Risk Assessment vs Method Statement: How the Information Chain Works

Most contractors can name the difference between a risk assessment and a method statement. Far fewer can prove, line by line, that their method statement actually implements the controls their risk assessment selected — and that gap is the single most common reason a principal contractor (PC) rejects a RAMS. This page is about the information chain: how each assessed control must become a traceable step, what a reviewer looks for, and the specific disconnects that fail review.

A two-minute recap before we trace the chain

A risk assessment identifies hazards, evaluates likelihood and severity, and selects control measures — the analytical layer. A method statement converts those controls into a sequenced, task-specific safe system of work — the procedural layer. A RAMS is the industry-standard document that bundles the two; it is not a defined term in UK statute.

That is as far as we go on definitions here. If you need the full comparison — what each document is, the legal-duty versus contractual-requirement distinction, who produces what, and a decision matrix for which document a job requires — read our companion guide: RAMS vs risk assessment: what is the difference. That page owns the definitions and the "which document do I need" decision. This page owns what happens between the two documents once you know you need both.

---

The core principle: every control must be traceable to a step

The risk assessment and the method statement are not two separate documents that happen to be stapled together. They are two views of the same safe system of work:

• The risk assessment states what control is required and why (which hazard it addresses).
• The method statement states how and when that control is applied, by whom, and with what equipment.

The link between them is traceability. Every control measure recorded in the risk assessment must reappear in the method statement — not as the same sentence, but transformed into an instruction. "Collective fall protection" in the RA becomes "erect edge protection to all open edges before any operative steps onto the roof" in the MS. Same control, different layer.

A simple test, which is also the exact test a PC reviewer applies: pick any control measure from your RA. Can you point to the specific numbered step in your method statement that implements it? If you cannot, your RAMS has a gap, and a competent reviewer — or an HSE inspector after an incident — will find it.

---

What a PC reviewer actually checks: forward and backward traceability

Reviewers do not read the two documents in isolation. They cross-read them in both directions:

Direction	The question being asked	What a failure looks like
Forward (RA → MS)	Does every control I selected in the RA appear as an instruction in the MS?	A control listed in the RA (e.g. "exclusion zone below work area") never appears in the work sequence
Backward (MS → RA)	Does every step in the MS trace back to an assessed hazard and a selected control?	A method step introduces an activity or piece of plant that was never assessed — an unassessed hazard has crept in
Consistency	Is the equipment, competence, and emergency provision consistent across both layers?	The RA relies on a harness; the MS lists no harness, or a different fall-protection method, with no explanation

Forward traceability proves the controls you promised are actually enforceable on site. Backward traceability proves you have not silently introduced work — and therefore hazards — that nobody assessed. A RAMS can pass one direction and fail the other; both must hold.

---

The four disconnects that get a RAMS rejected

These are the specific failure modes reviewers see most often. Each is a break in the information chain, not a definitional error.

1. Orphan control — a control in the RA that never appears in the sequence

The RA selects a control ("crawling boards across fragile areas"), but the method statement's numbered steps never mention laying them. The control exists on paper but has no mechanism for being implemented on the day. This is the single most common rejection reason. A control that is not written into the sequence is a control nobody is instructed to apply.

2. Orphan step — a method step with no corresponding assessed hazard

The reverse fault. The MS includes a step — say, "use a cherry picker to lift materials to roof level" — but the RA never assessed mobile elevating work platform (MEWP) operation, overhead lines, or ground-bearing capacity. The step has introduced significant new hazards that were never evaluated. The RA must be revisited before the MS can stand.

3. Drift between layers — control named in RA, contradicted in MS

The RA relies on collective protection (edge barriers); the MS quietly substitutes "operatives to wear harnesses clipped to a running line." That may or may not be acceptable, but the unexplained substitution signals the two documents were written separately and never reconciled. A reviewer cannot tell which control is actually intended.

4. Sequence-blind controls — right control, wrong (or no) timing

The control appears in the MS, but at the wrong point in the sequence — for example, edge protection erected after the first roof access rather than before, or an asbestos check placed after work begins. For hazards where order is the control (isolate before work, confirm before access), getting the sequence wrong defeats the control even though it is present.

---

Worked example — traced control-by-control: fragile roof light replacement

Scenario: A maintenance contractor must replace six polycarbonate roof lights on an existing single-storey commercial warehouse. The roof is profiled metal sheet. The client has not confirmed whether earlier roof sheets contain asbestos cement. Operatives must access the roof surface.

Why this is high-risk: According to HSE guidance HSG33, roof work accounts for a substantial share of construction deaths, and falls through fragile materials such as roof lights and asbestos cement sheets are a leading single cause. Many of those killed on roofs are not trained roofers but maintenance workers accessing roofs (HSG33). Falls from height remain one of the biggest causes of UK workplace fatalities (HSE INDG401).

Below, each assessed control is traced forward into the exact method step that implements it. This is the table a reviewer reconstructs in their head — building it for them is what makes a RAMS pass first time.

