Good Documentation Practices

(GDP)

Presented by: Jordan Desilva

www.fda.gov 2
 DEFINING GOOD DOCUMENTATION
PRACTICES

www.fda.gov 3
 Good Documentation Practices
• Good Documentation Practice (GDP) is a systematic
procedure of preparing, reviewing, approving,
issuing, recording, storing and archiving of
documents.
• GDP describe standards by which documents are
created and maintained.
www.fda.gov 4
 The Purpose of Good
Documentation
• Ensures reliable, consistent transfer of information.
• Ensures product quality and safety.
• Complies with regulatory requirements.
• Fulfills the basic premise that good science is reproducible.
• Helps prevent dishonesty and fraud; and is essential for
producing quality results.
• Provides control of processes and improves performance.
• Enables important messages to be communicated clearly and
accurately.
www.fda.gov 5
 What Are Good Documentation
Practices?
• Good Documentation Practices, commonly referred to as GDPs,
are the guidelines that one follows in recording information in a
legible, traceable and reproducible manner.
• A key to Good Documentation Practices is to consider these
questions each time you record your raw data:
1. Is it attributable? 5. Is it accurate?
2. Is it legible? 6. Is it complete?
3. Is it contemporaneous 7. Is it permanent?
4. Is it original?
www.fda.gov 6
 BACKGROUND

www.fda.gov 7
 Background –
Raw Data Definitions
• Good Laboratory Practices (GLP) (21 CFR Part 58):
– Any laboratory worksheets, records, memoranda, notes, or exact copies thereof,
that are the result of original observations and activities of a study and are
necessary for the reconstruction and evaluation of the report of that study. Raw
data may include photographs, microfilm or microfiche copies, computer
printouts, magnetic media, including dictated observations, and recorded data
from automated instruments (21 CFR 58.160.3 (7)).
• Good Clinical Practices (GCP) (VICH GL9 / Guidance for Industry #85):
– Any original worksheets, calibration data, records, memoranda and notes of first-
hand observations and activities of a study that are necessary for the
reconstruction and evaluation of the study. Raw data may include, but are not
limited to, photographic materials, magnetic, electronic or optical media,
information recorded from automated instruments, and hand-recorded
datasheets (Guidance for Industry (GFI) #85, VICH GL9, 1.24).
www.fda.gov 8
 Background –
What are Raw Data?
Both the GLPs and GCPs state how raw data should be
collected
• GLP:
– All raw data generated during the conduct of a study, except those that are
generated by automated data collection systems, shall be recorded directly,
promptly, and legibly in ink. All data entries shall be dated on the date of
entry and signed or initialed by the person entering the data (21 CFR
58.160.130 (e)).
• GCP:
– Raw data whether handwritten or electronic, should be attributable, original,
accurate, contemporaneous and legible (GFI #85, 8.3.1).

www.fda.gov 9
 Background – Original
Documentation is Raw Data
• Similarities between GLP and GCP definitions:
• Any worksheets, records, memoranda and notes of original and first-
hand observations and activities of a study that are necessary for the
reconstruction and evaluation of the study.
• The data collected for a study are there to confirm that the study
protocol was conducted accordingly and support the results and
conclusion of the study.
• Raw data supports any conclusions drawn from the study. High
quality raw data are the strongest support for study conclusions.
• How can raw data collected for a study meet these definitions for a
GLP or GCP compliant study?
www.fda.gov 10
 ALCOA

www.fda.gov 11
 What is ALCOA?
• Raw data should include these attributes:
– Attributable
– Legible
– Contemporaneous
– Original
– Accurate
• Better known as ALCOA.
• If your raw data includes these attributes, they would likely be
compliant with both GLPs and GCPs.

