The SDS You Approved Three Years Ago: Is It Still Valid?

The SDS you approved three years ago: is it still valid? It’s a question worth asking properly, because the honest answer for a lot of businesses is: nobody’s actually checked. An SDS isn’t a document you author once and file away. It’s a snapshot of a regulatory picture that keeps moving underneath it, and every year that passes without a review is another year the gap between what the SDS says and what’s actually true can quietly widen.

Why a compliant SDS doesn’t stay compliant on its own

When an SDS is authored, it reflects the classification, hazard data and regulatory status that applied at that exact moment: the Candidate List as it stood, the CLP classification in force, the GHS revision in use, the guidance ECHA had published to date. None of that is fixed. The regulatory framework around every SDS keeps evolving, and nothing about the document itself changes to keep up. It just sits there, becoming quietly less current, until something forces a review.

What actually moves underneath an SDS

• New SVHCs. If a substance in your formulation is added to the Candidate List (as we’ve covered in several recent posts), an SDS that made no mention of it can become out of date overnight, along with the Section 15 regulatory information and any Article 33 communications built on it.
• CLP classification updates. Harmonised classifications change, and self-classifications need periodic re-checking against current data and current guidance. A hazard classification that was correct three years ago isn’t guaranteed to still be correct today.
• GHS revision changes. The UN’s Globally Harmonized System is revised on a cycle and adopted into CLP with a lag. Hazard statements, pictograms and precautionary statements can shift as a result, even when the underlying science hasn’t changed.
• ECHA guidance updates. Even without a change in law, ECHA periodically revises its guidance on classification, labelling and SDS authoring itself, changing what “good” looks like without changing the regulation’s text.
• PFAS classification developments. This is one of the fastest-moving areas in chemical regulation right now, and it’s exactly the kind of thing that can turn a routine ingredient into a substance requiring active management with very little warning.

What actually breaks when this happens

It’s rarely the whole document. Usually it’s specific sections quietly going stale: Section 3 composition data no longer reflecting a substance’s current classification, Section 15 regulatory information referencing an outdated Candidate List position, or precautionary statements that haven’t caught up with a revised GHS category. Under REACH, there’s also a direct legal angle here: Article 31(9) requires an SDS to be updated without undue delay when significant new information affecting risk management measures, or new information on hazards or risks, becomes available, and for the revised version to reach recipients. An SDS that’s fallen behind isn’t just a quality issue; it can be a live compliance gap.

How to actually check

• Set a fixed review cadence for SDS documents rather than waiting for something to force the issue: annually at minimum for anything with SVHC or CLP exposure.
• Cross-check every substance in the formulation against the current Candidate List, not the one it was checked against at authoring.
• Re-verify classifications against current CLP entries and the GHS revision currently in force, rather than assuming the original classification still holds.
• Treat any Candidate List addition, restriction decision or authorisation update covering a substance you use as an automatic SDS review trigger, not something to fold into the next scheduled check.

How we can help

We review and re-author safety data sheets against the current regulatory position (not the one they were originally written against) and build a review cadence that catches drift before it becomes a compliance gap. If your SDS library hasn’t had a proper check recently, that’s worth finding out sooner rather than later.

← Back to Regulatory News