Acute Toxicity Estimates for Mixtures With Missing Data

An incomplete supplier SDS does not mean an ingredient is non-toxic. OSHA provides a defined method to calculate acute toxicity estimates, address unknown data, and disclose uncertainty on labels and SDSs.

A mixture may be classified even if a supplier has not provided a usable acute-toxicity value. Do not treat missing values as zero toxicity. OSHA prohibits guessing to fill data gaps, and lack of data does not indicate safety.

Under OSHA Appendix A, the acute toxicity estimate (ATE) for an untested mixture depends on its ingredients, their concentrations, the relevant exposure route, and the extent of unknown acute toxicity. The sequence of these checks is as important as the calculations.

Start with the evidence, not the formula

OSHA’s classification process starts with reliable toxicity data for the mixture. If relevant test results exist, they determine the classification. If not, assess whether an approved bridging principle, such as dilution or interpolation, applies. Only then use ingredient-based additivity. Valenc’s guide to GHS bridging principles addresses bridging; this article focuses on calculations when bridging is not applicable.

Assess acute toxicity by route: oral, dermal, or inhalation. Do not combine oral LD50, dermal LD50, and inhalation LC50 in one calculation. OSHA requires evaluating additional routes if evidence shows relevant acute toxicity through more than one route. Category limits and ATE units vary by route.

Decide which ingredient values can be used

For each ingredient, record its concentration and the most relevant acute-toxicity evidence. An actual LD50 or LC50 may provide the ATE. If no suitable numerical value exists but a valid acute-toxicity category or range is available, use the converted point estimates from Appendix A Table A.1.2. These converted values are not the same as category cutoffs.

For example, an ingredient classified as Acute Toxicity, Oral Category 3 receives a converted point estimate of 100 mg/kg for the additivity formula, not the Category 3 upper cutoff of 300 mg/kg. Category 4 converts to 500 mg/kg. Record the use of conversion values in the classification file.

A relevant ingredient is typically present at 1% or more. However, Appendix A requires considering ingredients below 1% if there is reason to believe they could affect the mixture’s classification, especially for highly toxic Category 1 or 2 ingredients. Concentrations below 1% are not automatically excluded.

Purely inert ingredients reasonably presumed not acutely toxic, such as water, are excluded from the additivity denominator. These are distinct from ingredients with missing toxicity data. This distinction ensures data gaps are not mistakenly treated as harmless.

Appendix A.1.3.6.1 specifies two additional cases. Ingredients with an oral or dermal LD50 above 2,000 and up to 5,000 mg/kg remain in the formula, even though this is outside the Category 4 range. Ingredients with only limit-dose test data at the Category 4 threshold showing no acute toxicity are excluded.

The standard ATE mixture calculation

When the necessary ingredient information is available, Appendix A.1.3.6.1 uses:

100 / ATE_mix = Σ [ C_i / ATE_i ]

Here, C_i is each relevant ingredient’s concentration as a percentage, and ATE_i is its acute toxicity estimate for the same exposure route. Sum the concentration-to-ATE ratios, then divide 100 by this sum to determine the mixture estimate.

Worked example: oral ATE from known ingredients

Consider an illustrative formulation with two oral-toxic ingredients and a nontoxic carrier:

100 / ATE = 0.10 + 0.04 = 0.14  →  ATE = 714.3 mg/kg

OSHA’s oral acute-toxicity bands classify values greater than 300 and up to 2,000 mg/kg as Category 4. This is a teaching example, not a classification of any commercial product. Actual classification must use applicable ingredient evidence and relevant interaction information.

Missing acute-toxicity data changes two decisions

An ingredient may have an SDS without a reliable ATE, or the provided toxicity information may be insufficient for a justified estimate. Appendix A.1.3.6.2 allows a technically qualified evaluator to determine if other evidence, including related-substance information, can support a derived value. Such extrapolation requires substantial expertise and does not permit inventing numerical values.

