1. What we mean by cyber security in healthcare
Cyber security is the set of measures that protect systems and data from unauthorised access, improper alteration and disruption. In a healthcare facility this means three things together: confidentiality, because clinical data is among the most sensitive that exists; integrity, because an altered report or therapy can cause direct harm; and availability, because a medical record that cannot be consulted on a ward halts work and delays care. The GDPR treats health data as a special category and requires stricter protections.
For anyone running a practice or clinic, the starting point is not the software to buy but a question: which systems process patient data, who uses them, who manages them, and what happens if they stop for a day. Answering requires a simple map, even a shared spreadsheet. Without it, any technical choice risks protecting the wrong thing.
2. How it works in practice: systems, data and people
A healthcare facility usually has a few central systems and many peripheral points. The central systems include patient and appointment management, the clinical record, the diagnostic imaging archive, email and backups. The peripheral points include ward workstations, printers, tablets, networked medical devices and external supplier access. Each element has a different risk profile, so the measures also differ.
People matter as much as technology. Doctors, nurses, administrative staff and external collaborators use the same systems with different needs. The most effective measures are those everyone understands: access tied to role, multi-factor authentication (MFA, a second proof of identity in addition to the password), a clear way to report a suspicion and precise rules for anyone leaving the facility.
3. Where it applies: from the practice to the largest facility
The same principle applies to a general practice with two workstations and to a clinic with wards, a laboratory and operating theatres. What changes is the number of systems, the people with access and the complexity of the network. In a practice, simplicity matters most: well-configured access, verified backups, active updates and a supplier who responds. In a larger facility, formalised roles, access logs and an incident procedure are also needed.
Facilities that are not hospitals also process large amounts of health data: rehabilitation centres, analysis laboratories, private outpatient clinics, dental practices and diagnostic services. More than size, the type of service matters, because it determines what data is collected and how long it is kept. Verify your case with the supplier and, if needed, with a privacy adviser, before deciding what is proportionate.
4. The obligations: it depends on the perimeter
Two regulatory frameworks almost always affect healthcare facilities. The first is the GDPR, which governs the processing of personal data and imposes stricter rules for health data: it requires you to know which data is processed, why, for how long, who accesses it and which measures protect it. The second is the NIS2 Directive, transposed in Italy by D.Lgs. 138/2024, which requires IT risk management across several sectors, including healthcare, but only for entities that exceed certain size and activity thresholds.
The useful question is therefore not which rule applies in the abstract, but whether your facility falls under NIS2 and which personal data processing it carries out. The answer must be verified with an adviser or the competent authority, not with a generic interpretation. In the meantime, it is best to work on measures needed in any case, such as access control, backups and logs, because they support both the GDPR and NIS2.
5. Common myths to dispel
The first myth is that an antivirus is enough. An antivirus helps against some malware, but it does not replace access control, system updates, email protection and backups. The second is that the external supplier takes care of everything. The supplier manages the technology, but responsibility for the data stays with the facility, which must set the rules, verify them and be able to explain them. The third is that security gets in the way of clinical work. A good configuration avoids this, because emergency procedures are planned in advance rather than improvised.
A fourth myth concerns data residency. Where data is stored is a different question from who can lawfully access it: a system may physically be in one place and still be subject to the access rules of other countries. It is not enough, therefore, to ask where the data is; you must also ask who can access it, under what contractual guarantees and with what records. Treating the two questions separately avoids reassuring but incomplete answers.
6. A generic example: a practice with three workstations
Imagine a practice with a management system for appointments and records, three workstations, a shared printer and a supplier who supports it remotely. Nobody has ever defined who has access to what. The part-time collaborator uses the practice owner's account, the supplier has kept a remote access open for years and the backups go to a disk connected to the main computer, which is always on.
A reasonable first intervention does not need a large project. Create a personal account for each person, activate the second proof of identity on the management system, close the remote access and reopen it only on request, with a record kept. Then move a copy of the backups to a separate medium and test restoring one file. Each step is small, but it reduces the risk in a concrete and verifiable way.
