A validation deviation is defined as any non-conformance that occurs during any validation or qualification process in relation to the validation protocol that has been previously approved, prescribed acceptance criteria, procedure that has been approved and anticipated conditions. Any validation deviations should be recorded, investigated, classified according to their importance and assessed in terms of their potency impact on the validity of the study, safety of the final product and its efficacy for patients and validity of the technology.
A validation deviation does not lead to invalidation of the entire validation. However, a positive outcome of the study does not ensure that this deviation is permitted or justified in compliance with facts or reasons from scientific practice supporting the outcome and applicable pharmaceutical quality management system.
The classification should refer to the consequences of the event and not the manageability of correction of the problem.
1. Critical deviations may involve the existence of the credible threat to patient safety, major data integrity failure, loss of control over the significant process, or the indication of incapability of the research to deliver its conclusions. Thus, for instance, if there are any unjustified changes of important validation data, the integrity of the whole research might be in jeopardy, which will require following the escalation process according to the site’s quality department.
2. Major deviations may drastically impact important parameters, acceptance criteria, sampling strategies, or the interpretation of results. For example, if one sampling site in a cleaning validation study is omitted, this omission may be considered serious if that sampling site is hard to clean and there are doubts regarding the residue at this site.
3. Minor deviations have little influence on the validity or quality of the study when there is supporting documentation. For example, if a mistake made in the administration does not interfere with the implementation of the study or the traceability of results, but the site's procedures allow to classify this mistake as minor, it will fall under this limitation.
These examples serve only for reference and are not an exhaustive guide for classification of deviations. What seems to be a minor deviation can turn into a serious deviation if, for example, several linked deviations occur repeatedly in the same study, or if the parameter in question involves quality.
The responsible group shall decide whether continuing with the action, isolating the impacted equipment/data/file, preserving the electronic data, locating the impacted materials, or alerting the QA and validation owner.
For instance, when a temperature sensor breaks down as part of thermal qualification testing, the team must ensure thorough documentation of the breakdown, find out what measurement period was affected, keep the existing records and establish whether the study can go on with providing enough proof of the required temperature distribution. If the team continues to run their tests without evaluating the consequences of the breakdown before running the test, a huge volume of meaningless data will be produced.
A deviation record should contain details about the date and time of the incident, study and protocol IDs, equipment used, involved personnel, actions taken and preliminary description of the problem.
To start with, one should analyze the actual implementation of the protocol, SOPs, condition of the equipment and acceptance criteria. It is recommended to base your investigation on original raw data and records, not just on people’s memories or validation report summary.
In accordance with what has happened, useful evidence could be the following:
For example, the acceptance criterion fails for a validation sample. Possible reasons may be the variability in the process, error in sampling, analytical issues, or even a true failure in the control. The investigation should check the situations taking into consideration the relevant records and testing that is justified scientifically.
1. Effects on Validity of Work: Evaluate whether it remains possible to assess the goals and objectives stated in the protocol. Also, it should be noted that the absence of a critical phase under monitoring or of some mandatory testing may create a big gap in the evidence and validity of the work.
2. Effects on Integrity of Data: Determine whether the data is complete, traceable, valid and credible. Lack of basic data raw data or of some explanation as to why some changes have taken place or why some experiments have been excluded should be taken into consideration.
3. Effects on Quality of Products and Patient Safety: Where the manufacturing or analytic processes can have an impact on product quality, one should evaluate if the batches, samples, or some materials have been affected. The decisions regarding the disposition of the product must be made according to the relevant quality procedures and must be grounded on the evidence available.
4. Effects on Validated/Qualified Status: It has to be determined whether evidence supports the equipment, utilities, processes, cleaning procedure, testing method, or computer-based system.
5. Wider and Historical Impact: Check if such another failure could affect other equipment, shifts, products, protocols, or already completed studies. The recurrence probably indicates existence of some broader weakness, such as those in procedures, training, maintenance, or control strategy.
