Why 'continue to monitor' fails every FDA audit (and what to write instead)
Every CAPA you write ends with an effectiveness check. It's the last line. It's also the line FDA inspectors read first — because it tells them whether you've actually defined what success looks like, or whether you're hoping to figure that out later. The single most-cited phrase in CAPA-inadequacy 483s isn't a missing root cause or a weak corrective action. It's four words: 'continue to monitor going forward.'
Why FDA flags 'continue to monitor'
An effectiveness check is supposed to answer a yes/no question: did the CAPA work? 'Continue to monitor' doesn't answer it. It defers it. There's no threshold, no sample size, no closure criterion, no time-box. The CAPA can never fail effectiveness — which is also why it can never demonstrate success. Under 21 CFR 820.100(a)(4), CAPAs must include 'verifying or validating the corrective and preventive action to ensure that such action is effective.' The verb is verifying. You cannot verify without a threshold to verify against. FDA writes Warning Letters citing this every quarter.
The Cipla Goa Warning Letter (paraphrased)
In FY2023, FDA cited Cipla's Goa facility for closing multiple CAPAs with 'monitor for 6 months' as the effectiveness verification — with no defined acceptance criteria, sample size, or closure threshold. The agency required revised CAPAs with measurable closure criteria. This isn't an exotic finding. It's the most-cited CAPA-effectiveness pattern in the FDA's Warning Letter corpus. Aspen Pharmacare, Sun Pharmaceutical, Lupin, Macleods Pharmaceuticals — all cited for the same pattern in different forms over the last 24 months.
What a defensible effectiveness check looks like
A defensible effectiveness check has four parts. Every part has a number. No part is optional.
- Window. A specific number of batches, weeks, or units. ('30 consecutive batches' beats 'over the next 6 months.')
- Sample size. The n that gives you statistical power to detect the goal target shift. ('~n = 8,000 vials across 30 batches' beats 'a sample of recent runs.')
- Threshold. The numeric pass/fail criterion. ('Reject rate < 2.0%' beats 'reject rate at industry standard.')
- Closure rule. When you'll close, and what auto-reopens it. ('QA closure decision at the 90-day review; auto-reopen if the 30- or 60-day checkpoint shows a regression of > 1 pp.')
Worked example — rewrite of a real-shape effectiveness check
The pre-flight habit
Before any CAPA leaves your inbox, search the effectiveness section for these phrases: 'continue to monitor,' 'monitor going forward,' 'review trend data,' 'track for recurrence.' If any of them appear, rewrite. The rewrite takes 90 seconds. The 483 you don't get takes 6-18 months to remediate and costs an average of $1-5M.
How Taktly catches it automatically
Taktly's senior-review audit is a deterministic detector for 'continue to monitor' language paired with the live FDA Warning Letter pattern library. Paste your CAPA at gettaktly.com — in 30 seconds you'll see whether your effectiveness check would pass a 21 CFR 820.100 review. If it wouldn't, the audit hands you back a rewrite in practitioner voice.