Data Integrity in the QC Laboratory

Excerpt from the GMP Compliance Adviser, Chapter 14.L, Data integrity in the quality control laboratory

5 min. reading time | by Markus Veit, PhD 
Published in LOGFILE 22/2025 

Data integrity plays a particularly important role in pharmaceutical quality control as the results generated here are the basis of product quality assessment and thus are relevant for batch release. This article explains, which data arise during HPLC analytics and how the ALCOA principles are linked to the process of data generation.

This article deals with the practical aspects of implementing the European requirements for data integrity. A HPLC laboratory for pharmaceutical quality control is used as an example in which data for the batch release of a finished medicinal product is generated. How does the release process work and what kind of data is generated? 

For batch certification the test results generated by quality control are compared to the specifications in the authorisation documents and subsequently filed in a Certificate of Analysis (CoA). 

In case of test results meeting the specifications, the batch can be released; the release is documented in the release register. It is part of the GMP documentation where all testing data is collected.

What types of data occur in a chromatography laboratory is shown in Figure 1. The list does not claim to be exhaustive.

Figure 1 | Data in a chromatography laboratory

All of this data should comply with the ALCOA principles – and, in an ideal situation, the extended ALCOA plus („ALCOA⁺⁺“) principles (see Figure 2).

• Attributable

• Legible

• Contemporaneous

• Original

• Accurate

Figure 2 | ALCOA / ALCOA++ principles of data integrity

A number of these requirements listed in Figure 2 are already contained in the EU GMP Guidelines, i.e. they are not newly introduced by the publication of data integrity specifications. In particular, the traceability of data connects the ALCOA elements (McDowall, R. D.: Is Traceability the Glue for ALCOA, ALCOA+, or ALCOA++? in: Spectroscopy, April 2022, Volume 37, Issue 4, p. 13–19; https://doi.org/10.56530/spectroscopy.up8185n1).

Key aspects of data integrity in the chronological order of data generation are examined below. 

Data on sampling and the handling of samples is frequently stored in a LIMS. In practice, however, hybrid systems are also used. When this is the case, special care must be taken to ensure that data is properly transferred. Data on sample preparation and preparation for chromatographic determination is generally documented on paper. Not all of the data must be transferred to the LIMS; however, it is important to ensure that the batch-related traceability of data does not depend on the type of documentation used. During chromatographic determination, only electronic data is generated. It is read in the form of integrated chromatograms and the resulting peak areas. Subsequent manual reintegration is not always avoidable and continues to be a controversial aspect with regard to data integrity.

The further processing of data during evaluation can be carried out using integrated evaluation software, LIMS or external programs. Different aspects of data integrity need to be observed. This is a common source of non-compliance that is often not recognised by the operator and/or cannot be influenced by him/her and also explains why the au-thorities stipulate that an audit trail review must be carried out prior to batch release.

When a report is generated, it must be ensured that it is an original document, regardless of whether it is paper-based or electronic. The storage and archiving of data must also be regulated in detail so that the data is protected against subsequent manipulation, unauthorised access or loss. It must also be ensured that the data can be retrieved and remains legible for the entire mandatory period of retention. 

Non-compliance with data integrity requirements e.g. result from insufficient access controls, incomplete qualification and validation, errors during audit trail review or during handling of changes and deviations.

For process development compliance with data integrity is not required comprehensively. Nevertheless, basic requirements e.g. data accuracy has to be fulfilled to achieve reli-able results. This includes correct handling of statistical analysis. 

Data integrity is a critical and very complex topic. Comprehensive trainings for everyone involved in the processes thus are mandatory. Compliance with data integrity should be reviewed by regular self-inspections.