7. The in-depth guides to read in order
This guide is the map. For details, first read the in-depth guides on the legal bases for health data and on GDPR measures, because they condition all other choices. Right after, the security checklist for practices and outpatient clinics is a practical starting point. Then the technical topics are covered: backups of clinical data, protection against ransomware and emergency access to data, that is, the break-glass procedure. Read in this order and apply the decisions at each step, because the later guides assume a map that is already up to date.
Other topics follow your case: the electronic health record if you feed data into the national system, connected medical devices if you have networked equipment, telemedicine if you conduct video consultations, and the rules on software as a medical device if you buy or develop such software. Read the ones that match your systems, not all of them together. Anyone who reads this guide before a meeting with the supplier arrives with more precise questions.
8. When the question becomes a decision
Security becomes a management decision when it requires spending, staff time or a change of supplier. The most common choices concern the service level to request, who responds in emergencies, how often restore tests are run and whether to manage in-house or outsource to a supplier. There is no single answer for everyone: it depends on the number of systems, the staff available and the type of data processed.
Before signing a contract, ask in writing what the supplier guarantees, what recovery times it declares, how it manages remote access, where and for how long it keeps logs and how it will notify you of an incident. Be wary of generic promises and compare the answers. A reliable supplier responds with verifiable details and clear references to responsibilities.
9. How to start
The useful first step this week is a list of all the systems that hold patient data or are needed to work with patients, with three columns: who uses it, who manages it, and what happens if it stops for a day. Set aside an hour with the administrative manager and the IT supplier, and write the answers next to each item, even when they are incomplete. If you cannot answer for an item, note who might know and by when you will ask.
With that list in hand, check the most important backup and review who has administrative access. If something does not add up, that is where to start. The rest of the guide and the linked in-depth guides will help you set priorities, but without an up-to-date map every measure remains a hypothesis that is hard to justify to management. The first meeting is for understanding the picture, not for signing a project.
Frequently Asked Questions
Does a small healthcare facility need to worry about cyber security?
Yes, because patient data is sensitive regardless of size. What changes is the depth of the measures, not the obligation to protect the data. In a small practice, simple controls done well are often enough: personal accounts, multi-factor authentication, verified backups and controlled remote access. Check with an adviser which legal obligations apply to you based on your activity.
Are the GDPR and NIS2 the same thing?
No. The GDPR concerns the protection of personal data and individuals' rights. NIS2 concerns IT risk management and the reporting of incidents for entities in certain sectors, including healthcare, when they exceed certain thresholds. Both may apply, and the same measures often help comply with both. Understanding which one applies to you requires an assessment of your specific case.
How long do I have to report a personal data breach?
The GDPR requires notification to the supervisory authority within 72 hours of the controller becoming aware of the breach, unless the breach is unlikely to result in a risk to individuals. For this reason a written procedure is needed that allows you to decide quickly. Whoever supports you must know whom to report a suspicion to, and the assessment must be documented even when no notification is made.
Can I hand all security over to my IT supplier?
You can delegate the technical management, but responsibility for the data and for decisions stays with the facility. The supplier should state in writing what it does, within what timeframes and with what access. You must know who does what, approve the main rules and keep evidence of the checks. A simple responsibility matrix prevents each side from assuming the other is handling a problem.
Is a cloud server in a European country enough to protect the data?
No. Where the data is stored is only part of the question. It also matters who can lawfully access it, what contractual guarantees are in place, and how logs and encryption keys are managed. Ask the supplier for a written answer on both aspects, and get an adviser to help you assess it.
Where do I start if I have little time?
Start with a map of the systems holding patient data, then check the backups and who has administrative access. These are three concrete actions that can be completed in a few weeks with the supplier and administrative staff. The checklist for practices and outpatient clinics offers a working order. The rest can follow, based on the risks identified in the map.