ISO 14971 risk estimation and evaluation process, risk acceptability criteria, and risk management file linking risks to controls.
CVSS-based security risk estimation distinct from safety probability/severity scales, acceptability threshold documented in the security risk management plan, with findings fed back to the ISO 14971 file.
Security risk assessment with CVSS scores and documented acceptability thresholds — applying safety probability scales to cybersecurity is a critical FDA and Notified Body finding.
Maps to
IEC 81001-5-1: §7.3 Estimation and evaluation of SECURITY risk
ISO 14971: §5.4 Identification of hazards and hazardous situations
Requirement text
Estimate the risk of the vulnerabilities identified, considering the adverse impact of that vulnerability to cybersecurity. This estimation can be supported by using vulnerability scoring, such as CVSS or the MITRE scoring rubric for medical devices. The scoring system can also be based on a likelihood/severity scheme used by the manufacturer for other risks (see e.g. ISO/IEC Guide 51 or ISO 14971). Evaluate estimated risks and, based on scoring, determine if the risk is acceptable or not. Inform the product risk management process about updates to the threat model.
Why this clause exists
Security risk estimation requires a framework different from ISO 14971 safety risk estimation because the key parameter in security risk — exploitability — is adversarially controlled rather than probabilistic. Safety risk estimation treats probability of harm occurrence as a property of the device and environment; security risk estimation must treat exploitability as a property of adversarial capability and access opportunity. FDA has explicitly rejected the use of ISO 14971-style probability scales for cybersecurity risk, noting that threat actors can increase their effort when attack payoffs are high. CVSS was developed specifically to provide a standardized, reproducible scoring methodology that captures exploitability factors (attack vector, attack complexity, privileges required, user interaction) independently from impact factors (confidentiality, integrity, availability impact), allowing organizations to estimate security risk in a way that regulators and operators can independently verify and compare. IEC 81001-5-1:2021 clause 7.3 requires security risk estimation and evaluation using such a methodology, and CVSS v3.1 represents the current industry standard for satisfying this requirement in the medical device context.
What changed
IEC 81001-5-1:2021 is the first standalone cybersecurity standard purpose-built for health software and medical device software. Published in December 2021, it was adapted from IEC 62443-4-1 (industrial control systems security) to address the unique safety and regulatory context of medical devices — adding health-specific requirements that account for patient safety, clinical workflows, and the manufacturer-HDO relationship.
The standard mirrors IEC 62304's lifecycle structure but adds security-specific activities at every phase — planning, development, testing, release, and maintenance. It requires security risk management to be integrated with ISO 14971 safety risk management, not treated as a separate IT concern. FDA recognized it as Consensus Standard 13-122 (date of entry December 19, 2022). FDA's February 2026 cybersecurity guidance, Section V, lists IEC 81001-5-1 among the "possible frameworks to consider" for secure product development, and FDA's recognition record for the standard notes that conformance "may not satisfy all the cybersecurity requirements outlined in Section 524B of FD&C Act".
EU MDR harmonization was originally targeted for May 2024 but postponed to May 2028.
Common gaps (what we see in audits)
- Risk estimation uses safety-derived scales inappropriate for cybersecurity — Cybersecurity risks are estimated using probability/severity scales designed for safety risks under ISO 14971. FDA rejects 'probability of occurrence' for cybersecurity because security exploits are not random events. Risk must be assessed based on exploitability, attacker capability, and attack complexity using frameworks like CVSS or SSVC.