If your organisation makes, tests or distributes medicines, much of the software it uses is subject to GxP rules. Computer system validation (CSV) is how you show regulators, and yourself, that this software can be trusted. This guide explains what validation means, when it applies and how to approach it without drowning in paperwork.
What is computer system validation?
Computer system validation is documented evidence that a computerised system does what it is intended to do, consistently and reliably, in its real environment. It covers the whole life of the system: from defining requirements and choosing a supplier to testing, operation, changes and, finally, retirement.
The aim is not the documents themselves but what they prove: that the system protects product quality, patient safety and the integrity of the data it handles.
When does validation apply?
Validation applies to systems used in GxP activities, the good practices that cover manufacturing (GMP), laboratories (GLP), clinical trials (GCP) and distribution (GDP). As a rule of thumb, a system needs validation if it:
- creates, changes, stores or reports records that regulators may inspect;
- controls or monitors the manufacture, testing or release of a product;
- supports decisions that affect product quality or patient safety.
Typical examples are laboratory information management systems (LIMS), quality management systems (QMS), manufacturing execution systems (MES), and the spreadsheets, databases and integrations around them.
The regulations to know
- EU GMP Annex 11 sets the expectations for computerised systems in the European Union, from risk management and supplier assessment to audit trails and business continuity. A revised version is in preparation.
- FDA 21 CFR Part 11 defines when electronic records and electronic signatures can be trusted as equivalent to paper records and handwritten signatures.
- GAMP 5 (Second Edition), published by ISPE, is the industry reference guide for a risk-based approach to validating computerised systems.
- Computer software assurance (CSA), promoted by the FDA, focuses testing effort on what matters most to patient safety and product quality, rather than on producing documents for their own sake.
A risk-based validation lifecycle
Modern validation scales the effort to the risk. A simple, widely used configurable product needs less testing than a custom-built system that controls a critical process. A typical lifecycle includes:
- User requirements: what the system must do, written in testable terms.
- Risk assessment: which functions affect patient safety, product quality or data integrity, and how much testing each one needs.
- Supplier assessment: how the supplier develops, tests and supports the software, so you can rely on its own work where appropriate.
- Specifications and configuration: how the system is set up to meet the requirements.
- Testing: evidence that the requirements are met, focused on the highest risks, with a traceability matrix that links each requirement to its tests.
- Validation report and release: a summary of the results and a formal decision that the system is fit for use.
- Operation: change control, periodic review, backups, access management and audit trail review keep the system in a validated state until it is retired.
Data integrity and ALCOA+
Regulators expect data to follow the ALCOA+ principles: attributable, legible, contemporaneous, original and accurate, as well as complete, consistent, enduring and available. In practice, software should provide unique user accounts, role-based access, secure time-stamped audit trails, protection against unauthorised changes and reliable backup and restore.
Common pitfalls
- Validating everything with the same intensity, instead of focusing on what carries risk.
- Writing requirements that cannot be tested.
- Treating validation as a one-off project and forgetting change control after go-live.
- Overlooking spreadsheets, interfaces and data migrations, which often hold critical data.
- Leaving audit trails switched off, or never reviewing them.
How Aryaz can help
We design and build software for pharma and life sciences with validation in mind from day one. We help you define clear requirements, build in data integrity controls, prepare test evidence and documentation for your quality team, and migrate data from legacy systems with full traceability.
Find out more about our software for pharma and life sciences, or talk to our team about your next system.
This article is general information, not regulatory advice. Always check the current version of the regulations that apply to you.