data integrity blog post cover

Data Integrity in your lab: The 6 steps you can take to avoid an FDA warning

6 min read

Implementing good data integrity practices not only ensures your data is reliable and accurate but is also critical for complying with regulatory bodies. With increasing FDA warning letters triggered by poor data integrity, what steps can your laboratory take to ensure compliance, and avoid being on the receiving end of these letters?

In this article, we will provide a brief overview of data integrity (with references to the FDA Guidance) and share the steps you can take to make sure you can trust your data.

What is data integrity?

Data integrity is not a new concept. However, with the increasing digitalization of data and documents, there is now renewed interest in the importance of integrity means. FDA’s Data Integrity and Compliance with CGMP (current good manufacturing practice) provides a definition that’s often referred to, and is typically interpreted as follows:

“Data integrity refers to the completeness, consistency, and accuracy of data. Complete, consistent, and accurate data should be attributable, legible, contemporaneously recorded, original or a true copy, and accurate (ALCOA).”

  1. Attributable – where records should be attributed to the specific person that collected the measurements or data. For electronic records, attribution can be in the form of an electronic signature.
  2. Legible – where records should be readable, clear, and accessible. For electronic records, legibility also includes the record being in a format that is accessible in the future, such as a PDF file.
  3. Contemporaneous – where records should be documented at the time when the activity is performed.
  4. Original – where the records should be original or a true copy. Electronic copies can be considered true copies as long as it includes the original data, context, as well as metadata.
  5. Accurate – where the records and data should be accurate and free of errors.

These guiding principles are often abbreviated to ALCOA, with various companies offering systems and instruments available to help ensure data complies to the ALCOA framework. Additional principles have more recently been added by WHO to form ‘ALCOA+’, which includes data being complete, consistent, and enduring. In addition to ALCOA/+ and good manufacturing practice (GMP), following good laboratory practice (GLP) is needed to ensure sufficient data integrity.

With the increasing amount of electronic documentation, FDA also provided guidance through 21 CFR Part 11, and outlined the requirements for validation, audit trail, legacy systems, copies of records, and record retention. ALCOA+ and 21 CFR Part 11 are closely intertwined when it comes to ensuring data integrity in a digital environment.

SciNote supports electronic signatures, electronic witnessing, audit trails etc. Get SciNote’s 21 CFR Part 11 guide now:


Why is data integrity important?

Generating good data integrity ensures that the data is reliable and accurate. Failing to meet adequate data integrity can lead to a range of issues, such as incorrect scientific findings and insight, legal implications, and costs in time and expense. The FDA guidance also states that good data integrity is needed “to ensure the safety, efficacy, and quality of drugs, and of FDA’s ability to protect the public health.” Ensuring good data integrity also builds and maintains trust between industry and regulatory bodies, as well as upholding a company’s reputation.

How can poor data integrity arise?

FDA has found an increasing number of data integrity failings and noncompliance during inspections. In addition, the proportion of FDA warning letters issued due to data integrity problems increased from 47% in 2019 to 65% in 2021. These failings are largely the result of not grasping the rationale behind data integrity requirements, not incorporating data integrity practices into day-to-day research activities, and not having a proper oversight structure within the organization.

What steps can you take to manage data integrity?

Adhering to good practice reduces the risk of inaccurate and unreliable results. A variety of approaches can be taken to improve data integrity:

1) Train employees

Training employees on how to identify and prevent data integrity errors can help reduce these errors from occurring and reoccurring. In addition, it is important to ensure that employees face a constructive and fair response when identifying or reporting errors, so that employees are comfortable reporting and tackling errors. A working environment – or a company culture –where employees have fears or concerns about reporting errors to colleagues or management could lead to data being falsified or altered, driving noncompliance.

2) Implement and review audit trails

FDA defines an audit trail as “a secure, computer-generated, time-stamped electronic record that allows for reconstruction of the course of events relating to the creation, modification, or deletion of an electronic record”. Audit trails allow records and modifications of records to be traced; regular reviews of the audit trail can help organizations identify and mitigate risk of errors.

SciNote features for GLP environments

SciNote’s Premium features (Essential, Validated and Platinum plans) and service support the GLP principles in assuring the consistent quality of the software as well as the security and integrity of your research data.

SciNote’s Premium features and service support the GLP/GLP principle

3) Assess data integrity risks

When it comes to assessing data integrity risks, the FDA Guidance suggests risk assessments “should include evaluation of data criticality, control mechanisms, and impact on product quality.” Assessing the risk of potential failings or errors in processes that generate data can help mitigate errors and improve compliance. Data integrity risk assessment tools have been developed, such as the IQ Consortium Data Integrity Risk Assessment Tool.

4) Backup data

The FDA uses the word “backup” to refer to “a true copy of the original record that is maintained securely throughout the record retention period” where backup data “must be exact, complete, and secure from alteration, inadvertent erasures, or loss.” Appropriate systems should be in place so that the original data can be accessed. However, access to changes to data should be restricted to authorized personnel to maintain data integrity.

5) Validation of processes

Validation can have different meanings depending on application or context. Here, the FDA refers to the definition by ICH guidance for industry, which states that “validation” means “providing assurance that a specific process, method, or system will consistently produce a result meeting predetermined acceptance criteria”. Effective validation of equipment, methods or systems mitigates risk of errors occurring, upholding good data integrity. FDA specifies that for validation of the entire workflow, the system must be validated in how it will be used or implemented during testing. Validated systems should also have appropriate controls and restricted access over people who can make changes to settings.

6) Implement a digital system

In many cases, data integrity failings including poor data handling and entry practice, incomplete data sets, unauthorized access, and traceability gaps (e.g., having inconsistent timestamps or revision information) can be attributed to using paper-based systems. Therefore, implementation a digital system, such as an electronic lab notebook (ELN) with data integrity features, could help address the many data integrity challenges a laboratory faces. To this end, look for an ELN with features that will help with maintaining data integrity and will meet requirements such as FDA 21 CFR part 11, including:

  • Built-in audit trail and timestamp to track inputs and revisions
  • Electronic signatures to streamline approval process
  • Automatic back up to prevent loss of data
  • Access control management to restrict data access
  • Additional validation support for GxP requirements

Data integrity will continue to be in the center of attention in the years to come. If you haven’t taken the steps to address data integrity issues in your research process, it is now time to start so you can avoid the dreadful FDA warning letters in the future.

Article by Dr. Brydie Thomas-Moore
Brydie Thomas-Moore is a freelance science/medical writer, information designer, and editor, with a biomedical PhD research background. Brydie has been working on a freelance basis in science communications for 4 years.

Need help with managing your data?

Talk to our team & get the top-rated customer service in the ELN industry.

Sales team