Cheat Sheet
- This is the third blog in a three-part series. You can read Part 1 here and Part 2 here.
- Chemicals of concern are omnipresent in industry.
- GHS hasn’t led to as much “harmonious” practices as expected.
- When evaluating substance data, it is essential to understand the quality and reliability of the available test data.
This is the third part of a three-part series regarding chemicals of concern. In Part 1, we discussed the Globally Harmonized System of Classification and Labelling of Chemicals (GHS) framework, the differing adoptions per jurisdiction and the impact this has had on classifying chemicals of concern. In Part 2, we discussed the different data types used to evaluate chemicals of concern. In the third and final installment of this blog series, we will share best practices for using the individual data types and also important best practices for the hazard evaluation process itself.
Best Practices: Substance Test Data
When evaluating substance data, it is essential to understand the quality and reliability of the available test data as well as if a study was conducted using sound scientific methods. This is an important factor when deciding if a study should be used to support a classification or how much weight a study should carry when a weight of evidence evaluation—assessing if the science supports the conclusion—is applied. In addition to generally evaluating if a study was conducted according to sound scientific methods there sometimes may be a Klimisch score associated with the test data. The Klimisch score is one method of evaluating the quality and reliability of a study, and while it might not always be accurate depending on who assigned the score, it does at least give you another opinion about the study’s reliability.
One of the best pieces of information to help determine that is having a test method or protocol identified as part of the study information. You can look for things like the Organization for Economic Co-operation and Development (OECD) test guideline number or an assay name, such as a Buehler assay.
Anytime there is a specific test method or protocol identified, always check the classification criteria or guidance you are following to see if the test method is mentioned there. If it is, then this can help validate that the test is applicable for classification. If a test method is provided with the study but that method is not mentioned in regulation or guidance, then it doesn’t necessarily mean it can’t be used but the quality and reliability of the study must be evaluated.
Anytime you have a test method that is not mentioned in the regulatory text you can also compare that test method with another one that is listed in the regulatory text to see if they are similar. For alternative test methods, there is still work being done to validate some of these. If you run across one in your research, you should rely on regulatory agency guidance for how, or if, data from an alternative test should be used. If the agency doesn’t specify, you can always contact the regulatory agency for clarification on the use of an alternative test method in a hazard assessment.
Another important consideration in evaluating test data is to determine if the result is applicable to be used for classification purposes. Some examples where a test result might not really be applicable are as follows:
- Example 1: There is a study in rats that indicates a chemical causes cancer in an organ that is not found in humans; this study is likely not relevant to determine a classification in humans.
- Example 2: The test was conducted via intravenous administration. IV is not a standard form of workplace exposure, so this would not be a test that you would want to consider in determining the hazards of a substance.
- Example 3: You have information that the substance is metabolized differently in the test species than in humans. If the particular metabolic pathway in the test species does not exist in humans, then the test is probably not one that should be considered in determining a classification. The perfect example of this is methanol, which is metabolized differently in primate species than in rats, so a rat study would not be a good test to use to evaluate methanol for acute toxicity, and primate data would be better, if available.
Best Practices: Mandatory List Data (GHS Classifications)
As we discussed in our last blog, another data type is mandatory list data. With these types of mandatory classification lists, it is important to understand whether the classifications are baseline classifications or a full and complete list. A baseline classification list is one where additional hazards can be added if you have data indicating additional hazards are applicable.
A complete classification list, on the other hand, is one where no amount of additional data would warrant adding additional hazard classification(s). It is also important that, when using a mandatory list, you are familiar with the regulatory text. Using these lists is usually not as simple as just looking at the classifications and using them. Be sure to notice any footnotes or references to other places in the regulatory text that could be associated with the classification itself and take the time to understand what they mean. Also, read any portions of the regulation that indicate how to use the classifications. For example, Annex VI uses single and sometimes double asterisks as well as notes and even specific concentration limits. The single asterisk means that the classification given is a minimum classification, and other data should be evaluated to determine if a more severe classification should be applied. You need to read through the regulations to ensure you are interpreting these symbols and notes appropriately and applying them correctly.
Best Practices: Advisory List Data
In addition to the GHS classification lists, there are other lists, such as the International Agency for Research on Cancer (IARC), that will not give a direct GHS classification but can be used as a reference list to help derive a GHS classification. You will need to review the information on these types of lists and determine how to use them when determining a GHS classification. Sometimes there are guidance documents provided by the regulatory agency to assist in using this data. For example, HazCom 2012 gives guidance on how to classify for carcinogens in the U.S. using both IARC and National Toxicology Program (NTP). For example, and IARC Group 1 carcinogen aligns with a GHS Category 1A making the use of this data straightforward when classifying for the United States. NTP on the other hand leaves some room for interpretation when it comes to classification. If you use this information, you will need to spend more time evaluating what the NTP “reasonably anticipated” classification means and whether your substance should be considered a GHS Carcinogen Category 1B or 2. This is an interesting example as you might make the business decision to use IARC data over NTP data because it is more clear in its alignment with OSHA’s HazCom as opposed to NTP where you might need to take more time to complete an evaluation.