#	Hazard (RA)	Control selected (RA)	Implementing step (MS) — where the control becomes an instruction
C1	Roof sheets of unknown composition — possible asbestos cement	Confirm asbestos status before any roof access; treat as ACM until cleared	Step 1: Site manager obtains and reviews the asbestos register before mobilisation. If absent or inconclusive, commission an intrusive survey; no roof access permitted until cleared (hold point)
C2	Fall from roof edge	Collective protection: edge protection before access	Step 2: Erect scaffold or proprietary edge protection to all open edges before any operative steps onto the roof surface
C3	Fall through fragile roof lights / fragile sheets	Stage loads across fragile areas; keep operatives off fragile material	Step 3: Lay crawling boards extending at least 600 mm beyond each roof light; operatives must not leave the boards. Step 5: remove and replace each roof light individually — never more than two open at once
C4	Persons struck by falling material below	Exclusion zone beneath the work area	Step 4: Ground worker establishes and maintains an exclusion zone directly below the work area; no persons admitted during roof work
C5	Dropped tools/materials during transfer	No manual carrying up ladders	Step 6: All equipment raised/lowered by rope-and-bucket or mechanical aid — no carrying up ladders
C6	Defective access equipment	Pre-use inspection of ladders/equipment	Step 7: User carries out a ladder pre-use check before each task and after any change (e.g. ladder dropped or moved), per HSE construction ladders guidance
C7	Suspended worker after a fall (backup PPE)	Rescue plan for fall-arrest	Emergency arrangements: designated rescuer identified; suspension-rescue procedure briefed before work starts; nearest A&E confirmed; first aider on site

Read the table left to right and you have forward traceability: every RA control (C1–C7) lands in a specific step. Read it right to left and you have backward traceability: every listed step exists to discharge an assessed control, with no orphan activity. That is a complete information chain. Notice that C1 and C2 are also sequence-critical — the asbestos hold point and the edge protection both have to come before access, or the control is defeated even though it is present (disconnect type 4 above).

---

Turning the chain into a check: the six-question traceability test

Run this before issuing any RAMS to a PC. Every "No" is a rework trigger — and maps to one of the four disconnects above.

#	Question	Disconnect it catches
1	Does every control measure selected in the RA translate into a specific, numbered step in the MS?	Orphan control (1)
2	Does every step in the MS trace back to a hazard the RA actually assessed?	Orphan step (2)
3	Is the equipment and fall-protection method in the MS the same as the RA relied on, with any substitution explained?	Drift between layers (3)
4	For sequence-critical controls (isolate-before-work, confirm-before-access), is the step in the correct position in the order?	Sequence-blind controls (4)
5	Do the MS emergency arrangements specifically address the highest-severity hazards the RA identified?	Orphan control (1)
6	Where the RA flagged a hazard needing specialist input (asbestos, live services), does the MS include a hold point that must be confirmed before work proceeds?	Orphan control (1) / Sequence (4)

---

Why the chain breaks: how documents drift apart in practice

The disconnects above are rarely deliberate. They come from a handful of working habits:

• Method statement written first, RA fitted afterwards. The RA must logically precede and feed the MS. When the MS is written from a generic template and the RA is back-filled at tender stage, controls and steps were never derived from one another — drift and orphans are almost guaranteed.
• Generic RA, specific MS (or the reverse). A RA written for "roof work" in general will not name the specific fragile materials, access route, or overhead services that the MS steps assume. The two layers describe different jobs.
• Static submission, changing site. A RAMS submitted once and never revisited breaks its own chain the moment the roof type differs, ACM is discovered, personnel change, or there is a near-miss. The RA review duty exists precisely to keep the chain intact.
• Copy-paste controls. Pasting a standard control block into the RA without writing the matching steps into the MS produces orphan controls at scale.

---

How RamsDocs keeps the information chain intact

RamsDocs is built around the chain described on this page. You start with the hazard-identification layer — task, location, people at risk, control hierarchy — and the platform then prompts you to translate each control you selected into a sequenced operational instruction, with fields for equipment, personnel, competencies, sequence position, and emergency arrangements. Because each method step is generated from a named control, the most common rejection cause — an orphan control that never reaches the sequence — is designed out, and the forward/backward traceability a PC reviewer checks is preserved by construction.

Important: All output from RamsDocs must be reviewed and adapted to your specific site, task, and contractor requirements by a competent person before use. No software-generated document removes the need for site-specific professional judgement.

---

Frequently asked questions

How should information flow between a risk assessment and a method statement? The risk assessment comes first and selects the control measures. The method statement then converts each of those controls into a specific, numbered step — same control, expressed as an instruction with who, when, and what equipment. Every control should be traceable to a step, and every step back to an assessed hazard.

What is the most common reason a RAMS is rejected? An orphan control: a control measure listed in the risk assessment that never appears as a step in the method statement. There is no mechanism to implement it on site, so the reviewer cannot rely on it. The reverse — a method step with no assessed hazard behind it — is the next most common.

What does a principal contractor's reviewer actually check? Traceability in both directions: that every RA control surfaces as an MS step (forward), and that every MS step traces back to an assessed hazard and control (backward), with consistent equipment, competence, and emergency provision across both layers.

How can I check my own RAMS before submitting it? Pick any control in your risk assessment and find the exact numbered step that implements it. Repeat for every control. Then read the method statement and confirm each step maps back to an assessed hazard. The six-question test above formalises this and links each failure to a specific disconnect.

Does combining the documents into a RAMS change the legal position? No. RAMS is an industry term for the combined document, not a statutory one. The underlying duties remain those of the individual documents. For the full legal-duty versus contractual-requirement breakdown, see the companion RAMS vs risk assessment guide.

---

This page is provided for general guidance only. All risk assessments and method statements must be reviewed and adapted to the specific site, task, and workforce by a competent person before use. Nothing on this page constitutes legal advice.