FDA Findings: Discarded and Shredded Original Records

The original record is the first capture of the data — the printout from the balance, the handwritten entry, the plate on the bench, the electronic file. Copies and transcriptions are only as good as the original they came from. When originals disappear, everything built on them becomes unverifiable. Three 2026 Warning Letters show how originals get lost: casually, systematically, and deliberately.

Background: data integrity and ALCOA+ in practice.

What FDA found

Company X1 — originals in the shred bins

An API manufacturer (FDA cites adulteration under FD&C Act section 501(a)(2)(B)). Under its quality-unit finding, the letter states that the investigator observed “original laboratory records inside the shred bins”, including instrument printouts and signed data review checklists that the firm could not attribute or link to batches. The firm’s review missed at least one bin, and it could not determine how much had already been destroyed by its third-party shredding provider.

FDA asked the firm to determine how long shred-bin disposal of GMP documents had been going on, to assess records destroyed before the inspection (the bins had been collected without QA review), and to report on every document found.

Company X2 — executed records in a garbage bag

Cited under 21 CFR 211.22. According to the letter, original records — including executed batch records, media preparation and sterilisation logbook pages, and sample and environmental monitoring forms — were found in a garbage bag. The firm also did not retain original data supporting test results used to release commercial batches. It attributed the incidents to a former manager; FDA found the response inadequate, stating it failed to describe immediate actions to prevent alteration or destruction of CGMP data.

Company X3 — plates discarded, pages replaced

Cited under 21 CFR 211.194(a). The letter describes, among other things, an allegation that a team leader told staff to discard bioburden plates showing significant growth, and that “pages were removed from your logbook and replaced with newly fabricated pages” made to look original. Because the problems extended to microbiology and environmental monitoring data, FDA considered the firm’s existing sterility and EM data unreliable as support for product safety.

Why this happens

Interpretation — the letters state findings, not root causes. These are common patterns, not claims about these firms.

• “It’s just a printout.” Staff transcribe a value into the official record and treat the printout as waste — not realising the printout is the original.
• Shredding without rules. Confidential-waste bins are emptied by a contractor, and nobody checks what goes into them.
• Rewriting instead of correcting. A messy or wrong page gets replaced with a clean one. The intention may be tidiness; the effect is destruction of the original.
• Pressure on a bad result. A failing plate or result that can be thrown away is a result that never has to be investigated.

What good looks like

Example: a fictional QC lab with paper worksheets, instrument printouts and an EM program.

1. Define “original” per record type. A record map lists, for each activity, what the original is (printout, worksheet, plate read, electronic file) and where it must be kept. Staff are trained on it.
2. Attach, don’t transcribe and toss. Instrument printouts are attached to the worksheet or batch record, or the data is captured electronically. If a value is transcribed, the original stays with the record and transcription is second-person verified.
3. No GMP paper in general waste. GMP records can only be destroyed through a documented process: after the retention period, with QA approval and a destruction log. Shred bins in GMP areas are controlled, or removed.
4. Correct, never replace. Errors are corrected on the original by single line, with initials, date and reason. Replacing a page requires a documented deviation, and the original page is retained.
5. Reconcile. Issued forms, logbook pages and test samples (including plates) are reconciled, so a missing item is visible.
6. Results are reported before they can be discarded. Microbiology plates and similar perishable originals are read and recorded at the bench, and any out-of-limit result is escalated before the material is disposed of.

Perspective: what FDA’s requests tell you

In X1, FDA’s first question was not “what did you find?” but how long the practice had been going on and what was destroyed before we arrived. That is the uncomfortable part of a discarded-records finding: the firm has to prove a negative about records that no longer exist. Prevention is far cheaper than that retrospective investigation — and a controlled destruction log is the evidence that turns “we think nothing important was lost” into “here is what was destroyed, when, and with whose approval.”

Regulatory basis

• 21 CFR 211.194(a) — laboratory records must include complete data, including all graphs, charts and spectra from laboratory instrumentation.
• 21 CFR 211.22 — quality control unit responsibilities, with written procedures that are followed.
• API manufacturers: FDA evaluates API CGMP against ICH Q7 expectations.
• FDA guidance: Data Integrity and Compliance With Drug CGMP — Questions and Answers (December 2018), question 12 (recording data on paper to be discarded is not acceptable) and question 18 (how data integrity problems should be addressed). Guidance is non-binding; the regulations are.

EU angle

Check current versions on EudraLex before relying on specific clauses. In EU GMP, retention and handling of original records sit mainly in EudraLex Volume 4, Part I, Chapter 4 (Documentation), currently under revision (draft published for consultation in July 2025).

Takeaways

• The printout is the original. The transcription is the copy.
• No GMP record leaves the system without an approved destruction log.
• Correct the original; never replace it.
• Once originals are gone, the burden is on you to prove nothing mattered.