Summary of public comments received on the science approach document for the chemical screening and prioritization: Health Canada’s automated workflow for prioritization (HAWPr)

Comments on the Science Approach Document for the Chemical Screening and Prioritization: Health Canada’s Automated Workflow for Prioritization (HAWPr) were submitted by Association de l’aluminium du Canada, Canadian Paints and Coatings Association, Chemistry Industry Association of Canada. Ecojustice and Environmental Defence, Fragrance Creators Association, Methylacrylate Producers Association, Mining Association of Canada with input from Cobalt Institute, Humane Society International, International Zinc Association and NiPERA, and an individual.

Summarized public comments and responses are provided below, organized by topic.

Overarching comments

Comment summary 1: The Government should better communicate that the HAWPr process will be used for prioritization and triaging substance groups, and not to pre-determine or inform risk assessment outcomes. More emphasis should be placed on this tool as a “first-level screening tool” and that the results are meant for prioritization and the triaging of substance groups. Also, comment on how HAWPr fits into the approach for the identification of risk assessment priorities (IRAP) is not clear in Fig1-1. Request that the IRAP expert review process be maintained to validate the future use of the HAWPr to inform decisions on future priorities.

Response 1: Additional text has been added in the science approach document (SciAD) to clarify the role of HAWPr in the overall prioritization process. HAWPr was designed to be conservative with respect to the identification of both potential hazard and exposure candidates for prioritization and that additional manual review and scoping will be required to confirm the results from this tool.

The broader IRAP review considers HAWPr’s results in conjunction with other pertinent information on the substance(s), such as, for example, the results of previous assessments conducted under the Chemicals Management Plan (CMP), as well as the conclusions of these risk assessments and any existing (or planned) risk management activities in place.

Comment summary 2: How does the Government intend to tackle the thousands of substances identified in the SciAD?

Response 2: HAWPr was designed to be conservative with respect to the identification of both potential hazard and exposure candidates for prioritization. It is recognized that additional manual review and scoping will be required to confirm the results from this tool.

Comment summary 3: The SciAD states that there is a “need to continue to innovate and advance the development of new approach methodologies (NAMs) to support ongoing data driven prioritization and assessment activities to address key data gaps”; however, it does not clearly articulate how this will be achieved. Stakeholders assert that emphasis should be on developing more robust NAMs and increasing confidence in existing NAMs. They suggest that the Government establish an interagency team composed of members from governmental agencies responsible for requiring, using, generating, and disseminating toxicological and safety testing information.

Response 3: The Government of Canada recognizes the need to replace, reduce or refine the use of vertebrate animal testing when assessing the potential harms that substances may pose to human health and the environment under the Canadian Environmental Protection Act (CEPA). Health Canada (HC) and Environment and Climate Change Canada (ECCC) are working to advance this work on several fronts, including through the development, standardization and incorporation of NAMs into hazard and risk assessment activities.

The HAWPr SciAD outlines the current use of NAMs, where available, for prioritization purposes. To guide continued efforts towards the replacement, reduction or refinement of vertebrate animal testing under CEPA, HC and ECCC have published a strategy to more explicitly articulate ongoing work in this area.

Comment summary 4: Comment requesting that the Government refrain from using animal models and testing in chemicals research, moving forward.

Response 4: On June 27, 2023, it was announced that the Food and Drugs Act (FDA) was amended to ban cosmetic animal testing. Beyond cosmetics, HC is working with the international scientific and regulatory community to develop, validate and implement effective alternatives to animal testing. Bill S-5, Strengthening Environmental Protection for a Healthier Canada Act, which received Royal Assent on June 13, 2023, added several provisions to CEPA. These provisions are aimed at replacing, reducing or refining the use of vertebrate animals in toxicity testing and a strategy has been developed to guide these efforts. When conducting an assessment, information is gathered from a wide range of sources, which may include existing studies that used animal testing.

Comment summary 5: Consider collaboration in risk assessment approaches as part of the Regulatory Cooperation Council (RCC).

Response 5: The Government of Canada will continue to work closely with partners in the US as well as others internationally to maximize efficiencies and ensure the best science is utilized in risk assessment and risk management work and decisions.

Comment summary 6: Comment on the ability of HAWPr to properly characterize hazard and exposure to metals, and deal with classes of substances, in particular metals.

Response 6: HAWPr was designed to be conservative with respect to the identification of both potential hazard and exposure candidates for prioritization and that additional manual review and scoping will be required to confirm the results from this tool. The results of HAWPr are one of many sources used to inform potential priorities. This is particularly true for classes of substances, such as metals, where it is recognized that the results of HAWPr may require further consideration of available information at a broader class level, including, but not limited to, previous or ongoing assessment and regulatory activity of these classes.

