Validation records

The standard requires validation results to be recorded and design failures to be documented in the validation records. 

Design verification measures (clause 4.4.7) Design validation records (clause 4.4.8.2)... 

Upon completion of the experimental phase of validation, the data are compiled and evaluated by qualified scientific personnel. The results may be summarized on a summary sheet, an example of which is shown in Table 2. Once a process has been validated, it must be controlled to assure that the process consistently produces a product within the specifications established by the validation studies. As shown in Table 2, documentation should present original validation records, a schedule of revalidation dates, and data from the revalidation studies. The interval between validation studies strictly depends on the judgment of the validation team based on the experience and history of the consistency of the process. 

Description of process validated load containing filling equipment and accessories not to exceed 102 kg Temperature set point for validation 121.0°C Temperature range for validation +0.5°C Cycle validated 35 min Validation records stored in archives A105-11 Revalidation records stored in archives C314-70... 

Time-based reviews should be planned for at defined intervals to check adherence to procedures and the currency of validation records. The frequency... 

The U.S. issued specific distribution requirements in 1990 to complement the basic provisions established earlier in its GMPs. Although computer validation is not specifically identified as a requirement, separate electronic record/signature legislation (21 CFR Part 11) requires computer systems handling defined distribution records to be validated. Records should be sufficient to track the origin and destination of medicinal products primarily in support of possible customer returns and product recall. Key GDP topics identified in the U.S. legislation that validation should address in relation to computer systems include ... 

System requirements must be traceable throughout validation records. 

Computer Aided Software Engineering used in context of supporting creation, ongoing maintenance, and/or retrieval of validation records... 

Validation records did not address wiring diagrams. [FDA Warning Letter, 2001]... 

What documentation generally exists to provide up-to-date descriptions of the systems and to show physical arrangements, data flows, interactions with other systems, and hfe-cycle and validation records Comment as to whether all of these systems have been fully documented and validated. 

Factory Acceptance Testing (FAT) is conducted within a simulated environment to demonstrate that the system meets the URS and FDS. The FAT will only be conducted if the software is a new development or major adaptation of the standard product. The FAT is the first opportunity for the pharmaceutical manufacturer to test the system in its entirety. The scope of the testing should be wide enough to ensure that most problems can be identified and rectified within the development enviromnent. Where tests are not dependent on the operating environment as, for example, in data entry validation, the FAT may serve as the validation record for that function. However, in order to adopt this approach, formal test specifications must be developed and approved prior to executing the tests, and results must be clearly recorded. 

Validation records of test methods, where applicable. 

To retain the validation records for 3 years after the date of recording. 

Validation records version of software to be recorded on documents... 

Whilst these cover the basic interfaces to the process, i.e., the process equipment and the operator, they do not cover the actual production environment itself. Many products require specific environmental conditions to be maintained, such as temperature, sterility, humidity and so on. These are often controlled by external systems, and the need to validate these conditions means that a separate, validated recording system is often provided for the air-conditioning system, which is known as a BMS. 

Performance goals/acceptance criteria outlined earlier align with FDA/EMA Bioanalytical Method Validation Guidances and published best practices for method validation. However, technologies and/or platforms used may require additional or alternative validation plans and/or target criteria. All plans, procedures, and data should be documented in validation records. 

The process validation should show that the chosen formulation, in combination with the chosen method of preparation, will lead to a product with a consistent quality. The results of the validation are recorded on a separate validation record. This record should contain information on the conditions during preparation and on the method of sampling for analysis, together with the date of drawing up the corresponding BPI. 

In order to release a product - and all its updates - for clinical use, it is mandatory to re-evaluate all the open defects for their level of risk none with severe safety risks can be left open, all those remaining have to be formally evaluated and accepted, documenting the reasons for the acceptance and/or the workaround to be applied. The formal defect disposition is the last step of the validation process and an input to release process. Defect list, risk evaluation, disposition and justification are part of the validation records. In special cases a version can be approved for release if the validation documents are not formally completed in no case can a version be approved if the risk evaluation is not completed. 

It is important to recognize the distinction between data and other information that may be in a computer data base, and a permanent record of that information. It is not uncommon to acquire primary data electronically and accumulate information in data bases. These data bases are sometimes described as electronic notebooks. Such a method has the merit of enabling rapid retrieval and facile work-up of data. But there is no way to guarantee the integrity of a data base, neither as regards content, time, person nor place in the absence of some kind of hard copy. Only a timely print-out of a computer file or a data base—signed, dated, witnessed and secured—can serve as a record. In the future, courts may accept other criteria for valid records. For the present. Research Notebooks shall be in paper or other permanent, physical form. 

---