www.fda.gov 12
 Attributable
• Documentation should provide information on:
– what activity is being documented,
– who is performing the activity,
– materials and equipment used in the activity, as well as
– the activities preceding and following the activity.
• Documentation should be traceable to the individual(s) performing an activity.
• Data should be directly recorded into the data capture form (DCF) or in the
electronic data capture system (EDCs), and always sign and date each entry as
required.
• If a procedure requires transcription of the raw data to an EDCs or paper record,
keep the raw data as part of the record.
• GLPs require the individual to initial and date data entries.
www.fda.gov 13
 Attributable
• Should a photocopy of a record be needed, the individual
making the copy will confirm that the copy is a true and
exact reproduction of the original document.
• The copy will be stamped or labeled as “True and Exact Copy”, and
the individual making this confirmation will initial and date the
copy near or in the “True and Exact” stamp.
• Copies of multiple pages should include the “True and
Exact” stamp on the first page, along with initials and date
of the individual who made the copy.
www.fda.gov 14
 Attributable
Signatures and Initials:
• Signatures and/or initials on a document indicates that:
• The individual is authorized to perform, verify, review and/or approve
the activity he/she signed for.
• The individual agrees with the legitimacy of the data content and its
documentation.
• Confirms your unique identity (signatures/initials should be authentic
and traceable to a specific individual).
• Sign and/or initial for yourself only. Never allow someone to sign or
initial any document for you, unless that person is authorized in
writing to do so and according to procedure.
www.fda.gov 15
 Attributable
Signatures and Initials contd.:
• Handwritten and electronic signatures should be unique and legible.
• A signature and/or initials is not valid without a date. Dates should be
documented at the time of signing any record. Multi-page data
capture documentation should have attribution on each page.
• Initials and/or signatures entered into an electronic data capture (EDC)
system should be consistent with log-on credentials to ensure that
electronic records can be attributed to a unique individual.
• An audit trail can serve the purpose of traceability if a history of changes can
be retrieved and viewed as part of the current record.
www.fda.gov 16
 Attributable
• If an EDC system is used for documentation:
– Unique single sign-on identifiers can be used and if
used, should only be used by the individual to which
it is assigned.
• Sign on identifiers should not be shared among personnel
– Assignment of each sign on identifier should be
documented.
www.fda.gov 17
 Attributable Example

www.fda.gov 18
 Legible
• Documentation should be easy to read and
permanent.
• Handwritten entries, signatures/initials and dates on
records should be clear enough to identify the
individual who signed the record and should be legible
to reviewers of the documents.
• The documentation should be clear to limit
misinterpretation of what was performed and
recorded.
www.fda.gov 19
 Legible Example

www.fda.gov 20
 Contemporaneous
• Sign and date records with the date and/or time the data
is collected/entered.
• Data entries, initials/signatures, dates, and/or times into
documentation/records are made as the task is
performed or observed.
• Late entries should be fully justified and supported by
other documents.
• Data entries are not back-dated, post-dated, or pre-
completed.
www.fda.gov 21
 Contemporaneous Example

www.fda.gov 22
 Original
• The documentation that captures the first-time data are
recorded is the original raw data.
• Data entries documented in the appropriate data capture form
or EDCs as the first capture of information.
• If procedures require transcription of the data to a paper record
or data entry into a computer, reference the original raw data to
confirm its accurate via signature/initials and date. If possible,
keep the original raw data as part of the raw data package.
• Original study documentation should be appropriately
maintained.
www.fda.gov 23
 Original Example

www.fda.gov 24
 Accurate
• Documentation contains reliable data and an exact account of events
conducted in an effort to comply with the protocol.
• Data makes sense.
• No approximation, but the actual values.
• Values recorded from an instrument are recorded as presented by the
instrument, i.e units not converted or rounded prior to being recorded
and compliant with the protocol requirements.
• If an instrument is used to collect the data, is the instrument
calibrated and validated as demonstrated by appropriate
documentation?
www.fda.gov 25
 Accurate Example

www.fda.gov 26
 Complete
• All required information should be included in the document,
including all relevant details of description.
• Activities should be completely documented in their entirety
so that events can be reconstructed and understood at a later
time.
• Unused spaces or data entry lines should be marked with a
single line through all the blank spaces, along with “N/A”
written along this line; and initials and date to ensure that
data entries cannot be added to it at a later time.
www.fda.gov 27
 Permanent
• Information cannot be erasable nor be obscured in any way.
• Never use pencil or ink that can be erased.
• Never use correction fluid, correction tape, or cross out entries
so that the original entry becomes unreadable.
• The recorded data should be dark enough to be legible on the
original and any copies.
• Data and the medium it is collected on (paper or electronic)
should be able to be stored and retain its integrity over time.

