Data integrity and ALCOA+ in the QC lab

Data integrity is the degree to which data are complete, consistent and accurate throughout their whole life — from the moment a result is generated, through processing, review and reporting, to archiving and final destruction. In a QC lab, the data are the evidence that a batch is fit for release. If the data cannot be trusted, neither can the batch decision.

Regulators do not see data integrity as a new rule. It is the basic expectation of GMP: that records are true, were made when the work was done, and show everything that happened. What changed in the last decade is focus. Inspectors now look at audit trails, instrument folders and deleted files, not only at the printed report.

Data integrity is a shared responsibility. Analysts generate honest data, reviewers check the full record, IT and system owners configure computerised systems correctly, and management builds a culture where people can report errors without fear.

A useful test: could someone who was not in the lab reconstruct exactly what you did, when, with which sample, instrument and standard, and why — from the records alone?

ALCOA is the short checklist used across GMP to describe good data. The “+” adds four more attributes that the WHO, MHRA and PIC/S guidance documents make explicit. Some companies also talk about ALCOA++ and add “traceable”; the idea is the same.

Raw data are the original records of an activity, in the form in which they were first generated. For a balance this may be the printout; for an HPLC it is the electronic data file in the CDS, not the PDF report. Guidance from FDA, MHRA and WHO is clear that for computerised systems the electronic record is usually the original, and a paper printout alone is not a complete copy.

Metadata are the “data about the data” that give a result its meaning: who acquired it, when, on which instrument, with which method version, processing parameters, integration events and audit trail entries. A peak area without its metadata cannot be fully reconstructed or checked.

A true copy is a copy that keeps the content and meaning of the original, including relevant metadata, and has been verified as such by a documented process. Scanning a notebook page can make a true copy; printing a summary table from a CDS does not.

• Static records (a paper printout, a fixed PDF) cannot be reprocessed; dynamic records (a chromatogram file) can be. Dynamic data should normally be kept in their dynamic form.
• Define in your SOP which record is the original for each instrument: printout, electronic file, or both.
• Hybrid systems (paper plus electronic) need clear rules linking the signed paper to the exact electronic data set.

An audit trail is a secure, computer-generated, time-stamped record of who created, changed or deleted data, when, and why. It must be switched on, and ordinary users must not be able to switch it off or edit it. In a CDS it shows events such as sequence creation, method changes, manual integration, reprocessing and data deletion attempts.

Having an audit trail is not enough: someone must look at it. Regulators expect a risk-based review of audit trails for critical data, normally as part of reviewing results before they are approved. The reviewer looks for changes that could alter the reported result and checks that each one was justified.

1. 1Confirm the sequence in the CDS matches the sequence on the worksheet — same samples, standards and number of injections.
2. 2Look for injections that were run but not reported, aborted runs, or extra sequences on the same day.
3. 3Check processing history: how many times was the data reprocessed, and were integration parameters or methods changed?
4. 4Check every manual integration has a reason that matches the site SOP and does not just move a result into specification.
5. 5Look for deletions, renamed files, or date and time changes on the instrument PC.
6. 6Record that the audit trail review was done, by whom, and any findings, as defined in your SOP.

Audit trail review is part of the result review, not a separate yearly exercise. A result is not fully reviewed until its audit trail has been checked.

In the US, 21 CFR Part 11 sets the conditions under which FDA accepts electronic records and electronic signatures as equivalent to paper. In Europe, EU GMP Annex 11 covers computerised systems used in GMP. The details differ, but the expectations overlap strongly.

In practice both require: validated systems that do what they claim; unique user accounts with role-based access; secure, time-stamped audit trails; protection of records from loss or change; and electronic signatures that are linked to the record and show the signer's name, date, time and the meaning of the signature (for example “performed”, “reviewed”, “approved”).

• Analysts should not have administrator rights on systems that hold their own data.
• Passwords are personal; never share them or let a colleague work under your login.
• Data should be saved automatically to a secure server, not to a local desktop or USB stick where they can be lost or altered.
• Instrument PC clocks should be locked and synchronised so time stamps are reliable.
• Backups and archive restores should be tested, not assumed.

Warning letters and inspection reports repeat the same lab findings year after year. Most do not start as deliberate fraud. They start as shortcuts, poor system set-up or pressure to release batches. Knowing them helps you avoid them.

If you make a mistake, record it and tell your supervisor. An honest, documented error is easy to deal with. A hidden one becomes a data integrity finding.

Good documentation practice (GDP) is the day-to-day habit that turns ALCOA+ into reality on paper. Most labs use controlled, issued worksheets or bound notebooks with numbered pages so that pages cannot be removed or added without trace.

• Write in permanent ink at the time you do the work. Never use pencil, correction fluid or erasable pens.
• Correct an error with a single line so the original is still readable, then add the correct entry, your initials, the date and a reason where your SOP requires it.
• Do not leave blank spaces: line through unused fields and mark them “NA” or as defined in your SOP.
• Attach instrument printouts (balance, pH meter) directly, and sign or initial across the printout and the page where required.
• Record the identity of every instrument, column, reference standard, reagent lot and solution used, with expiry dates.
• Never write results on loose paper first and transcribe later.
• Use the date and time format defined by your site so entries are unambiguous.
• Reviewers check the full record against the raw data, not just the final result.

Inspectors often start at the instrument, not at the report. They may ask an analyst to show the CDS project for a released batch, open the audit trail, and compare it with the paper worksheet. They also look at the lab culture: whether people feel able to report failures, and whether management reviews data integrity metrics.

• User access lists: who has administrator rights, and are accounts unique and current?
• Audit trails switched on, and evidence that they are reviewed with each result set.
• Failed, aborted or unreported runs, and how they were handled.
• Manual integration frequency and justification.
• Balance and pH printouts against the worksheet entries.
• Issuance and reconciliation of blank worksheets and notebooks.
• OOS investigations closed as “lab error” without a clear root cause.
• Backup, archive and restore records for electronic data.

The FDA (2018), MHRA (2018), WHO TRS 1033 Annex 4 (2021) and PIC/S PI 041 documents all describe these expectations in detail. They are free to download and worth reading in full.