Whether on the site or sponsor level, in clinical research we are asked to use Good Documentation Practices (GDP) during the conduct of a clinical trial. One might assume that a quick visit to the FDA website would produce the list of practices. However, there is no Code of Federal Regulations for GDP.
If there is no specific guidance or regulation, why do Sponsors insist on GDP for their research studies? Why is GDP so important in the clinical research world? Most importantly, what are some good documentation practices?
The following are some “good documentation practices” as listed by the Dana-Farber/ Harvard Cancer Center and should be applied throughout the course of a clinical study:
- Maintain adequate records (21 CFR 812.120 (a))
This is sited time and time again in FDA 483 and warning letters: “Failed to prepare and maintain adequate and accurate case histories”, “No documentation of protocol-required procedures: no proof labs were completed”, “Inconsistencies in source and CRFs”
- Source documentation is where the information is first recorded (ICH GCP E6 1.52)
That’s right! It doesn’t matter where it’s written- just where it is first recorded. A colleague recently told me that as a research coordinator she was required to enter all “source” data into the medical records. If it wasn’t in the medical records, it didn’t happen as true source were medical records only. This is not what ICH GCP says. ICH GCP states true source is where the information is first written regardless of if it’s the medical record, a source worksheet or a post-it note.
- Data must be verifiable and follow an audit trail
Documentation should tell the whole story. It should also not contradict any other source data. The following was sited in an FDA warning letter “source documentation and CRFs contain discrepant information.”
- ALCOA- data should be attributable, legible, contemporaneous, original and accurate
This is an important guideline for those in clinical research, and quality assurance professionals. FDA auditors are taught to use this guideline during inspections.
- Corrections to source documents and CRFs should be lined through, initialed and dated (ICH GCP 4.9.3) and never use white out
In a 483 warning letter, one investigator was cited for “CRFs being incomplete with various cross-outs and changes by multiple authors, and an occasional use of white-out.” “Altered source documents with no explanation”
- Never destroy original documents
As sited in a 483 warning letter, “you failed to maintain documents evidencing informed consent.” Remember the statement, “if it’s not documented- it’s not done”- this is the truth! If you are doing something, take credit for your work and document everything. In this case there is no such thing as too much information.
- Keep study records secure yet accessible
It’s important to know who has access to study records. These should be limited to the study research team and kept in a safe area away from public access. Remember anything with patient information is subject to HIPAA and HITECH rules.
In another great article, Good Documentation Practice in Clinical Research suggests a reason for so many FDA 483 warning letters for poor documentation is due to lack of training and experience in good clinical practice and good documentation requirements. IMARC has published a whitepaper on Good Clinical Practice to better understand how to take “good” to “great”!
How does your company or site enforce GDP? Please share your experiences with us?
Photo Credit: Ruth and Dave