What is a clinical safety case?

In short: A clinical safety case is a structured argument — backed by evidence — that a health IT system is acceptably safe for its intended use in its intended setting. It is the central deliverable of DCB0129, owned by a named Clinical Safety Officer, and it travels with the product into deployment.

Where the requirement comes from

The clinical safety case is not a voluntary best practice — it is the central output of the UK health system clinical risk management standards, DCB0129 and DCB0160, published under section 250 of the Health and Social Care Act 2012. DCB0129 places the duty on the manufacturer of a health IT system; DCB0160 places a parallel duty on the organisation deploying it. Both require a documented clinical risk management system, a named Clinical Safety Officer, and a safety case that is assessed during procurement — increasingly through the Digital Technology Assessment Criteria (DTAC).

What it contains

In practice the safety case is a small family of linked artefacts rather than a single file. The two that reviewers ask for by name are the Clinical Risk Management Plan (how risk will be managed across the lifecycle) and the Clinical Safety Case Report, which presents the argument and points to the evidence. Underpinning both is the hazard log. Together they cover:

• System & intended use — what the product does, for whom, and where; the scope statement everything else is judged against.
• Risk management approach — the Clinical Risk Management Plan: how clinical risk is identified, scored, controlled and reviewed.
• Hazard log — each hazard, its causes, clinical effect, controls and residual risk, with an initial and a post-control score.
• Safety argument — the Clinical Safety Case Report: a clear, evidenced argument for why the residual risk is acceptable.
• CSO sign-off — accountability by a named, suitably qualified and experienced Clinical Safety Officer (a registered clinician).

Manufacturer or deployer? Two safety cases, not one

A frequent source of confusion is who has to produce what. The manufacturer's DCB0129 safety case argues that the product is acceptably safe for its stated intended use. The deploying organisation's DCB0160 safety case argues that the product is acceptably safe as it is actually configured, integrated and used locally — with its real interfaces, its real users and its real patient population. The deployer's work builds on the manufacturer's evidence; it does not replace it. Suppliers that hand over a clean, current DCB0129 safety case make their customers' DCB0160 work far lighter, which is a genuine competitive advantage in UK health system sales.

Inside the hazard log

The hazard log is where assessors spend most of their time, because it shows whether risk has actually been thought through. Each hazard is described in clinical terms — not "the API times out", but "the medication list fails to load and a clinician prescribes without seeing an allergy". Causes, the potential clinical effect, and the controls that reduce it are recorded, and risk is scored on a severity-by-likelihood matrix both before and after controls. A credible log shows residual risk driven down to an acceptable level by specific, testable controls — some technical, some procedural (training, alerts, fallback workflows) — rather than asserted to be low.

Why safety cases fail review

Most safety cases that stall in procurement fail for the same handful of reasons: the intended-use statement is vague, so the scope of the argument is unclear; the hazard log lists generic IT failures instead of hazards expressed as clinical harm; controls are claimed without evidence they work; the document is stale, signed off against a version of the product that has since changed; or there is no genuinely independent, named Clinical Safety Officer standing behind it. Each of these is avoidable, and each is something a buyer's clinical-safety team will probe directly.

Why it matters commercially

For health-system buyers, a clear, current clinical safety case is a precondition, not a nice-to-have. It is assessed within DTAC and feeds the deploying organisation's DCB0160 work. A weak or stale safety case is one of the most common reasons adoption stalls.

Keeping it alive

Treat the safety case as a living artefact: update it when the product, intended use or context changes, and review it on a schedule. Vendors that hand deployers a clean, current safety case make procurement dramatically smoother.