Safety case reports and missing as-built records: closing the evidence gap

Government guidance on safety case reports for higher-risk buildings asks for two things that sound simple: a description of how the building is constructed, including its structural elements and materials, and evidence of its structural condition, including the surveys that support it. For a recently built tower with a complete digital record, both are an exercise in assembly. For a large share of the social housing stock the regime actually covers, they are not.

This post is about that gap — why it exists, why it cannot be papered over, and how structural investigation closes it in a way the Building Safety Regulator can accept. If the regime is new to you, start with our plain-English guide to building safety cases.

Why the records are missing

Buildings over 18 metres or seven storeys have been going up in England since long before anyone was required to keep a golden thread of information about them. Much of the qualifying stock dates from the 1960s to the 1980s: system-built and large panel system blocks, in-situ concrete frames, and hybrid forms altered repeatedly over the decades. Stock transferred between councils and housing associations, sometimes more than once. Drawings were lost in office moves, mergers and archive clearances. What survives is often a design drawing, not a record of what was built — and the difference between the two is precisely where structural risk hides.

The result is that the building safety manager drafting the report is asked to describe a structure nobody fully documented. An honest report says so. But honesty about a gap is only the first step; the regulator’s question is what you are doing about it.

Assumption is not evidence

There is a temptation to fill the gap with typology: “buildings of this era and form are generally constructed as follows.” That has a place in framing an assessment, but a safety case built on it is fragile. The structural failure risk the report must address lives in the specifics — the actual reinforcement in the actual panels, the actual bearing of the actual floor units, the actual condition of forty-year-old concrete.

Government has been clear that landlords can draw on existing surveys rather than starting from scratch, and that is worth doing. But the existing material has to be current, relevant, and reviewed by someone competent — a stock condition survey carried out for asset planning in 2011 is not a structural appraisal, and presenting it as one is the kind of thing that gets a report sent back.

The engineering profession has held this line since long before the Act: an appraisal of an existing structure is only as good as the information it rests on, and where records cannot be relied upon, testing is how the information is obtained. The safety case regime turns that professional norm into a regulatory one.

What investigation actually provides

The useful thing about the evidence gap is that it is closable, and closable without pulling the building apart around its residents. A structured programme of non-destructive investigation can establish most of what a safety case report needs to say about the structure:

• Reinforcement detail. Ferro scanning and GPR locate reinforcement, confirm spacing and arrangement, and measure cover depth — turning an assumed detail into a measured one, non-destructively and floor by floor if needed.
• Element geometry. Slab and panel thickness, wall construction and hidden voids can be established from one face, which matters in occupied buildings where access to both sides is rarely possible.
• Material condition. Concrete condition surveys — carbonation, chloride, half-cell potential — establish whether the material still has the durability the original design assumed, and where reinforcement corrosion is likely to be active.
• In-situ concrete strength. Pull-out testing to BS EN 12504-3, alongside rebound hammer and where necessary core sampling, gives the engineer a defensible in-situ strength value to assess against. Where the question is instead whether a specific anchor or fixing will hold — cladding restraint, plant, safety-critical attachments — proof-load testing to BS 8539 is the relevant test, and it is a different exercise.
• Geometry and record drawings. Where no reliable drawings exist at all, a measured as-built survey creates the baseline record the golden thread is supposed to contain.

Each result does double duty. It feeds the engineer’s structural assessment, and it becomes a document in the safety case — dated, methodical, attributable evidence of the building’s real condition.

Proportionate, not exhaustive

Nobody scans every square metre of a 20-storey block, and the regulator does not expect it. The investigation should be targeted by the engineer’s assessment: which elements are safety-critical, which assumptions matter most if wrong, where the typology is least trustworthy. A sensible programme samples representative elements, confirms or corrects the assumed construction, and escalates only where the findings warrant it.

That is also the answer to the cost question, which for a social landlord with a dozen blocks in scope is not a small one. Investigation scoped this way is a modest line against the alternatives: a report the regulator rejects and you resubmit, or remediation designed against a structure nobody actually understood.

Practical notes for commissioning

Three things make these programmes go well. First, involve the structural engineer in scoping — the survey exists to answer their questions, and a well-written brief starts there. Second, plan for occupation: this work happens in residents’ homes and common parts, and non-destructive methods, sensible hours, proper notice and clear communication are what keep it from becoming a resident engagement problem — which the safety case also has to account for. Third, specify the deliverable as safety case evidence from the outset: findings referenced to location, method and date, in a format that drops into the report and the golden thread rather than a PDF that summarises vaguely.

The safety case regime asks landlords to prove what their buildings are. For buildings whose paperwork cannot prove anything, the structure itself is the record — and investigation is how you read it.