Concept↦Data Management↦Clinical Data Mgmt. System↦Computer System Validation
What is it? Why is it important?
A Computer System Validation (CSV) is a documented validation-process, verifying that a computer-based system does exactly what it is designed to do (e.g. consistent, reproducible, accurate)
In studies, the aim is to demonstrate that the study data is correctly handled from collection/storage (CDMS) to analysis (e.g. statistical software).
Problems must be identified and controlled, as it may compromise data:
Completeness: a complete dataset contains:
- All relevant information for a given purpose
- No missing, duplicated or irrelevant data
Accuracy: requires trained users of software systems, the implementation of data checks (e.g. source data verification), and query management (e.g. data monitoring)
Reliability: requires that data is reproducible:
- Upon double entry same datasets are produced
- Upon repeated statistical analysis same study results are produced
- Data records are retained as originally stored
Integrity: data is neither distorted nor lost due to;
- Changes to the software (e.g. software updates)
- Data being migrated/processed/transferred
What do I need to do?
As SP-INV, prepare the following documents:
-
- User Requirement (USR): describe stakeholder and CDMS expectations.
- Functional Specifications (FS) describe what end-users want the system to do (i.e. system capabilities, appearance and interactions with users)
- Design Specification (DS) describe how FSs are translated in software modules (i.e. collection of building blocks configured and adapted for different user requirements)
Document URS in a traceability matrix, and relate every USR to one or more FS, and link every FS to a given software module.
Describe software validation steps:
-
- Installation Qualification (IQ) describe how software modules are installed, connected to each other, and tested step by step
- Operational Qualification (OQ) describe tests related to any implemented FS of the CDMS itself as a unique system
- Performance Qualification (PQ) describe the implementation of a clinical trial with the described system (e.g. through screenshots of the developed form)
More
Describe in an SOP(s) the data-life-cycle or data-flow from data source (e.g. patient file, lab results), through CDMS, to the statistical software
The PQ is the only document that should be prepared ex-novo for every clinical trial, as it represents the data collection specificity of every study. In the event of study modification (e.g. protocol amendment), changes must be validated.
Where can I get help?
Your local CTU↧ can support you with experienced staff regarding this topic
Basel, Departement Klinische Forschung, CTU, dkf.unibas.ch
Lugano, Clinical Trials Unit, CTU-EOC, www.ctueoc.ch
Bern, Clinical Trials Unit, CTU, www.ctu.unibe.ch
Geneva, Clinical Research Center, CRC, crc.hug.ch
Lausanne, Clinical Research Center, CRC, www.chuv.ch
St. Gallen, Clinical Trials Unit, CTU, www.kssg.ch
Zürich, Clinical Trials Center, CTC, www.usz.ch
External Links
ECRIN – see in particular
- Services/Data Centre Certification
GCDMP – see in particular
- Chapter “Database Validation, Programming and Standards”
FDA – General Principles of Software Validation
Swiss Law
ICH GCP E6(R2) – see in particular guidelines
- 5.5 Trial Management, data handling, and record-keeping