www.fda.gov 28
 ERROR CORRECTIONS

www.fda.gov 29
 Error Corrections
• How to make corrections
– GLPs state “any change in entries shall be made so
as not to obscure the original entry, shall indicate
the reason for such change, and shall be dated and
signed or identified at the time of the change” 21
CFR 58.160.130 (e)

www.fda.gov 30
 Error Corrections
• Correction of any entry should allow the original
entry to be readable and should provide
information on the date of the change, and the
person making the change.
• Therefore, if a mistake is made, do not correct it
with whiteout or anything else that would hide or
obliterate the original entry.
• Erasures and overwriting are never acceptable.
www.fda.gov 31
 Error Corrections
• The individual who made the mistake should line out the mistake by
putting a single line through the entry, write the corrected information
next to the entry, then initial and date the correction.
• All corrections should be clear and legible.
• Date of the correction should be the date the correction was made;
not the date the error was made.
• Back-dating or post-dating of information is not allowed.
• If there is insufficient room to write the correction next to the entry,
then the footnote method will be used to document the correction.
www.fda.gov 32
 Error Corrections
• Footnotes:
• Are used when there is not enough space/room to complete the correction
of an error that was crossed/lined out.
• Provide explanations/comments when correcting errors.
• Make multiple corrections on the same page of a document.
• Footnotes should:
• Correspond with a unique footnote. However, a footnote can be used
multiple times if it is the same person, same type of error, AND is being
corrected at the same time.
• Be noted on the same day.
• Not be shared with others.
www.fda.gov 33
 Error Corrections
• Example of making a correction using a footnote:
1. Draw a single line through the mistake.
2. Add a footnote (a “1” with a circle around it) next to the mistake.
3. On the same page, find sufficient room to write the correction
and draw the circled numeral you chose earlier.
4. Write the reason for change next to the footnote.
5. Initial and date the correction.

www.fda.gov 34
 Error Corrections
• For electronic records, an audit trail may serve the purpose of
traceability if the history can be retrieved and viewed as part of the
current record.
• Any changes to the data, forms, records/documents after it has been
signed and dated as reviewed, verified, or signed (wet ink or
electronic signature), invalidates the signature. The document should
be reviewed, and/or verified again and re-signed.
• Modification(s) to verified records should be limited to authorized
individuals; documentation of modification to critical data should
include a reason for the change.
www.fda.gov 35
 COMMON GDP ERRORS

www.fda.gov 36
 Common GDP Errors
• Illegible and unclear documentation.
• Use of scrap paper or non-official forms in documentation.
• Failure to maintain original documentation (raw data).
• Obliterations or write-overs.
• Lack of corrections or excessive changes or corrections.
• Use of outdated or uncontrolled forms for documentation.
• Study procedures not initialed and dated by person performing the task.
• Incomplete study records or forms.
• Incomplete explanation of changes to data entries and how the correct data
entry was confirmed.
www.fda.gov 37
 GDP Best Practices
• Full study documentation fully complies with the protocol and
standard of conduct.
• All study documentation was appropriately maintained and will
be able to maintain its integrity when archived.
• Unexpected events and deviations are fully described in the
study documentation.
• Study documentation fully supports all statements and
conclusions in the final study report.
• All study documentation possesses all the attributes of ALCOA.
www.fda.gov 38
 GDP Best Practices
• All study documentation possesses all attributes of ALCOA:
– Each individual that recorded data is clearly identified.
– If someone other than the recorder conducted the observation or observed data
point, that should be documented. The person who performed the task should
review what was recorded for them by the recorder.
– Manually recorded data is recorded clearly and legibly in indelible ink.
– Data is accurately recorded at the time it was performed or observed, including the
date and time.
– The data make sense and any meta data, such as units for values, are included and
documented.
– Narrative documentation of study procedures, events, communication and notes to
file should be clear and provide a complete, accurate description of each occurrence
and be fully attributable.
www.fda.gov 39
 THINGS TO CONSIDER

www.fda.gov 40
 Things to Consider
• Quality Control (QC) procedures should be in place
to ensure data are reviewed in a timely manner to
ensure quality and integrity.
• Facility or study specific correction codes can be
used for reasons for correction.
– If the code does not clearly describe the reason for
change, then additional explanation should be provided
if the code does not clearly describe the reason for
change.
www.fda.gov 41
 Things to Consider
• Raw data collected in an EDC System would also
need to meet the attributes of ALCOA.
– Collecting raw data in an EDC system should mimic how
raw data are collected on paper.
• Individuals recording data should ensure that all
data generated adheres to good documentation
practices or when it does not, is documented as a
deviation.
www.fda.gov 42
 REFERENCES

www.fda.gov 43
 References
• FDA 21 CFR Part 58 (FDA GLPs)
• Good Clinical Practices (GCPs)
• FDA Data Quality Resources Webpage

www.fda.gov 44
 QUESTIONS?

www.fda.gov 45