What clinical risk is and why managing it is mandatory
In the literature on safety of care, clinical risk is the probability that a patient will be the victim of an adverse event, that is, will suffer harm or discomfort — even if only potential — attributable to the healthcare received rather than to the underlying disease. Clinical risk management is the set of methods, roles and tools through which a healthcare organisation systematically identifies, analyses, treats and monitors these risks, with the aim of reducing the probability and severity of harmful events.
The basic taxonomy distinguishes three categories of event, in increasing order of severity:
- Near miss: an error that could have caused harm but did not, because it was intercepted in time or owing to fortunate circumstances. It is the most valuable source of information: it reveals the weaknesses of the system without the human cost of the event;
- Adverse event: harm that has actually occurred to the patient, attributable to care. It may be preventable or non-preventable;
- Sentinel event: a particularly serious adverse event — potentially indicative of a serious system malfunction — that requires an immediate investigation and activation of the measures set out in the Ministerial Recommendations.
In Italy clinical risk management is not an optional good practice: it is a mandatory function. Law no. 208 of 28 December 2015 (art. 1, paragraphs 538-540) required all organisations, public and private, to set up a function for monitoring, preventing and managing healthcare risk. Law no. 24 of 8 March 2017 (Gelli-Bianco) then raised the principle to system level: art. 1 defines safety of care as a constituent part of the right to health, art. 2 establishes the Regional Centres for healthcare risk management and art. 3 the National Observatory of good practices on healthcare safety at AGENAS. With the implementing decrees — in particular Ministerial Decree (D.M.) 232/2023 on insurance — the law has been fully in force since 16 March 2026: the obligations are no longer a future prospect, they are enforceable.
There is, however, a deeper reason than the legal obligation: clinical risk management works. A mature system reduces preventable events, contains litigation, improves the internal climate and makes the organisation more credible to patients, insurers and supervisory bodies. The sections that follow describe how to build it, piece by piece.
Governance: risk manager, committee, Regional Centres
The first organisational decision concerns who does what. The legislation does not impose a single model, but established practice in Italian healthcare organisations is structured on three levels.
The first level is the risk management function, led by the clinical risk manager: a formally appointed figure, with a written mandate from the directorate, dedicated resources and direct access to top management. The risk manager coordinates the collection and analysis of reports, conducts or facilitates event analyses, drafts the annual plan and the year-end report, and oversees training. This may be an internal figure or one supported by specialist consultancy: what matters is that the function exists, is recognisable and produces evidence.
The second level is collegial: the patient safety committee (or risk management unit), which brings together the medical directorate, risk management, quality, healthcare professions, pharmacy and — where relevant — the legal office and the insurance area. This is where priorities are validated, cross-cutting corrective actions are approved and claims management is integrated with prevention, often through the Claims Assessment Committee (Comitato Valutazione Sinistri). In many organisations the committee also involves the DPO: adverse events and reports contain health data, and data protection is part of the system's design, not an afterthought.
The third level is external to the organisation: the Regional Centres for healthcare risk management and patient safety (art. 2 of the Gelli-Bianco Law), which collect data from the organisations in their region and pass them on to the National Observatory of good practices at AGENAS (art. 3). This cascading flow has a precise operational consequence: the data the organisation produces internally must be structured, classified and aggregable, because sooner or later someone will ask for them in that format.
A final element of governance, often underestimated, is the protection of internal analysis: art. 16 of the Gelli-Bianco Law states that the minutes and documents resulting from clinical risk management activities cannot be acquired or used in legal proceedings. This is the guarantee that makes a non-punitive reporting culture possible: those who analyse an event in order to learn must be able to do so without that work becoming evidence for the prosecution.
Reactive methodologies: learning from what has happened
Reactive methodologies start from an event that has already occurred — or nearly occurred — and reconstruct its causes to prevent it from happening again. They are the system's learning engine.