HAWPr Process

Comment summary 7: Clarification is needed to determine whether existing NAMs presently used in risk prioritization, including Bioactivity Exposure Ratio (BER) and Threshold of Toxicological Concern (TTC) fall within the HAWPr context or outside of it. The Government should include BER and TTC tools in the HAWPr process.

Response 7: HAWPr uses select NAMs for flagging potential hazards, however the TTC and BER based approaches are not currently deployed in HAWPr. Both the TTC and BER approaches require quantitative estimates for human exposure which occur during more comprehensive risk assessment stages. The TTC and BER approaches remain viable tools in the CMP risk assessment toolbox.

Comment summary 8: Stakeholders assert that unless thresholds are adequately sensitive, health criteria are comprehensive, data sources are weighted towards independent science, and transparency is dramatically increased, this tool may end up reinforcing and replicating weaknesses observed in previous assessments. Stakeholders assert that the lack of data on the many undisclosed substances in products limits the ability of the HAWPr process to adequately characterize the exposures that drive prioritization. HAWPr’s scanning tools must be able to identify biased data and de-emphasize their influence on its scanning and prioritization.

Response 8: A goal of HAWPr was to increase the transparency of the data and approaches used in the prioritization process to facilitate review and reproducibility. This will continue to be improved as additional data sources are identified and incorporated into the tool. Within the SciAD, steps were taken to check the results of the tool against previous more manual techniques used in previous cycles of IRAP. HAWPr performed well in identifying substances that were previously identified for further consideration.

Within the hazard and exposure modules, HAWPr triages data sources with various levels of associated confidence and bases the indicators on the best information available. The hazard and exposure indicator criteria can be continually refined as experience is gained in subsequent IRAP reviews. Manual review of data for substances identified for further work will help to reduce the impact of biased data on the selected outcome.

Hazard Module

Comment summary 9: Classification data should be weighted differently depending on the confidence associated with those classifications. For example, known carcinogens should be weighted higher than suspected carcinogens.

Response 9: IARC classifications used within HAWPr are considered fit for the purposes of prioritization. Regardless of hazard endpoint, additional manual review, triage and scoping of substances identified as having hazard indicators by HAWPr would include, but are not limited to, consideration of previous or ongoing assessment activity both domestically and internationally, additional hazard information not captured by HAWPr sources, and existing risk management measures already in place (if applicable). Moreover, any analysis conducted as part of an IRAP is reviewed by Health Canada during the risk assessment phase prior to decisions on regulatory outcomes.

Comment summary 10: The Government should consider a separate metal-specific screening approach for endocrine disruption since metal interactions with the endocrine system are very complex and some metals are essential for physiological systems.

Response 10: Substance specific attributes and toxicities are typically considered during the assessment phase where a more in-depth analysis is conducted. Nevertheless, as data becomes available, HAWPr may be expanded in the future to better address metal compounds.

Comment summary 11: The Government should provide links to the specific references used to make each hazard determination, rather than list generic descriptions.

Response 11: Examining the level of information in the summary output files will be part of future updates to HAWPr. Given HAWPr queries thousands of data points it will be important to achieve an appropriate balance between providing an overwhelming amount of output data and critical data points used for the decision on each indicator.

Comment summary 12: The Government should include other potentially important hazard endpoints of highest concern, such as immunotoxicity and neurotoxicity.

Response 12: Given the modular nature of HAWPr, it will be possible to add other hazard endpoints like immunotoxicity and neurotoxicity in the future as datasets become available. Ongoing research and emerging data in these areas is being monitored, including in NAM development, and various assays and data sources will be added as they become available for future cycles of IRAP.

Comment summary 13: The Government should remain flexible and adjust dose thresholds over time, based on real-world experience.

Response 13: The thresholds for repeated dose toxicity were selected to be conservative while aligning with GHS criteria to the extent possible for use in a high-throughput prioritization process within HAWPr. As experience is gained, it is possible that the thresholds will be revisited in future applications of the tool.

Comment summary 14: The Government should include input from industry and academic experts in the endocrine disrupting compounds (EDC) module . It should also run a number of naturally occurring EDCs through the process as a proof-of-concept and to develop a foundation of relative risk.

Response 14: The current use of the EDC modules for estrogen receptor (ER) and androgen receptor (AR) modalities is considered fit for purpose which is to identify the potential for EDC activity. The EDC module may be further refined based on scientific advancements for the testing and assessment of EDCs. For example, Health Canada, collaborating with researchers from regulatory, industry and academic groups, recently participated in the development of an in silico approach for endocrine activity assessment (Johnson et al., 2025). Elements of this approach are currently reflected in HAWPr and will continue to evolve as science progresses. For chemicals on the priority list, a more fulsome determination of EDC potential, including an evaluation of evidence for adversity, would be part of the risk assessment process.

