Process Validation × Continued Process Verification
Stage 2 proves a process can perform; Stage 3 proves it still does. The overlap is the handoff between 'validated' and 'stays validated' — and the difference between a process in specification and a process in control.
What this page does not claim
An intersection covers what happens only where two axes overlap. It does not restate what either parent page says, and it is not a substitute for reading them.
WHAT ONLY EXISTS IN THE OVERLAP
- Process qualification is a snapshot; continued process verification is a moving claim. The overlap is the handoff between them — the point where the process knowledge built in design has to define exactly what ongoing monitoring watches, or Stage 3 measures the wrong things forever.
- A process can pass every release specification while drifting out of statistical control. Continued verification exists to catch the drift that specification limits cannot see, which makes process-monitoring capability — the ability to tell in-spec from in-control — the load-bearing skill of the intersection.
- FDA's Stage 3 continued process verification and EU GMP Annex 15's ongoing process verification are the same duty written by two authorities. A global site runs one programme that answers both, and the overlap is where that alignment is designed rather than duplicated.
- Monitoring that never triggers action is not verification. The signals continued verification produces have to feed ICH Q9 risk re-assessment and ICH Q10 management review and continual improvement, or the process is being watched without ever being verified.
- Continued verification changes what the defensible record must contain: not a periodic re-qualification, but a connected chain from monitored signal to assessment to action to outcome — which is a different evidentiary shape than a validation report.
Where 'validated' becomes 'stays validated'
FDA's 2011 process validation guidance recast validation as a lifecycle in three stages — process design, process qualification, and continued process verification — and the third stage is where this intersection lives. The trap is to read Stage 3 as a lighter repeat of Stage 2: run the qualification again, periodically, at reduced intensity. It is not that. Process performance qualification is a snapshot taken under heightened scrutiny to demonstrate the process can reproducibly deliver quality; continued process verification is the ongoing demonstration that it actually does, batch after batch, under routine conditions. The first answers 'can this process perform'; the second answers 'is this process still performing', and the honest transition between those two questions is the work the overlap owns.
The handoff is more than sequential — it is a transfer of knowledge with strict content. What Stage 3 must monitor is not decided in Stage 3; it is inherited from the process understanding built in Stage 1, where the critical quality attributes and the critical process parameters that drive them were identified, and from the control strategy that qualification confirmed. A continued-verification programme that monitors whatever is convenient to measure, rather than what the design stage identified as consequential, has severed the handoff: it collects data diligently about parameters that do not matter and stays blind to the ones that do. The mark of a sound intersection is a Stage 3 monitoring set that traces directly back to Stage 1's determination of what makes this product's quality vary.
In-spec is not in-control
The distinction that gives continued process verification its whole reason to exist is one that release testing cannot make: the difference between a process that is in specification and a process that is in a state of control. Specification limits are the boundary of acceptable product; statistical control describes whether the process is behaving consistently with its own history. A process can produce batch after batch that passes every specification while its centre drifts steadily toward a limit, or while its variability widens — trends that are invisible to release testing because every individual result is still acceptable, and lethal to detect late because by the time a result finally fails, the process has been deteriorating for a long time. Continued verification is the mechanism that sees this, and it sees it only through process-monitoring capability applied to trends rather than to pass/fail dispositions.
This is why the intersection is carried by a monitoring competence, not by a document. Reading a process for control means trending the right parameters over time, distinguishing common-cause variation from special-cause signals, and acting on a drift while every batch is still passing. An organisation that equates continued verification with confirming that results stayed inside specification has kept the form and lost the function — it will report a process under control right up to the batch that fails, and the failure will look sudden only because nobody was reading the trend that had been announcing it. The capability to tell in-spec from in-control is what separates a continued-verification programme that prevents problems from one that merely documents them after release.
One duty, two regulators' words
The same obligation appears under two names. FDA's lifecycle guidance calls the third stage continued process verification; EU GMP Annex 15 requires ongoing process verification, and the pharmaceutical quality system expectations of ICH Q10 frame continual monitoring and improvement as a standing responsibility of the quality system rather than a validation appendix. A manufacturer supplying globally does not run three programmes; it runs one, designed so that a single monitoring and trending system satisfies each authority's expression of the same underlying duty — that a process demonstrated capable at qualification is shown to remain in a state of control across its commercial life.