Best Practices: Raw Material Data
As we discussed in our last blog, there a two paths companies can take when choosing how to use raw material Safety Data Sheet information. Regardless of the path chosen, most will at least use data from Section 3, which is the composition or ingredient section of the SDS. There are some important things to understand when using this information. First you should determine which jurisdiction the SDS was written for. This is important because different jurisdictions have different substance classifications and mixture cut-offs, which, for example, can lead to different ingredients being listed in section 3 for a U.S. Occupational Safety and Health Administration (OSHA)-compliant SDS compared to a European Union-compliant SDS for the same product. Disclosure requirements also vary between jurisdictions, which can affect the substances listed in this section.
If choosing to use the raw material SDS for evaluation purposes, you should also determine the quality of that SDS. One of the best methods for achieving this is to reverse engineer the SDS. This is because a well-written SDS should have data in various SDS sections that support what is represented in other sections. Therefore, attempting to reverse engineer the SDS is a quick way to check the quality and reliability of the information provided.
At a minimum you should check the following:
- Material Classifications (in Section 2) against Ingredient Information (in Section 3): If classifications for the ingredients are listed, ask yourself: Do the classifications and ingredient percentages align with the appropriate cut-offs for the agency the SDS was written for?
- Section 9: If the material is a liquid and there is a flashpoint, verify that it was evaluated correctly and a classification if warranted is listed in Section 2.
- Check Section 11 and 12 to see what kind of test data is listed: If there is acute toxicity data or ecotoxicity data for the product listed in Section 11 or 12, is it consistent with the classifications in Section 2?
- Section 14: If there is a transportation classification, is it consistent with Section 2 classifications for those that align?
Section 3 – if classification information is listed for the ingredients, compare that data to any component test data listed in Section 11 and 12 to ensure they align. The last step in determining SDS quality is to check substance data and classifications against any available substance data and list classifications found in research. This will bring out discrepancies between substance data or classifications vs. what is listed on the SDS. There could be legitimate reasons for the discrepancies. (Maybe an outdated version of Annex VI was used on the SDS so you are not seeing the latest classifications).
If more information on ingredients is needed or inconsistencies are found in the SDS review, you might need more information from the supplier. Having good relationships with those suppliers will make them more likely to provide you with the necessary information.
Depending on the discrepancy and the regulation that is applicable to the classification being completed, you might need to reach out to the supplier to get more information to correctly determine the hazards in your product. There might also be other data you would want to request from the supplier such as the full formulation of the raw material since SDSs are technically only required to disclose the hazardous components per the jurisdiction that the SDS is written to comply with.
Although these are best practices, sometimes resources don’t permit this to happen or suppliers won’t cooperate with addressing the inconsistencies. This is where process and documentation become particularly important, which we will talk about in our last section: best practices during the hazard evaluation process.
Best Practices: Product Data
Although testing of products is not required under the GHS framework, there are some scenarios where it may be worth the time and money to generate data on a product. One example of this is if it based on the substances in the product, the mixture cut-off indicates that the product is corrosive. This is a new product and the product experts who reviewed the classification don’t think it is corrosive, but they also don’t have any data to substantiate that to be the case.
In this example, conducting a quick and inexpensive Corrositex test may confirm that a less-severe material classification is applicable. This could potentially translate to reduced shipping cost for your company and reduced usage cost, (less expensive/lower hazard personal protective equipment) for your customer. Physical property testing is also worth considering because it is relatively low cost, quick and, when determining physical hazard classifications, the criteria for classification only uses product data. There are no mixture rules for physical hazards.
GHS (and many regulations that have adopted GHS) also include criteria for some hazard classes called bridging principles. Bridging principles incorporate data of other “similar mixtures” in determining a new mixture classification and can be used when data for your specific product is not available.
Finally, I would like to end this series with some best practices that you may find useful during the hazard evaluation process.
Best Practices: Hazard Evaluation Process
Software: If using a software system to author your SDSs, know how to complete the calculations by hand. Utilizing software systems can make life a lot easier and more efficient, but they are not perfect. You need to be able to recognize if the output on the SDS is not accurate. If you don’t understand how the calculations are done, you will not be able to make a quick assessment to determine whether the classification provided on the SDS is accurate or not. Along these same lines, you need to understand how the software is using data. For example, is there a precedence in the type of data used for calculations and where did that data come from? By knowing this, you will be able to decide if you need to supplement certain endpoints with additional data or classifications.
Defined Process: Processes are very important, and the only way to ensure that multiple people are doing something in the same way is to document a process and then follow up to ensure that it is being followed. These processes will make it easier for you to train new employees or explain to an authority how you came about a certain classification.
Who classifies the substances? You don’t want just anyone doing substance hazard evaluations. These are the basis for all other classifications in the system or process. Changes to substances, other than mandatory regulatory updates coming from list classifications, should be well-evaluated as they will have a ripple effect on all materials containing them in a composition. Any changes to substances could impact the SDS and subsequently the label if the product classification changes. Anyone completing substance classifications should be well-trained and highly experienced.
Documentation: Anytime an evaluation for a substance or a mixture is completed, there is always a lot of data reviewed and many decisions made. It is important that the data used to classify, and any professional judgement decisions made are well-documented. This documentation should provide a snapshot of what the data looks like at the point when the evaluation was completed. The goal of the documentation is to be able to easily and quickly explain both why something was and why something was not classified. This is important because new data becomes available from time to time and the regulations change. Years down the road a company may need to defend the classifications and show why something was classified a certain way at a certain point in time. Or perhaps a customer is asking about a current classification for a material. This documentation is invaluable should a question about a classification ever arise.
In a world that is becoming increasingly more regulated, chemicals and chemicals of concern are no exception.