For every dimension, the rationale must be documented. Stating “no impact on product quality” is simply insufficient unless investigation shows proof supporting the conclusion drawn therein.
Examples of possible actions are equipment repair or equipment calibration, revision of a procedure, improvement of worker’s training, upgrading of an alarm or regulation, re-engineering of a sampling method and protocol improvement. Each action should be assigned to a responsible person with a due date and documented completion.
Validation team and QA then should assess whether the study can be accepted as carried out, requires some further justified testing, needs full or partial repetition, or does not allow drawing intended conclusions. Further testing should bridge an evidence gap without neglecting unfavourable information.
The validation report should include references to deviation records, investigations and impact assessments summary, CAPA descriptions and conclusion explanations. If the issue of the importance remains unanswered, the report should not state that validation has been carried out successfully without any justifying argument.
Common problems are shallow root-cause explanations, uneven classification for like events, absence of connections between deviations and validation reports and conclusions focused mostly on acceptable results.
A valuable internal review must confirm that each deviation has an accurate event description, an evidence-based classification system, a documented investigation process, a logical impact evaluation and legitimate disposal of action. Trends have to be monitored so that repetitive failures that are not obvious may be seen.
Classifying Validation Deviations by Risk and Significance
There is not a universal classification system which must be applied by all pharmaceutical companies to validation deviations. Companies should set definitions and criteria for their approved deviation management procedures. Some organizations use the categories of major, minor and critical deviations, while some may use risk-based classification systems.The classification should refer to the consequences of the event and not the manageability of correction of the problem.
1. Critical deviations may involve the existence of the credible threat to patient safety, major data integrity failure, loss of control over the significant process, or the indication of incapability of the research to deliver its conclusions. Thus, for instance, if there are any unjustified changes of important validation data, the integrity of the whole research might be in jeopardy, which will require following the escalation process according to the site’s quality department.
2. Major deviations may drastically impact important parameters, acceptance criteria, sampling strategies, or the interpretation of results. For example, if one sampling site in a cleaning validation study is omitted, this omission may be considered serious if that sampling site is hard to clean and there are doubts regarding the residue at this site.
3. Minor deviations have little influence on the validity or quality of the study when there is supporting documentation. For example, if a mistake made in the administration does not interfere with the implementation of the study or the traceability of results, but the site's procedures allow to classify this mistake as minor, it will fall under this limitation.
These examples serve only for reference and are not an exhaustive guide for classification of deviations. What seems to be a minor deviation can turn into a serious deviation if, for example, several linked deviations occur repeatedly in the same study, or if the parameter in question involves quality.
Immediate Actions: Control the Event Before Investigating
In an event of a deviation, the first thing to do is to document the event that has happened and act to stop further uncontrolled implementation of the activity. The action taken will vary depending on the specific situation inhand and level of risk involved.The responsible group shall decide whether continuing with the action, isolating the impacted equipment/data/file, preserving the electronic data, locating the impacted materials, or alerting the QA and validation owner.
For instance, when a temperature sensor breaks down as part of thermal qualification testing, the team must ensure thorough documentation of the breakdown, find out what measurement period was affected, keep the existing records and establish whether the study can go on with providing enough proof of the required temperature distribution. If the team continues to run their tests without evaluating the consequences of the breakdown before running the test, a huge volume of meaningless data will be produced.
A deviation record should contain details about the date and time of the incident, study and protocol IDs, equipment used, involved personnel, actions taken and preliminary description of the problem.
Investigating the Cause with Evidence
Proper investigation tells about the circumstances of the incident, its reason and whether there is a danger of the same happening again in the future in other places. The level of investigation should correspond to the amount of risk involved and the complexity of the case.To start with, one should analyze the actual implementation of the protocol, SOPs, condition of the equipment and acceptance criteria. It is recommended to base your investigation on original raw data and records, not just on people’s memories or validation report summary.
In accordance with what has happened, useful evidence could be the following:
- Executed protocol and protocols pages, worksheets, sampling records and printouts from devices.
