SDS Section 4: First aid measures

Written for whoever is standing there

Every other section of a safety data sheet assumes a reader with time. Section 4 does not, and REACH Annex II says so before it says anything else. The section must describe the initial care in such a way that an untrained responder can understand and provide it, without the use of sophisticated equipment and without the availability of a wide selection of medications. If medical attention is required, the instructions must state this — including its urgency.

That opening sentence is the test the rest of the section has to pass. It rules out clinical vocabulary. It rules out equipment the room does not have. And it rules out “consult a physician” standing in for a statement of how fast. UN GHS Annex 4 frames the section the same way and adds a suggested shape: the immediate effects by route of exposure, then the immediate treatment, then possible delayed effects with the medical surveillance they require.

One thing this page is not. What follows describes what the regulation requires the sheet to contain. It is not first aid guidance, and it is no substitute for the sheet supplied with the product in front of you or for professional medical advice.

4.1 — instructions someone can actually follow

Route by route, because that is how exposure happens

Subsection 4.1.1 does not ask for first aid advice in general. It asks for first aid instructions by relevant route of exposure, and it requires subdivisions — a visible heading per route — naming inhalation, skin, eye and ingestion as examples. OSHA’s Appendix D asks for the same subdivision in item (a), with a marginally different list: inhalation, skin and eye contact, and ingestion.

The insistence on splitting is not formatting pedantry. The right first response diverges sharply by route, and an undivided paragraph makes the reader work out which sentence applies to them while the clock runs. A splash in the eye and a swallowed mouthful of the same product are two different situations with two different answers, and the point of section 4 is that the supplier separated them in advance so nobody has to do it under pressure.

The word relevant carries weight in the EU list, which is introduced by “such as” — an illustrative list, not a closed one. A route the product cannot realistically take does not need an invented procedure. But point 0.4 of Annex II states plainly that the safety data sheet shall not contain blank subsections, so an inapplicable route has to be written out as inapplicable rather than left as white space.

The four things 4.1.2 has to settle

Subsection 4.1.2 is a list of decisions the sheet makes on the reader’s behalf, so that nobody has to make them in the moment. The Annex requires advice as to whether:

• (a) immediate medical attention is required, and whether delayed effects can be expected after exposure. Two findings, not one — an exposure can need no hospital in the first minute and still need observation afterwards, and the sheet has to say which case it is in.
• (b) movement of the exposed individual from the area to fresh air is recommended.
• (c) removal and handling of clothing and shoes from the individual is recommended. The Annex says removal and handling, because contaminated clothing carries the exposure with it and the question of what happens to it does not end at taking it off.
• (d) personal protective equipment is recommended for first aid responders.

That last one deserves its own note, because it is both the most-skipped item in the section and the one with no counterpart anywhere else on the sheet. The PPE specified in section 8 is chosen for an operator working under engineering controls in a planned task. A first aid responder arrives into an unplanned situation, often without controls, often with a contaminated surface or a contaminated person between them and the problem. Section 4 is the only place the regulation asks the supplier to say whether that person needs protection.

OSHA’s Appendix D names none of these four. That is the largest single gap between the two texts in this section: on the EU side, 4.1.2 is four explicit questions with four required answers; on the US side, item (a) asks for a description of the necessary measures and stops.

4.2 and 4.3 — the subsections addressed to other readers

4.2 — symptoms, briefly summarised

The wording of 4.2 is doing more work than it looks. Briefly summarised information on the most important symptoms and effects, both acute and delayed, from exposure. Both qualifiers are requirements, and they push in the same direction: 4.2 is a recognition aid, not a data dump.

It has two audiences. A colleague uses it to connect what they are seeing to the product — to move from “something is wrong” to “this is the chemical involved”. A clinician uses it as an opening hypothesis before the full record arrives. Neither is helped by a page of endpoints, which is why the evidence lives in section 11 and only the conclusion lives here.

The delayed half is the one that gets dropped. Where the classification already implies a lag — a respiratory sensitiser, a specific target organ effect after single exposure — a section 4 that reports only what happens in the first minute is inconsistent with the sheet it is printed in, and point 0.2.4 of Annex II forbids statements inconsistent with the classification.

4.3 — the subsection written for a clinician

Subsection 4.3, Indication of any immediate medical attention and special treatment needed, is where the technical content is allowed to be technical. Where appropriate, the sheet gives information on clinical testing and medical monitoring for delayed effects, specific details on antidotes where they are known, and contraindications. It closes with a requirement that has no US equivalent: for some substances or mixtures it may be important to emphasise that special means to provide specific and immediate treatment shall be available at the workplace — a prompt to the employer, not to the injured person.

Read 4.1 and 4.3 together and the architecture becomes obvious. The two subsections are addressed to different people on purpose. 4.1 stays inside what a bystander can do; 4.3 holds everything that requires a professional. A sheet that pushes clinical detail up into 4.1 has not added rigour — it has broken the standard the section opens with. Appendix D collapses all of this into one heading, item (c), and enumerates nothing.

Which version of section 4 you are looking at

The format usually called the new safety data sheet is the one set by Commission Regulation (EU) 2020/878, which applies from 1 January 2021; sheets that did not comply with it could be provided until 31 December 2022. On any current EU sheet, section 4 must therefore appear with the numbered subsections 4.1, 4.2 and 4.3. A US sheet written to Appendix D will more often print the three lettered subheadings instead, which is why the same section can look structurally different depending on which market the document was authored for.

How section 4 connects to the rest of the sheet

Section 4 is a conclusion drawn from other parts of the document, and it is checkable against them:

• The classification in section 2 decides which routes matter and which delayed effects have to be flagged. The response-type P-statements on the label are the same information compressed to a few words.
• The identities and concentrations in section 3 are what a poison centre or a treating physician will ask for first; section 4 is useless to them without it.
• Section 5 and section 6 handle the other two emergencies — the fire and the spill. Section 4 is about the person; keeping the three distinct is what stops each one becoming a summary of the others.
• Section 8 specifies protection for planned work. Subsection 4.1.2(d) specifies it for the unplanned arrival. The two answers are allowed to differ, and often should.
• Section 11 carries the evidence behind every symptom named in 4.2. If a symptom here has no basis there, one of the two is wrong.
• The emergency telephone number in section 1 is the escalation route that 4.1.2(a) points at when it says medical attention is required.

What a good section 4 looks like

Three tests, all answerable by reading the sheet:

1. Could an untrained person act on it in the first thirty seconds? No jargon, no equipment the site does not have, no instruction that requires a decision the reader is not qualified to make. This is the Annex’s own test, not an editorial preference.
2. Is every route either answered or explicitly accounted for? Four subdivisions, each carrying either a procedure or a stated reason it does not apply. No blanks, and no “not applicable” where the route plainly is.
3. Does it agree with sections 2 and 11? The symptoms in 4.2 should map onto the health hazard classes in the classification, and the delayed effects flagged in 4.1.2(a) should map onto what section 11 records. Where they diverge, the sheet has been assembled rather than written.