FOI release

FOI ATI1138 - Risk assessment process

Published 5 February 2026

1. ATI1138 Request

Request for all files, information and data detailing the risk assessment process, criteria, assumptions, limitations and conclusions that have been used to assess the risk to children, both through short term and long term exposure, that are exposed to topical, i.e. spot on, pet treatments for fleas and ticks that contain Fipronil. This to include consideration of the metabolites of fipronil degradation such as fipronil sulphone. These documents to include the rationale behind the risk assessment outcomes and decision making in relation to the authorisation and classification of topical pet treatments for fleas and ticks containing fipronil.

2. Our reply

• The risk assessment process including criteria, assumptions, limitations, and conclusions that have been used to assess the risk to children, both through short-term and long-term exposure: 

A user risk assessment (URA) is conducted following the available user safety guidance. The URA is a stepwise process which begins with the identification of the hazard(s), and an exposure assessment, these data are then used to conduct the risk characterisation and, if required, risk mitigation measures are implemented.  

For the hazard identification, a battery of pharmacological and toxicological data on the active substance(s), and formulation are assessed (see EU Regulation 2021/805 for the exact data requirements). These include both acute and chronic toxicity study data, as well as study data to evaluate the risk to potentially vulnerable sub-populations, such as pregnant women, including unborn children. From these study data, suitable points of departure (PoD) are identified (usually the no observable adverse effect level (NOAEL)) for comparison with reasonable worst-case exposure scenarios for the users and householders in the risk characterisation (see below).  

For fipronil, NOAELs are typically based on general toxicity (e.g., decreased bodyweight/bodyweight gain, increased liver/thyroid weight) and/or neurological signs after repeated oral/dermal administration to rodents or other laboratory species. Additionally, it should be noted that the breakdown, or metabolism, of active substances is considered during the assessment process. For example, fipronil is primarily metabolised to fipronil sulfone in mammals (including humans) and, therefore, the toxicological study data on fipronil (generated in laboratory mammalian species) investigate the toxicity of exposure to not only fipronil, but also any metabolites, including fipronil sulfone. 

The main guideline used for the URA is the CVMP ‘Guideline on user safety for pharmaceutical veterinary medicinal products’ (EMA/CVMP/543/03-Rev.1), However, for products which are applied topically (e.g. spot-ons, sprays, and collars for the treatment of fleas/ticks) a second, complimentary guideline is also used, the CVMP ‘Guideline on user safety of topically administered veterinary medicinal products’ (EMA/CVMP/SWP/721059/2014).  

This guideline focuses specifically on how to assess user safety for topically administered VMPs that may remain on the surface of the animal’s body for an extended period after application. It is recognised that exposure to topically administered products may occur via direct exposure to the product (e.g., accidental ingestion or accidental spillage), or when householders come into contact with the treated animal(s) in the time after administration. The assessment of exposure is divided into both short-term exposure (a single exposure event or exposure for a short time following treatment) and long-term exposure, as a consequence of contact with a pet in the days and weeks after treatment. Worst case exposure is estimated based on conservative default assumptions on the transfer of residues from the treated animal to the owner/child and this is done for the period of time spanning the duration of claimed efficacy (e.g. 28 days). More accurate estimations of exposure may be achieved through the generation of experimental data using the product; these include ‘wipe-tests’ in which treated animals are stroked using a gloved hand at regular timepoints after treatment, to measure the transferable residues remaining on the treated animal.  

The risk characterisation for each exposure scenario involves comparing an appropriate PoD with a reasonable worst case exposure estimate, resulting in a margin of exposure (MoE) calculation for each exposure scenario.   

The risk assessment process has conservatism built into each step, whether that be with the use of reasonable worst-case exposure estimates for each exposure scenario, or the incorporation of uncertainty factors to determine an acceptable threshold against which the MoE is compared (e.g. to account for intra- and inter-species variability, the severity of effects seen in the study data, and/or the quality of the available data).  If the MoE is below the acceptable threshold and a risk is identified, appropriate risk mitigation measures (RMM), such as user safety warnings and/or child-resistant closures, are then implemented (on the SPC and product information).  For flea and tick products, these RMMs typically include warnings to mitigate risks associated with acute exposure (e.g. advice on what to do in the event of accidental ingestion) and also later exposure as a result of contact with the treated animal (e.g. ensuring that animals are treated in the evening so that children avoid contact with the application site and that owners (especially children) should not be sleeping with treated animals).  

• The rationale behind the risk assessment outcomes and decision making in relation to the authorisation and classification: 

Veterinary Medicinal Products are evaluated on a case-by-case basis and may have different warnings, depending on the product and its identified risks. All VMPs are authorised on the basis of a positive benefit:risk balance, which takes into account the therapeutic benefits to the animals being treated, the additional benefits to those that come into contact with the treated animals, including the potential for a reduced risk of zoonotic diseases transmitted by fleas and ticks, as well as the risks to users and the environment from exposure to the therapeutic substances, which are usually mitigated as described above.