Data Integrity in QC: Practical Step-by-Step Guide

How to Implement Data Integrity in the QC Laboratory: Operational Step-by-Step Guideline

Introduction

To be truly audit-ready, a QC laboratory must integrate Data Integrity into every phase of data generation, review and archiving.
This practical guide follows the operational approach described in the reference document, transforming regulatory principles into daily, actionable practices.

Phase 1: Data Generation (Analyst)

1. Contemporaneous recording

  • Record every activity at the moment it is performed (ALCOA – Contemporaneous).
  • Avoid temporary notes on uncontrolled paper.

2. Complete raw data

  • Immediately save chromatographic files, sequences and metadata.
  • Ensure the system does not allow destructive modifications to raw data files.

3. Peak integration

  • Default to automatic integration.
  • Perform manual reintegration only when necessary, and always document:
  • reason,
  • parameters modified,
  • impact on the result.

4. Prevent common errors

Errors to avoid:

  • manual file saving to USB devices
  • use of shared accounts
  • replacing files or performing unauthorised reprocessing

Phase 2: Data Review (QC Supervisor / QA)

Essential Data Reviewer Checklist

  • Are all raw data present?
  • Were audit trails reviewed, with no unjustified critical events?
  • Are reintegrations documented and approved?
  • Is there consistency between the lab notebook, CDS sequence and final report?
  • Are methods up to date according to change control?

The guide provides specific templates for reviewing HPLC, GC, Empower, LIMS and ELN data.

Phase 3: Handling Deviations, OOS and OOT

  • Do not repeat tests without a formal investigation.
  • Correctly connect Data Integrity with OOS/OOT:
    A reintegration that changes the result requires the opening of an investigation.
  • Preserve all result set versions generated.

How to manage a nonconformity

  1. Identify the event (e.g., incorrectly integrated peak).
  2. Assess impact on the result.
  3. Document fully in the system.
  4. Involve QA.
  5. Define CAPA to prevent recurrence (e.g., update the method).

Common Errors and Operational Solutions

Error Risk Solution
Missing audit trail review Critical finding Mandatory SOP + checklists
Manual backups Data loss Centralised automatic backups
Iterative reprocessing Manipulation suspicion Restrict privileges + track each execution
Missing data Incomplete dataset Daily data reconciliation scheme

GMP Best Practice

“If it’s critical, make it visible.”
Any activity that can influence data must leave a verifiable and traceable footprint.

Realistic Scenario

An analyst reintegrates an impurity peak to correct faulty automatic integration.
The modification, however, changes the value from OOT to compliant.

→ Correct actions according to the guide:

  • immediate entry in LIMS
  • supervisor review and approval
  • opening of an OOS
  • verification of the integration method
  • complete recalculation via a versioned result set

Explore the Full Guide

Learn how to apply Data Integrity correctly in QC labs: procedures, checks, common pitfalls and operational checklists — audit-ready and fully compliant.

Back to blog

Looking for something specific?

DISCOVER THE FULL GXP CATALOG