If no reliable estimate can be derived, distinguish the ingredient with missing data from genuinely nontoxic ingredients. OSHA’s current Appendix A establishes two thresholds with different functions:

• At 1% or more: if a relevant ingredient has unknown acute toxicity and the mixture was not tested as a whole, the label and SDS must state what percentage of the mixture consists of ingredient(s) of unknown acute toxicity (Appendix A.1.3.6.2.3 and Appendix C.3.3).
• Above 10%: adjust the additivity formula for the combined concentration of relevant unknown-toxicity ingredients. At 10% or less, use the standard additivity formula.

For the adjusted calculation in Appendix A.1.3.6.2.4:

[ 100 − total % of relevant unknown-toxicity ingredients ] / ATE_mix = Σ [ C_i / ATE_i ]

The denominator still includes only appropriate known-toxicity terms. This adjustment does not convert unknown ingredients into known ones or provide a definitive estimate of the mixture’s true toxicity.

Worked example: 15% unknown toxicity changes the category

Suppose an untested formulation contains the following ingredients, evaluated for oral toxicity:

The sum for known ingredients is 0.32. Since relevant unknown toxicity totals 15%, the adjusted numerator is 100 minus 15, which equals 85.

85 / ATE = 0.32  →  ATE = 85 / 0.32 = 265.6 mg/kg

This result falls in Oral Category 3 (greater than 50 and up to 300 mg/kg). If 100 were incorrectly used as the numerator, the result would be 312.5 mg/kg, which is Category 4. Omitting the 10% adjustment would change the reported category.

This estimate classifies the mixture based on known ingredients; it does not indicate that the 15% unknown ingredient is harmless. The missing data and calculation method must remain clear to the reviewer.

Where the unknown-toxicity statement belongs

Appendix C.3.3 and Appendix D provide the label and SDS requirements for ingredients of unknown acute toxicity. In the worked example, the applicable disclosure is that 15% of the mixture consists of ingredient(s) of unknown acute toxicity (oral).

OSHA’s HazCom Q&A states that this disclosure applies when a mixture is classified as acutely toxic and the classification is not based on whole-mixture testing. The statement may specify individual exposure routes, as OSHA recommends, or use the route with the highest unknown percentage if multiple routes have unknown toxicity. For a classified mixture, the unknown-data statement does not replace the required hazard classification, signal word, pictogram, or precautionary information.

Include this information in SDS Section 2, where the unknown-acute-toxicity statement belongs. Do not place it only in Section 11 while leaving Section 2 blank. Section 11 should still report available toxicological findings and explain relevant data gaps.

For how Section 2 should line up with Sections 3 and 11, see The 16 Sections of an SDS and our guide to SDS Section 2 across HazCom, WHMIS and CLP.

Four errors that produce an unreviewable ATE

1. Using the wrong route or units. An oral dose in mg/kg is not an inhalation concentration in mg/L; an inhalation value must also be interpreted using the applicable exposure duration and gas/vapor/dust criteria.

2. Using the category limit as the conversion estimate. Appendix A Table A.1.2 has specific converted values that must be used when only a category is known.

3. Classifying unknown data as nontoxic. The absence of an LC50 on a supplier SDS is a data gap, not a reassuring test result.

4. Forgetting the adjusted numerator. A relevant unknown fraction exceeding 10% can change the calculation and even the classification category.

Keep a calculation record that another reviewer can reproduce

An adequate review file should include the formulation version, ingredient names and identifiers, actual concentrations, supplier SDS revisions, selected route, source of each ATE or converted point estimate, treatment of inactive and unknown ingredients, the equation used, and the final category. Additionally, document the reasons whole-mixture testing or bridging were not used and identify who approved any expert judgment.

Manufacturers reviewing multiple mixtures often face challenges linking inputs to outputs. Valenc SDS Indexing provides supplier values as structured, reviewable fields. Valenc GHS classification applies published rules and retains both the inputs and the logic behind the results. While AI can extract data from supplier documents, it should not estimate missing LD50 values.

Questions reviewers ask

Classification should show where the uncertainty remains

A good ATE calculation is not simply a number in a spreadsheet. It is a documented decision about which data were used, which values could not be established, and which OSHA rule applies. Retain the evidence, the equation, and the unknown-toxicity statement together.