SDS Section 16: Other information

The section that dates the other fifteen

Section 16 contains no hazard, no limit and no instruction. It contains the document’s own paperwork: when it was written, which version it replaces, what moved since then, what its abbreviations stand for, and where its numbers came from. That reads like housekeeping, and it is — but it is the housekeeping that determines whether anything above it can be relied on.

Which is why an auditor opens it first. A flash point in section 9 is a fact about the substance; a flash point on a sheet last revised before the amendment that changed the product’s classification is a fact about a document nobody has looked at. Neither REACH nor the Hazard Communication Standard sets a review interval — there is no rule that a safety data sheet expires after three years. Both are event-driven instead, and that is precisely what makes the date diagnostic. An old date is not automatically wrong. It is a question: has genuinely nothing happened to this product since?

The date, and the three places the three texts put it

EU — page one carries the date, section 16 carries the changes

REACH Annex II splits the job, and a lot of sheets get the split wrong. Part A subsection 0.2.5 puts the date of compilation on the first page; on a revised sheet it must appear there as Revision: (date), together with “one or more indications of which version is replaced, such as version number, revision number, or supersedes date”. Section 16’s own duty, under 16 (a), is different in kind: a clear indication of where the changes were made, unless that is shown elsewhere, plus the supplier’s standing obligation to be able to explain them on request.

Article 31(9) of REACH is what forces the date to move. The supplier updates without delay as soon as new information which may affect risk management measures or new hazard information becomes available, once an authorisation is granted or refused, and once a restriction is imposed. The new dated version then goes free of charge to every recipient supplied in the preceding 12 months — the obligation reaches backwards, to customers who have already bought and stored the product.

US — the date is the section

OSHA’s Appendix D names the section “Other information, including date of preparation or last revision”, and the whole of its content requirement is one line: “The date of preparation of the SDS or the last change to it.”

Two things follow. First, section 16 is mandatory in the United States. Appendix D requires the Table D.1 information for sections 1-11 and 16, and it is sections 12 to 15 that “may be included in the SDS, but are not mandatory”. Section 16 is enforced even though the four sections immediately before it are not — worth saying plainly, because the tail of the sheet is widely assumed to be optional throughout. Second, the clock runs differently: §1910.1200(g)(5) allows three months from the preparer becoming newly aware of significant information regarding the hazards or ways to protect against them, and §1910.1200(g)(6)(i) delivers the new sheet with the first shipment after the update rather than to past customers.

UN — A4.3.16

GHS Annex 4 asks for the date of preparation of the latest revision and, where revisions have been made, a clear indication of where the changes are unless indicated elsewhere; suppliers “should maintain an explanation of the changes and be willing to provide it upon request”. It adds the abbreviations key and the literature references, then softens the last one: references are not necessary, but may be included if desired. The EU text carries no such let-out.

What the EU puts here that the US does not

The full text the rest of the sheet left in code

This is the requirement that makes section 16 load-bearing rather than decorative. Subsection 2.1 allows the classification, hazard statements included, to be given in short form — provided “reference shall be made to section 16 where the full text of each classification, including each hazard statement, shall be given”. Subsection 3.2.3 does the same for the ingredient table: codes are sufficient there, and the full text of each relevant hazard statement is listed in section 16. Subsection 16 (e) closes the loop from the other end — write out in full any statement not written out in full under sections 2 to 15.

Note that 3.2.3 covers supplemental hazard statements too, which is where the EUH codes live. A sheet carrying EUH208 in section 3 and an empty section 16 has reduced its sensitiser warning to six characters. The wordings sit at /h-statements/ and /p-statements/. Nothing analogous arises under Appendix D: section 2 there requires the label elements themselves, so no code is left unexpanded.

A key to the abbreviations — and why one is needed at all

Subsection 0.2.4 instructs the author that the language “shall be simple, clear and precise, avoiding jargon, acronyms and abbreviations”. Subsection 16 (b) is the fallback for the acronyms that could not be avoided: DNEL and PNEC in section 8, study and endpoint abbreviations in section 11, PBT, vPvB and BCF in section 12. The test is whether every abbreviation actually printed on this sheet appears in the key — not whether the key looks respectable.

The classification method, for mixtures

Subsection 16 (d) requires an indication of which of the methods of evaluating information referred to in Article 9 of CLP was used for the purpose of classification. Article 9 sets the routes out in order: apply the criteria to the information available; where only ingredient information exists for a mixture, apply the bridging principles in section 1.1.3 and in each section of Parts 3 and 4 of Annex I; and where the information does not allow bridging, use the other methods described in those Parts — the calculation approaches. Those routes carry very different evidential weight, and 16 (d) is what lets a reader tell them apart.

References and training advice

16 (c) asks for the key literature references and sources for the data — the audit trail behind sections 9, 11 and 12. 16 (f) asks for advice on any training appropriate for workers. Neither has a US counterpart: under HazCom, training is an employer duty in §1910.1200(h), not a field on the sheet.

How section 16 connects to the rest of the sheet

• Section 2 is allowed to print bare classification and H-codes only because 2.1 points here for the full text. Section 16 is the other half of section 2.
• Section 3 does the same under 3.2.3, for ingredient hazard statements and supplemental EUH statements alike.
• The classification method named under 16 (d) is the method that produced everything in section 2 — bridging and calculation are visible nowhere else on the sheet.
• The abbreviations key resolves terms coined in section 8, section 11 and section 12, which are the three sections that generate them.
• Section 15 is the usual reason the date moves: an authorisation, a restriction or a new entry there is exactly the Article 31(9) trigger that obliges a reissue.

What a good section 16 looks like

1. Can you date the sheet and place it in a sequence? A date alone tells you when. A date plus “supersedes version 2.1” tells you whether the copy in your own file is the current one.
2. Can you read every code on the sheet without leaving it? Take three H-codes and two abbreviations from sections 2, 3, 8 and 11 and look them up here. If any of the five is missing, 16 (b) or 16 (e) has failed.
3. Does it say what changed, in terms someone can check? “Section 8 exposure limits updated; section 14 packing group corrected” is an answer. “Editorial revision” is a placeholder, and 16 (a) asks the supplier to be able to do better on request.