How do you check CAPA effectiveness?
By Savi C., Co-Founder · Published Nov 16, 2026
Implementation is not effectiveness
A CAPA has two checks that are easy to confuse. Verification of implementation confirms the action was done: the SOP was revised, the training was delivered, the equipment was modified. Verification of effectiveness confirms the action worked: the deviation it was meant to prevent has not recurred, and nothing else got worse because of it.
Closing a CAPA once the action is implemented, with no evidence that it worked, is a common weakness. EU GMP Chapter 1 also asks for the effectiveness of CAPAs to be reviewed in the product quality review, and ISO 13485 clause 8.5.2 asks device manufacturers to verify that a corrective action does not adversely affect the ability to meet regulatory requirements or the safety and performance of the device.
Planning an effectiveness check
Write these into the CAPA record when it is opened, not when it is due to close.
What to measure
- The specific failure the root cause produced, not a general quality indicator
- A measure the action should move: recurrence of the deviation, a defect rate, a trend in a process parameter, audit findings on the same requirement
- Where the data will come from, and confirmation that it is collected consistently
When and how much
- A period or number of batches long enough for the problem to have recurred if the fix did not work
- A larger window for rare events, or a targeted check such as a focused audit
- Depth proportionate to risk, as ICH Q10 and EU GMP expect
What counts as success
- A success criterion set in advance, such as no recurrence across a defined number of batches
- What happens if the check fails: reopen the investigation, not only extend the CAPA
- A check that the action did not introduce a new problem
Weak and strong effectiveness checks
Weak
Training completed, CAPA closed
Stronger
No recurrence of the error across the next defined set of batches using the revised procedure
Weak
SOP revised and approved
Stronger
Review of executed records shows the revised step performed correctly
Weak
No complaints received in 30 days
Stronger
A period chosen from how often the problem occurred before, and justified
Weak
Effectiveness assumed because the root cause was addressed
Stronger
Data compared before and after the action, using the same measure
Weak
Each CAPA judged on its own
Stronger
Recurrences trended across products, sites and causes
Why record data matters here
Most effectiveness checks depend on one thing: being able to see whether the same failure happened again. On paper, that often means someone pulls batch records by hand and reads them looking for one kind of error, which is slow and easy to under-sample. When record review produces structured findings for every batch, recurrence becomes something you can count across every batch instead of a sample.
The same data feeds the next level up. ICH Q10 expects CAPA to draw on trends from process performance and product quality monitoring, and EU GMP Chapter 1 expects the product quality review to cover significant deviations and the effectiveness of the CAPAs that followed. A check that only ever looks at one CAPA at a time misses the pattern where the same cause keeps returning under different deviation numbers.
How Litewave fits
Litewave reads paper or electronic records, checks every entry against your master record and specifications, and shows QA only the exceptions, each linked to the page it came from. Because the same checks run on every batch, exceptions can be counted and trended over time, which gives an effectiveness check evidence from every batch rather than a hand-picked sample. Litewave proposes, qualified people approve, including the decision that a CAPA is effective.
Questions we get asked
How long should a CAPA effectiveness check run?
Long enough for the problem to have recurred if the action had not worked. There is no regulatory period. Base it on how often the problem occurred before, and record the rationale in the CAPA.
Is retraining an effective CAPA?
Sometimes, but it is often used as a default. EU GMP Chapter 1 asks for human error conclusions to be justified after checking for process, procedural or system causes. If retraining is the action, the effectiveness check should show the error stopped, not that the training was attended.
What happens if the effectiveness check fails?
Treat it as evidence that the root cause was wrong or the action was insufficient. Reopen or open a new investigation, revisit the root cause and assess whether batches made since the CAPA are affected.
