A laboratory may have approved procedures, trained analysts, electronic records and a documented audit trail review process, yet still have important weaknesses in data integrity.
This happens because the existence of a control does not necessarily demonstrate that the control works.
Effective GMP laboratory data integrity depends on whether the organization can reliably establish what happened during testing, whether records accurately represent the work performed and whether quality decisions are supported by complete and trustworthy evidence.
For laboratories preparing for regulatory inspections, the focus should be on operational control rather than documentation volume.
Understand what data integrity means in practice
Data integrity concerns the completeness, consistency and accuracy of data throughout their lifecycle.
The ALCOA principles provide a familiar foundation:
- Attributable: It should be clear who performed an activity or generated a record.
- Legible: Information must remain readable and understandable.
- Contemporaneous: Activities should be recorded when they occur.
- Original: Appropriate original records or verified true copies should be retained.
- Accurate: Records should correctly represent the activity performed.
The commonly used ALCOA+ formulation adds complete, consistent, enduring and available.
These principles are useful for evaluating practical controls, but ALCOA+ is not itself a separate federal regulation.
FDA explains the relevant expectations in Data Integrity and Compliance With Drug CGMP: Questions and Answers.
The UK's Medicines and Healthcare products Regulatory Agency provides additional interpretation through its GxP Data Integrity Guidance.
The MHRA publication also notes that an OECD advisory document on data integrity takes precedence over the older MHRA guidance within the applicable OECD framework. Organizations should therefore verify which guidance is relevant to their activities.
Follow the complete laboratory data lifecycle
A useful inspection-readiness exercise is to follow one analytical result from sample receipt to final reporting.
The review might examine:
- Sample identification and receipt.
- Preparation of samples and standards.
- Instrument configuration and suitability.
- Data acquisition and original electronic records.
- Processing and calculations.
- Review of results and relevant metadata.
- Investigation of unusual observations.
- Approval, reporting, retention and retrieval.
At each point, ask whether the activity can be reconstructed from the available records.
For example, a final chromatographic report may show a passing result without displaying every processing action performed before the result was approved.
If relevant original data and processing history cannot be retrieved, the final report alone may not establish that the analytical activity was appropriately controlled.
Distinguish raw data from summary reports
A common weakness occurs when reviewers rely only on printed reports or exported PDFs.
Electronic laboratory systems may contain additional information that is important for understanding how a result was generated.
Depending on the system, this can include acquisition files, processing parameters, integration changes, sequence information, system events and audit trail entries.
The information required for an adequate review depends on the system and the regulated activity.
A PDF summary is not automatically inadequate, but organizations need to demonstrate that relevant original records, metadata and processing history are appropriately retained and available.
Quality personnel should understand what their electronic systems record and what information is lost when data are exported.
Make audit trail review meaningful
An audit trail is not useful simply because a software feature is enabled.
A well-designed review process identifies which events have significance for data integrity and quality decisions.
Potentially important events include:
- Changes to analytical methods or processing parameters.
- Modifications to results after initial calculation.
- Deletion or exclusion of data.
- Unusual user-access activity.
- Changes to system time or configuration.
- Repeated analyses requiring investigation or explanation.
Not every recorded event needs the same level of scrutiny.
For example, an ordinary login event may be less significant than an unexplained modification to a result used for batch disposition.
The review strategy should consider the criticality of the data, the system's capabilities and the potential consequences of inappropriate changes.
The reviewer should also have sufficient training and access to understand the events being reviewed.
Evaluate access controls and segregation of duties
Access management is a fundamental part of data integrity.
If analysts can modify system settings, delete original records or alter audit trail configurations without appropriate authorization, the organization may have difficulty demonstrating reliable control.
A practical review should identify:
User accounts: Are access rights assigned to named individuals?
Privileges: Are users limited to functions appropriate for their responsibilities?
Administrative access: Is elevated access controlled and monitored?
Account changes: Are access rights removed or modified promptly when roles change?
Review independence: Are reviewers sufficiently independent of the activities they assess?
Shared accounts can make attribution difficult because several individuals may be able to perform actions under the same identity.
The objective is not to prevent legitimate laboratory work but to maintain accountability and protect regulated information from inappropriate changes.
Address hybrid records and spreadsheets
Many laboratories still use a combination of electronic systems, paper records and spreadsheets.
A hybrid process can introduce gaps when information moves between formats.
Consider a hypothetical laboratory that exports instrument results into an Excel workbook for final calculations.
If formulas are accidentally overwritten, the resulting report may be incorrect even when the original instrument data remain accurate.
Controls might include restricted formula cells, approved workbook versions, independent verification, appropriate record retention and testing of calculation logic.
However, ordinary spreadsheet protection is not necessarily sufficient for all applicable electronic-record requirements.
The organization must determine whether the workbook and associated workflow require additional controls, taking account of intended use and applicable regulations.
Investigate unusual results transparently
Unexpected analytical results should not be resolved by repeatedly testing until a preferred result appears.
The appropriate investigation must consider the original observation, available evidence, potential laboratory error and possible product-quality implications.
An investigation should distinguish confirmed evidence from hypotheses.
Hypothetical example: An analyst obtains an unexpected assay result and suspects a sample preparation error.
The investigation should identify supporting evidence, such as preparation records, calculations, instrument information or independently verified observations.
A general statement that the result was caused by "analyst error" is not equivalent to demonstrating the cause.
FDA's CGMP questions and answers provide additional regulatory context for laboratory and manufacturing quality activities.
Test oversight rather than relying on training completion
Training records are important, but completing annual data integrity training does not prove that laboratory controls are effective.
Organizations should also assess whether personnel apply procedures correctly.
Possible evaluation methods include direct observation, targeted record review, periodic access assessments, investigation trend analysis and review of recurring audit trail findings.
For instance, repeated unexplained integration changes across several analysts may indicate a systemic weakness in method control, training or supervisory oversight.
A meaningful quality system should recognize such patterns and respond appropriately.
Conduct a practical inspection-readiness exercise
Select a recently completed analytical test and ask an independent reviewer to reconstruct the complete activity.
Can the reviewer identify the sample, method, instrument, user, acquisition history, processing decisions and final approved result?
Can unusual events be explained?
Are supporting records available without substantial manual reconstruction?
Where gaps occur, determine whether they reflect missing documentation, inadequate system configuration, unclear procedures or a broader governance problem.
This exercise can reveal weaknesses that routine procedural reviews overlook.
Practical takeaway
Effective data integrity rests on three related forms of control:
Technical control: Systems must protect and preserve reliable records.
Procedural control: Workflows must define how data are generated, reviewed and retained.
Management oversight: Quality leadership must confirm that those controls operate effectively.
A laboratory that can demonstrate all three is better positioned to defend the reliability of its results than one that relies exclusively on procedures and training records.
Explore the related EduCraft session
For a broader examination of electronic records, audit trail review, laboratory vulnerabilities and inspection-oriented quality oversight, explore GMP Laboratory Data Integrity: Practical Strategies for FDA and MHRA Inspection Readiness. Check the course page for current access options.