Incident reporting: the system's front door
Incident reporting is the structured flow through which staff report near misses, adverse events and sentinel events. It is the primary source of knowledge about the organisation's real risk, and its quality depends on three conditions: reporting must be easy (a few minutes, from any workstation), safe (guarantees of non-punishment and confidentiality) and useful (those who report must see that reports lead to analysis and action, otherwise they stop). A rising reporting rate in the early years does not indicate a more dangerous organisation: it indicates an organisation that has started to see. We have devoted an in-depth article to the topic: Incident reporting in healthcare: how to build a reporting system that works.
Alongside the internal flow there is the ministerial SIMES flow (Ministerial Decree of 11 December 2009), which collects sentinel events and claims notifications at national level: the most serious events must therefore be handled on a dual track, internal (analysis and actions) and external (reporting obligations).
Root Cause Analysis (RCA)
RCA is the systematic analysis of the root causes of a significant adverse event or a sentinel event. The guiding principle is shifting the focus from the person to the system: not "who made the mistake", but "which organisational conditions made the error possible or allowed it to go unintercepted". The typical process involves reconstructing the timeline of the event, identifying contributing factors (communication, training, technology, workload, environment, procedures) with techniques such as the "five whys" or the cause-and-effect diagram, and defining corrective actions with owners and deadlines. An RCA that ends with "remind the operator to follow the procedure" is a failed RCA: effective actions change the system, not the goodwill of individuals.
Clinical audit
Clinical audit compares actual practice with explicit standards — SNLG guidelines (the Italian National Guidelines System, art. 5 of the Gelli-Bianco Law), Ministerial Recommendations, internal procedures — on a defined topic: antibiotic prophylaxis, documentation management, patient identification. The cycle is always the same: define criteria and standards, measure practice, analyse deviations, introduce corrective measures, measure again (re-audit). The protection of art. 16 also applies to audit: the minutes of risk management activities cannot be used in court, precisely to allow frank professional discussion.
Mortality & Morbidity review
The M&M review is the structured, periodic meeting in which a team discusses cases of death or complications selected for their learning potential. Compared with RCA it is lighter and more frequent: it does not require a sentinel event to be triggered and keeps the habit of critical peer review alive. For it to work, clear rules are needed: cases chosen in advance, discussion without disciplinary purposes, conclusions recorded and followed up at subsequent meetings.
Claims analysis
Compensation claims, complaints and litigation are, to all intents and purposes, a source of information on risk: they tell the story of events that reached the patient and their legal protections. Systematic claims analysis — by area, type, outcome — feeds the priorities of the annual plan and closes the loop between the insurance and legal dimension and the preventive one. The natural forum for this integration is the Claims Assessment Committee, where one has been set up.
Proactive methodologies: anticipating error
Proactive methodologies do not wait for the event: they analyse processes to identify where an error could occur and prevent it. A mature system balances the two families: purely reactive, it is always chasing; purely proactive, it loses touch with operational reality.
FMEA / FMECA: failure mode analysis
FMEA (Failure Mode and Effects Analysis) and its quantitative variant FMECA (which adds criticality analysis) break a process down into its steps and, for each step, ask: what can go wrong (failure mode), with what effects, and for what causes? Each failure mode is then assessed by a multidisciplinary group on three dimensions — severity of the effect, probability of occurrence and detectability of the error before it reaches the patient — typically on numerical scales whose product generates a risk priority number (RPN, in Italian IPR). The RPN ranks failure modes and focuses countermeasures where risk is highest and least visible.
A classic application is the medication management process on the ward. The working group breaks the process down into prescribing, preparation, administration and monitoring; for each step it identifies the failure modes — mix-ups between LASA (look-alike, sound-alike) medicines, dosing errors during preparation, a missed administration, incorrect patient identification — and assesses them for severity, probability and detectability. A LASA mix-up at administration, for example, will typically emerge as critical because it is serious and hard to detect once it has occurred: the resulting actions will act on the system (physical separation and labelling of LASA medicines, independent double checks for high-risk medicines, active verification of patient identity before administration), not on individual attention. This is precisely the strength of FMEA: it makes explicit and shared a risk that each professional knew only in part.
