend a first risk characterisation based on extrapolated values. This is very much in line with the prescription in the TDG for chemicals. However, as this initial ERA of sediment is not associated with extra costs, it should be considered to employ this assessment already in Tier A, on equal terms with fresh and marine waters, for all veterinary substances released directly into the aquatic environment. When concern for the benthic fauna is identified by such an initial risk characterisation or if the Kow is higher than 1000, additional information using tests with spiked sediment is required in Tier B.
A parallel to the discussion above on secondary poisoning for the terrestrial environment can be made for the aquatic environment. It is not in line with the ERA of chemicals, not to have at least a first risk assessment of secondary poisoning in Tier A. Accumulation studies have to be carried out if the pharmacokinetic studies in fish for veterinary drugs indicate a potential for bioaccumulation. From the concentration of residues that are considered likely to occur in fish, on basis of the bioaccumulation studies, secondary poisoning is assessed by assuming a worst case situation where a 2 kg heron eats 500 g of contaminated fish pr. day. The risk characterisation follows the one for wildlife in the terrestrial environment, using the same assessment factors.
On basis of the initial studies in Tier B, the assessor has to estimate whether the substances pose a moderate to high risk to the environment or only pose a low or no risk at all to the environment. The EMEA guideline do not provide any criteria for deciding the level of acceptable risk, leading to a completion of the ERA, or a non acceptable risk leading to further studies, e.g. fate studies in sediment or aquatic field or mesocosm studies.
Box 5.2
Risk management and risk acceptance
The risk characterisation and the following risk classification of new and existing chemicals may necessitate a reduction of risk. For new and existing chemicals the process of developing strategies for risk reduction are given in a special Risk Reduction Guidance . For priority substances the repporteur shall suggest a strategy for limiting risk of the substance for man and environment, including control measures and/or surveillance programmes. Risk reduction can take many forms and many risk management decisions is often made on voluntary basis. Whether risk should be reduced or not is depending on the result of a detailed risk-benefit analysis. The overall objective of such a process is to identify risks and benefits by a single substances and a comparison of these to the risk and benefit of alternative substitutes. Economy is an integrated part of such a risk-benefit analysis. A substitution or ban may generate major economical impact. Protection of man and environment may hence be very cost-intensive and the highest possible protection should therefore be achieved by the money lost. The loss represent in this way the price society is willing to pay to reduce risk (risk acceptance). The guidance document on risk assessment of new veterinary medicinal products do somehow accept a lower margin of safety compared to industrial chemicals. . This may also in some circumstances be the case for pesticides. An argument often presented for accepting higher environmental risk by pesticides, is that they are released to the environment with the purpose to destroy specific harmful target organisms and hence application is associated with large beneficial result for the growth of crops. This may also be the case for fish medicine released directly into the water, but not for most veterinary substances released by dung or sewage sludge to the terrestrial environment. Although the use of veterinary substances without any doubt is associated with large benefit for the producers of domestic animals, the benefit is not a result of the release to the environment. In principle risk management should not influence the processes and assumptions made in the risk assessment. Therefore both veterinary substances and pesticides should be evaluated on the same basis as other chemicals If regulators or society, due to obvious beneficial results, are willing to accept a higher degree of environmental risk this could be realised through changes in the risk management and risk reduction strategies. |
5.6 Summary and conclusions
The risk assessment procedure for new and existing substances (chemicals) has briefly been reviewed and compared to the risk assessment of veterinary medicinal substances (drugs). A comparison reveal a number of differences of more less important character. The most important being:
 | No guidance exist on how and when to assess the environmental risk of already existing veterinary drugs, and apparently there is no current national or international regulatory strategy on how to deal with environmental risk of existing veterinary drugs. |
| The estimation of PEC for drugs is different from that of new and existing chemicals and plant protection products. The major environmental release is from treated animals, and is influenced by e.g. treatment intervals, dosages, formulation of drugs, metabolisation of drugs and by the species of animal treated. |
| The ERA procedure of drugs not used as fish medicine may stop further assessment on the basis of arbitrary cut-off values for the environmental concentration, i.e. a risk characterisation for drugs found in concentrations below 10 µg/kg in dung or soil can be left out because no risk is assumed. A half-life shorter than 30 days in soil of the parent compound may also stop the need for further assessment. |
| When making the risk characterisation by comparing the exposure level with the effect level of drugs (risk quotient), this is not done by directly estimating the PEC/PNEC ratio, but rather as a direct comparison of effect data with the PEC (e.g. PEC/LC50). The required assessment factors used for deriving PNEC for the terrestrial environment is at least a factor of 10 lower than the assessment factors prescribed for assessing chemicals, and in cases of compounds with half-life shorter than 60 days it is a factor of 100 lower. In the ERA of fish medicine the assessment factor is generally a factor of 10 lower than that for chemicals. |
| When conducting the initial risk characterisation for VMP, risk to the sediment and risk for secondary poisoning is not considered. Such a risk characterisation will only be made if the compounds are submitted for a further assessment in a Tier B of Phase II. |