Though always a salient issue, the MHRA’s Data Integrity guidance document further highlights data integrity as a top compliance area cited in FDA warning letters and Form 483 Inspectional Observations. Even with several guidance documents published by various agencies, companies and associated staff continue to struggle with data integrity issues.
Data integrity can be one of the most complex areas of compliance, but a few simple steps can ensure a company maintains fundamental Attributable, Legible, Contemporaneous, Original, Accurate (ALCOA) principles and minimizes the risk of compromised data integrity.
At USDM, we recommend the following action plan:
1. Data Integrity
: Many data integrity issues can be traced back to human error. Therefore, ensuring data integrity starts with the user. Staff should be trained on data integrity, data entry and fundamental ALCOA principles. An active and evolving training program should be in place based on the operational needs of the staff and business processes. Procedures regarding data (i.e. data entry, review and approval) should be easy to understand and available for reference. Programs should also be put in place to ensure System Administrators provide the correct level of user access based on training and role.
2. Understand your Process Workflow and Data Lifecycle
: From sample hand-off to release of results to archival of data, your laboratory process workflow, and consequently your data flow, should be well understood and documented. A simple data map for a laboratory can help consolidate workflows, highlight potential risks, work as a road map for potential process improvements and ultimately mitigate data integrity risks.
3. Automate Data Workflows
: Manual data entry or manual transcription increases data integrity risks and can lead to poor data integrity behavior. For example, data that is jotted down on a sticky not during testing and then transferred to an approved form, or manually entered in an electronic system. The transferred data is not original, contemporaneous, and potentially inaccurate. A data map can be used to identify areas where these types of risks to data integrity exist. Automated solutions should be put in place to ensure real-time data and metadata capture and do away with paper-based or hybrid (paper and electronic) processes.
4. Conduct Quality Reviews
: When data errors go unnoticed, the potential for much larger issues grows. It may seem like an obvious and simple solution, but review of critical data should be conducted by a quality reviewer with expert knowledge of the operation. Reviews should be done in near real-time to ensure data integrity discrepancies are dealt with and escalated prior to larger issues arising.
I discussed Data Integrity trends in a recent webinar
, so make sure to check it out to learn more, and as always, please contact us
with any questions you may have.
Project Manager at USDM Life Sciences Hovsep has over 12 years of experience managing projects in the life science industry, with specific focus on regulatory compliance, validation, equipment lifecycle & sustainability, laboratory operations and data management across all phases of the product life cycle.