Patch Management Policy

OT patch management policy

Operational technology is patched during planned outages after testing. NIST SP 800-82 Rev. 3 says to test on a sandbox where possible and to deploy during scheduled maintenance windows; FDA's 2016 guidance expects manufacturers to tell customers within 30 days and fix within 60 days in its stated circumstances. Read 19 Nov 2026.

What NIST asks of an OT patch process

NIST SP 800-82 Rev. 3 (September 2023) says organizations should have “a systematic, accountable, and documented OT patch management process,” which covers how to monitor for patches, when to apply them, how to test them, “and how to select compensating controls to limit exposure of the vulnerable system when patching is delayed.” It warns that a patch may introduce “a greater risk from a production or safety perspective” and may change how the OS works with control applications, that many OT systems run unsupported operating systems with no patch, and that knowledgeable OT personnel decide when and how to deploy. It suggests a separate automated process for OT, and prioritizing components in DMZ environments and vulnerabilities that threaten availability, integrity or remote access (NIST SP 800-82 Rev. 3). It names NIST SP 800-40 Rev. 4, NERC CIP-007 and ISA 62443-2-3 as sources for procedures; the latter two were not read for this page.

Where the vulnerabilities come from

The same document notes that many OT vulnerabilities are published as CISA advisories but not all vendors report to CISA, so subscribe to vendor-specific notices as well. The monitoring step is in monitor for releases. Where no patch exists, the asset moves to the unpatchable group in systems that cannot reboot.

Medical devices: whose clock is it

FDA's Postmarket Management of Cybersecurity in Medical Devices guidance (issued 28 Dec 2016) is addressed to manufacturers, and it matters to a hospital policy because it tells you what to expect. For a vulnerability with uncontrolled risk, FDA does not intend to enforce part 806 reporting if there are no known serious adverse events or deaths, the manufacturer communicates with customers and identifies interim compensating controls as soon as possible and no later than 30 days after learning of it, and it distributes a validated fix as soon as possible and no later than 60 days after learning of it. Cybersecurity “routine updates and patches” are changes for controlled-risk vulnerabilities and are generally not reportable as corrections under 21 CFR part 806 (FDA guidance PDF).

Source: FDA guidance (December 2016; the PDF read carries a later note on the Quality Management System Regulation), read 19 Nov 2026. The guidance is nonbinding.
QuestionSource saysPolicy line
Who patches a regulated deviceThe manufacturer validates changes; FDA's guidance is addressed to manufacturersDo not patch outside the vendor's process; ask for the validated update
When is a customer toldWithin 30 days of the manufacturer learning of an uncontrolled-risk vulnerability, in the conditions statedRecord the notice date and the interim controls
When is a fix expectedWithin 60 days under the same conditionsTrack the date; use controls in the meantime

Window, testing and evidence

Set the window by the process owner's outage calendar, test offline or with the vendor, keep a recovery plan, and record the vendor's approval for any change. Evidence: the vendor notice and the date received, the test record, the outage ticket, and the compensating controls while patching is delayed. The record of such controls belongs in exceptions clause.

Sources

Last reviewed 19 Nov 2026