Safety Walkaround
The Safety Walkaround takes the directorate physically onto the wards to talk about safety with frontline staff: scheduled visits, simple questions — what situations worry you, what nearly made you make a mistake recently, what would you need — and commitments that are tracked. The value is twofold: risks emerge that no reporting form captures, and staff see that the directorate treats safety as its own issue, not something delegated. The condition for credibility is follow-up: a walkaround whose commitments produce nothing erodes trust more than the visit built it.
Surgical safety checklist
The surgical safety checklist, promoted by the WHO and adopted in Italy by the Ministry of Health, is probably the best-known proactive tool: a structured check at three moments — before induction of anaesthesia (sign in), before skin incision (time out), before the patient leaves the operating theatre (sign out) — covering patient identity, site and procedure, allergies, availability of blood components, and swab and instrument counts. Its effectiveness depends on genuine execution, not on ticking boxes: a time out carried out "out loud, with everyone stopping" is a barrier; a checklist filled in at the end of the operation is just a form. Monitoring adherence to the checklist is, in turn, a typical indicator in annual risk management plans.
The PARM: the plan that holds the system together
The Annual Risk Management Plan (PARM, Piano Annuale di Risk Management) is the planning document that turns the methodologies described so far into a work programme: it analyses the context and the previous year's data (reports, claims, audits, indicators), defines objectives and priorities, assigns responsibilities and deadlines, plans training and establishes how progress will be measured. At year end, the annual report accounts for what has been done and what the data show.
The PARM is also where the risk management function stakes its credibility with the directorate and the supervisory bodies: a generic plan, photocopied year after year, falls apart at the first comparison with the organisation's real data; a plan built on internal evidence — which events actually occurred, which areas reported most, which of the previous year's actions were closed — is documentary proof that the mandatory function required by Law 208/2015 does not exist only on the organisation chart. We have devoted a specific guide to building the PARM, with a standard structure and annual cycle: PARM: how to build the Annual Risk Management Plan.
Professional liability and insurance after the Gelli-Bianco Law
Clinical risk management operates within a framework of legal liability that the Gelli-Bianco Law has redesigned. In a nutshell: the organisation is liable to the patient in contract, while the professional — when working within the organisation without a direct contractual obligation to the patient — is liable in tort. This dual track shifts the centre of gravity of litigation onto the organisation and makes the quality of its risk management system a concrete element of defence: the ability to document procedures, event analyses and corrective actions carries weight in reconstructing what the organisation had done to prevent the harm.
On the insurance front, art. 10 of the law requires healthcare and social care organisations to have mandatory insurance cover — or "equivalent measures" of direct risk retention — for civil liability towards third parties and towards those providing services. With the implementing decree D.M. 232/2023 and the law becoming fully in force from 16 March 2026, these obligations have become fully enforceable, and with them the need to demonstrate — to insurers, supervisory bodies and counterparties — an effective and documented risk management framework. It is no coincidence that the quality of the risk management system increasingly also affects the organisation's insurability terms. For a full analysis of the operational impact of the deadline, see our dedicated article: Gelli-Bianco Law fully in force from 16 March 2026: what changes for healthcare organisations.
Indicators and measurement of clinical risk
What is not measured cannot be governed: without indicators, risk management remains a set of activities without direction. A well-built dashboard combines three families of measures.
- Process indicators: they measure how the system itself is working — reporting rate by area and event category, time to take charge and to close analyses, percentage of corrective actions completed by the deadline, adherence to the surgical safety checklist, training coverage;
- Outcome indicators: they measure what happens to patients — the classic type is the event rate relative to activity, such as adverse events or falls per 1,000 patient days, or healthcare-associated infections in at-risk patients. The ratio to the activity denominator is essential: absolute values do not allow comparisons between periods or between wards with different volumes;
- Culture indicators: periodic surveys of the safety climate as perceived by staff, which anticipate signals that the other indicators will only show later.
Two methodological caveats. First: indicators must be read as trends and in context, not as absolute values — an increase in reports after a training campaign is a success for the system, not a deterioration in safety. Second: every indicator must have an owner, a defined data source and a review frequency, typically linked to the cycle of the PARM and the annual report. An indicator that nobody looks at is just a number.