- Equipment alerts, calibration status, maintenance history and qualification data.
- Electronic records and audit trails.
- Training records, applicable procedures and interviews with employees.
- Environmental trends, process deviations and other similar research.
- Records from a laboratory, chromatographs, calculations done and sample logistics.
For example, the acceptance criterion fails for a validation sample. Possible reasons may be the variability in the process, error in sampling, analytical issues, or even a true failure in the control. The investigation should check the situations taking into consideration the relevant records and testing that is justified scientifically.
Impact Assessment: Determine What the Deviation Changes
Impact assessment is a crucial process that needs to be assessed in case of any deviation during the validation exercise. It determines whether the event has negatively impacted the integrity of data, the conclusion of the study, the reliability of the process or system as well as the eventual quality of the products made. The following issues should be examined when evaluating the impact.1. Effects on Validity of Work: Evaluate whether it remains possible to assess the goals and objectives stated in the protocol. Also, it should be noted that the absence of a critical phase under monitoring or of some mandatory testing may create a big gap in the evidence and validity of the work.
2. Effects on Integrity of Data: Determine whether the data is complete, traceable, valid and credible. Lack of basic data raw data or of some explanation as to why some changes have taken place or why some experiments have been excluded should be taken into consideration.
3. Effects on Quality of Products and Patient Safety: Where the manufacturing or analytic processes can have an impact on product quality, one should evaluate if the batches, samples, or some materials have been affected. The decisions regarding the disposition of the product must be made according to the relevant quality procedures and must be grounded on the evidence available.
4. Effects on Validated/Qualified Status: It has to be determined whether evidence supports the equipment, utilities, processes, cleaning procedure, testing method, or computer-based system.
5. Wider and Historical Impact: Check if such another failure could affect other equipment, shifts, products, protocols, or already completed studies. The recurrence probably indicates existence of some broader weakness, such as those in procedures, training, maintenance, or control strategy.
For every dimension, the rationale must be documented. Stating “no impact on product quality” is simply insufficient unless investigation shows proof supporting the conclusion drawn therein.
Corrective Action, CAPA and Validation Decisions
Correction deals only with the immediate issue whereas corrective and preventive actions deal with the root cause and help reduce their possibility of happening again. Not every deviation needs to initiate a big CAPA, but the process of whether to initiate a CAPA or not should reflect the site procedure and risk assessment.Examples of possible actions are equipment repair or equipment calibration, revision of a procedure, improvement of worker’s training, upgrading of an alarm or regulation, re-engineering of a sampling method and protocol improvement. Each action should be assigned to a responsible person with a due date and documented completion.
Validation team and QA then should assess whether the study can be accepted as carried out, requires some further justified testing, needs full or partial repetition, or does not allow drawing intended conclusions. Further testing should bridge an evidence gap without neglecting unfavourable information.
The validation report should include references to deviation records, investigations and impact assessments summary, CAPA descriptions and conclusion explanations. If the issue of the importance remains unanswered, the report should not state that validation has been carried out successfully without any justifying argument.
What Inspectors Can Investigate
Various matters may be handled by the inspector during the inspection. For instance, inspectors often carry interval checks of validation documents and map the history of deviation according to the genesis of the issue, investigation, impact assessment and ultimate closure. They may investigate why the deviation was qualified as minor, why tests continued even after failure and how the organization managed to prove that the initial validation was not affected.Common problems are shallow root-cause explanations, uneven classification for like events, absence of connections between deviations and validation reports and conclusions focused mostly on acceptable results.
A valuable internal review must confirm that each deviation has an accurate event description, an evidence-based classification system, a documented investigation process, a logical impact evaluation and legitimate disposal of action. Trends have to be monitored so that repetitive failures that are not obvious may be seen.
Get ready to use editable Validation Protocols in MS-Word FormatView List
No comments:
Post a Comment
Please don't spam. Comments having links would not be published.