Comment summary 15: Reliance on international data sources that have not been fully reviewed, and classifications taken from Safety Data Sheets (SDS) that have not been verified may result in overly conservative outcomes.

Response 15: The SDS were not used in HAWPr as a source of hazard information (including classification data), but rather as indicators of potential exposure.

Comment summary 16: Clarification is requested for nanoparticles, which do not behave the same way as larger particles in both in vitro and in vivo studies.

Response 16: HAWPr does not explicitly consider nanoparticles and nanomaterials. Instead, the assumption is that data being used represents information on the “bulk” form of a substance from a given listing on the Domestic Substances List (DSL). The Government is committed to assessing nanoscale forms of DSL substances in a separate process. The recently published CMP Plan of Priorities includes 3 nanoscale substance assessments that are currently underway. The Government also plans to publish a DSL nanomaterials strategy document that will outline the selected prioritization approach and list of nanoscale substances prioritized for assessment, with a focus on those 53 DSL nanomaterials that have been previously identified as being in commerce in Canada. This list currently includes 5 CAS RNs that contain the aluminum moiety.

Exposure Module

Comment summary 17: Thresholds used for characterizing ingredients in SDS will result in the approach failing to capture substances. Issues of using SDS of products that may be available or used in Canada.

Response 17: The SDS used by HAWPr reflect available sources of information from retailers that sell consumer products to people in Canada. Preference is given to Canadian retailers directly, however, in some cases, retailers that operate in Canada may have SDS available from US operations. In the latter case, these are leveraged as indicators of potential exposure for people in Canada and attributed with lower confidence scores in HAWPr. Focus at present is on the occurrence of substances as ingredients listed in SDS, and not on specific concentrations. It is recognized that many SDS do not capture ingredients present below certain thresholds (for example, SDS do not typically list substances present at a concentration of less than 1%). This is a known uncertainty with the use of this data source.

Comment summary 18: Stakeholders have provided additional information sources for exposure estimation of fragrance chemicals.

Response 18: While it is recognized that fragrances typically result in low exposure levels, the information available for use in HAWPr is not sufficient to ensure that the substance is only used in this manner and so to be conservative these substances are identified as having potential exposure. The information provided by the stakeholders will be further considered in the manual triaging and scoping of the substances prior to recommending overall outcomes.

Comment summary 19: Stakeholders assert that the lack of Canada-specific exposure data is a limitation to HAWPr. They note that although the collection of additional data can be used to strengthen risk-based decision making and prioritization, the value of the data requested, as well as the associated costs and timelines must be considered.

Response 19: Noted.

Comment summary 20: Stakeholders request clarification on why all metals appear to be classified as high-risk, and if this is influenced by production quantity which does not necessarily equal exposure quantity.

Response 20: The exposure module, like the hazard module, is meant as a screening tool to identify potential indicators of exposure and or hazard, which prioritizes substances for review as part of a broader IRAP approach. To that end, HAWPr is a tool to efficiently and transparently collate and prioritize information on 1000’s of substances to identify indicators of potential risk that require further review, consideration of additional data and information etc. Input from interested parties and stakeholders, in particular with respect to additional data to incorporate into the tool, is welcomed at any time.

HAWPr outcomes and future work

Comment summary 21: HAWPr still needs more development, testing and validation. Could industry be involved in further development of the tool.

Response 21: The exposure and hazard modules are meant as a screening tool to identify potential indicators of exposure and or hazard, which prioritizes substances for review as part of a broader IRAP approach. To that end, the results of HAWPr do not represent a prioritization based on comprehensive data evaluation and the determination of “true risks”, but rather a tool to collate, screen and prioritize information on 1000’s of substances, efficiently and transparently, to identify indicators of potential risk that require further review, and consideration of additional data and information. Input from interested parties and stakeholders, in particular with respect to additional data to incorporate into the tool, is welcomed at any time.

Comment summary 22: Stakeholders note that a conservative threshold could impact the overall assessment of priority substances. It is suggested that a ‘low confidence’ positive by itself should not be used to differentiate such an outcome from a ‘non-priority’ hazard group.

Response 22: The 2900 additional substances identified by HAWPr that were not captured by IRAP in previous iterations reflects the impact of developing a tool that is able to consider a much larger amount of data than IRAP was capable of previously. The ability for HAWPr to look at study data directly, in conjunction with developing a system to bin this data using thresholds and criteria, allowed the tool to screen more substances for potential indicators of hazard, that may require further investigation and review.

Comment summary 23: The Government should conduct a pilot project using HAWPr to prioritize options relative to vulnerable populations and consider the addition of indirect exposures to modelling.

Response 23: Noted. Work is ongoing to determine what role indirect exposures may play in future iterations of HAWPr and how the approach can better consider populations that are disproportionately impacted.