The role of technology: why Excel spreadsheets are not enough
Many organisations still manage reports, analyses and plans on paper forms and spreadsheets. As long as the system is young and volumes are low, this seems to work. But an Excel spreadsheet has structural limits that emerge precisely when the system starts to get serious:
- No workflow: taking charge of a report, assigning the analysis and chasing overdue actions depend on people's memory, not on the system;
- No traceability: who changed what, and when? In a context where the ability to demonstrate is part of compliance, the absence of an audit trail is a vulnerability;
- Manual aggregation: every annual report, every data flow to the Regional Centre, every request from the directorate becomes a hand-crafted copy-and-paste job, with errors and diverging versions;
- Fragile confidentiality: reports contain health data; files shared on the network or sent by e-mail do not offer the access control that the GDPR and common sense require;
- No scalability: when reports grow — which is the goal — the manual system collapses at the very moment of success.
When choosing risk management software, the criteria that matter are coverage of the complete cycle (from report to corrective action to reporting), consistency of taxonomies with ministerial and regional flows, configurability of workflows to the real organisation, usability for the occasional reporter, security and GDPR compliance, and the ability to automatically produce the evidence that the PARM and the annual report require. We have gathered criteria, questions to ask suppliers and pitfalls to avoid in a dedicated guide: the buyer's guide to healthcare risk management software.
One final, increasingly topical consideration: clinical risk and cyber risk are converging. Healthcare organisations fall within the scope of NIS2, and a cyber incident that brings down clinical systems is, to all intents and purposes, also a patient safety risk — with notification obligations to ACN (the Italian National Cybersecurity Agency) set within tight windows. The two disciplines share the same grammar: timely reporting, root cause analysis, corrective actions, documentary evidence. For the cyber side, see NIS2 incident notification: the 24h / 72h / 1-month procedure to ACN.
EasyRisk: the platform for clinical risk management
EasyRisk is Easy Health's platform that digitises the clinical risk management cycle described in this guide: the risk register classified by area, process, probability and impact; structured incident reporting of near misses, adverse events and sentinel events, with a fully tracked workflow for taking charge, analysis and closure; monitoring of ministerial priority areas; the PARM as a dynamic tool with responsibilities and progress; the annual report generated by aggregating the data collected; integration between complaints, patient feedback and claims management. The aim is not to replace the organisation's organisational choices, but to make them measurable, traceable and demonstrable — which is exactly what the legislation, now fully in force, requires.
Clinical risk FAQs
What is the difference between clinical risk and safety of care?
They are two sides of the same issue: clinical risk is the probability of harm, safety of care is the condition one seeks to guarantee by reducing that risk. The Gelli-Bianco Law (art. 1) uses the second formulation and defines it as a constituent part of the right to health: clinical risk management is the operational tool through which safety of care is achieved.
Is the risk management function also mandatory for private and outpatient facilities?
Yes. Law 208/2015 (art. 1, paragraphs 538-540) requires the function for monitoring, preventing and managing healthcare risk to be set up in all organisations, public and private. The scale of the tools must be proportionate to the complexity of the organisation, but the principle — an identified function, event reporting, documented analysis and actions — applies to everyone.
Is it better to start with reactive or proactive methodologies?
With reactive ones, almost always: a functioning incident reporting system is the prerequisite for everything else, because it generates the data on which credible priorities can be built. Proactive methodologies (FMEA, walkarounds, checklists) fit well into a system that has already learnt to collect and analyse events — and the PARM is the tool for balancing the two families year by year.
Can internal event analyses be used against the organisation in court?
Art. 16 of the Gelli-Bianco Law states that the minutes and documents resulting from clinical risk management activities cannot be acquired or used in legal proceedings. This protection is designed precisely to allow frank analysis and a non-punitive reporting culture. The ordinary rules on the patient's clinical documentation, of course, still apply.
Want to structure clinical risk management in your organisation?
The Easy Health team supports medical directorates and risk managers in building the risk management system: methodologies, annual plan and software platform to make everything traceable and demonstrable.
Discover the EasyRisk platform or contact us to arrange a dedicated call.