The overlap is where that single programme is deliberately designed rather than accidentally duplicated. The alignment is not merely administrative convenience: reading the FDA lifecycle, the EU ongoing-verification requirement, and the Q10 quality-system frame together clarifies what the programme is for, because each contributes an emphasis the others assume. The lifecycle guidance supplies the staged logic and the tie back to process design; Annex 15 anchors it in the qualification-and-validation obligations of the EU quality system; ICH Q10 places continual monitoring inside the machinery of management review and improvement. A site that treats these as separate compliance boxes ends up with parallel, partly redundant records that agree on the easy cases and diverge on the hard ones; a site that treats them as one duty builds a single continued-verification system whose outputs are legible to any of the three inspectorates.
The loop back to risk and improvement
Continued verification that only produces charts is monitoring, not verification. The word verification implies a judgement returned and, where warranted, an action taken — and that is where the intersection connects to quality risk management and the pharmaceutical quality system. A signal from the monitoring programme — a parameter trending toward a limit, variability widening, a run of results on one side of the centre — is an input to ICH Q9 risk re-assessment: does this change the risk picture the control strategy was built on, and does the control strategy need to change in response. A continued-verification programme wired to this loop uses its signals to revisit the assumptions of the original validation; one that is not wired to it accumulates trend data that no decision ever consumes.
The same signals are inputs to the ICH Q10 management-review and continual-improvement machinery, which is what turns continued verification from a defensive record into a source of process knowledge. This also changes the evidentiary shape of what a manufacturer must be able to show. Under a periodic-requalification mentality, the defensible artefact was a re-qualification report at intervals. Under continued verification, the defensible artefact is a connected chain — the monitored signal, the assessment it prompted, the action taken, and the outcome — demonstrating that the process was not merely watched but verified, and that when it moved, the quality system responded. An inspector at this intersection asks not for the trend charts alone but for an instance where a trend triggered an assessment and an action, because that chain is the proof the loop is closed. A programme that can show the charts and not the chain has documented monitoring and skipped the verification.
Derived from the 4 standards SPEQ maps to this intersection, across 3 regulatory bodies: FDA, EMA, ICH.
FREQUENTLY ASKED
Is continued process verification just a periodic repeat of process qualification?
No — reading Stage 3 as a lighter, periodic re-run of Stage 2 is the classic misread. Process performance qualification is a snapshot taken under heightened scrutiny to show the process can reproducibly deliver quality; continued process verification is the ongoing demonstration that it actually does, batch after batch, under routine conditions. The first answers whether the process can perform, the second whether it is still performing. They also differ in content: what Stage 3 monitors is inherited from the process understanding built in Stage 1 — the critical quality attributes and the parameters that drive them — not chosen afresh. A programme that monitors whatever is convenient rather than what design identified as consequential has broken the handoff between the two stages.
What does 'in specification is not in control' mean for continued verification?
It means release testing and process control answer different questions, and continued verification exists to answer the one release testing cannot. Specification limits mark the boundary of acceptable product; statistical control describes whether the process is behaving consistently with its own history. A process can pass every specification while its centre drifts toward a limit or its variability widens — trends invisible to release testing because each individual result is still acceptable. Continued verification catches these through trend analysis, distinguishing common-cause from special-cause variation and acting while batches still pass. An organisation that equates it with confirming results stayed in specification will report a process in control right up to the batch that fails, and the failure will look sudden only because nobody read the trend announcing it.
How do FDA's continued process verification and EU Annex 15 relate?
They are the same duty written by two authorities. FDA's lifecycle guidance names it continued process verification as the third stage; EU GMP Annex 15 requires ongoing process verification; and ICH Q10 frames continual monitoring and improvement as a standing quality-system responsibility rather than a validation appendix. A manufacturer supplying globally runs one programme designed so a single monitoring and trending system satisfies each expression of the underlying obligation — that a process shown capable at qualification remains in a state of control across its commercial life. Read together the three clarify the programme's purpose, because each supplies an emphasis the others assume: staged logic and the tie to design, the EU qualification-and-validation anchor, and the placement inside management review.
Why isn't trend monitoring on its own enough to count as verification?
Because verification implies a judgement returned and, where warranted, an action taken — charts alone are monitoring. The signals a continued-verification programme produces have to feed two loops to earn the name. They are inputs to ICH Q9 risk re-assessment: a parameter trending toward a limit or widening variability asks whether the risk picture the control strategy was built on has changed. And they are inputs to the ICH Q10 management-review and continual-improvement machinery. This also sets the evidentiary standard: the defensible artefact is not a periodic re-qualification report but a connected chain from monitored signal to assessment to action to outcome. An inspector asks for an instance where a trend triggered an assessment and an action, because that chain is the proof the loop is closed.