The document summarizes key topics from Session 1 of a workshop on electronic clinical trials archiving, including: 1) The need to select file formats that are open, ubiquitous, stable and support long-term preservation when archiving electronically. 2) The unique challenges of eArchiving compared to traditional archiving, such as the irrelevant of original/copy and need to retain information on digital formats. 3) Criteria for selecting preservation-suitable file formats, such as open standards, stability and interoperability.

1. Workshop about electronic clinical trials, Session I
Standard based
Electronic Archiving
For Clinical eTrials
e-Clinbical Trials, London, 15.05.2009
Wolfgang Kuchinke, University
Duesseldorf, Dueseldorf, Germany

2. Agenda
Identifying the key
issues of electronic
archiving
01
Data and documents02
Analysis of Trial
Master Files
03
The Archiving
Challenge
04
http://www.free-powerpoint-templates-design.com

3. Session 1
Requirements and challenges of
electronic archiving on the way to
an eArchive

4. Identifying the key issues in
electronic archiving

5. The need for digital preservation
●
Digital storage media are vulnerable to damage or deterioration, and
hardware and software become obsolete, so the time frame for
making preservation decisions is short compared to paper-based
materials.
●
To preserve digital materials the key needs are
• protect the integrity of the data streams
• maintain a way of displaying and using the materials that retains their authenticity.
●
Related needs
• to record enough metadata to manage preservation
• to manage associated rights issues
• to monitor technology for changes that could lead to loss of access

6. Maintain accessibility
●
The key challenge is the future use of materials when technological
changes make original formats and operating environments obsolete
●
Steps that can be taken :
– to retain the hardware and software required to access the materials
– this is only a short-term solution, as hardware will be impossible to maintain indefinitely
●
If possible, produce materials in open, well-supported standard formats
for which software access tools may remain available for an extended
period
●
Transform materials to standard formats. If transforming, retain the
original data in case the chosen migration path becomes unsustainable

7. What is unique in eArchiving
●
eArchiving is essentially identical to traditional archiving with a few
exceptions:
– The concept of “original” and “copy” is irrelevant
– The records can not be inspected by humans without complex
tools
– In addition to retaining media and knowledge on how to interpret
the records, information on digital formats must also be retained
●
Consequences:
– The records are much more vulnerable to loss, since formats have
very short life cycles

8. W. Kuchinke (2009)
Criteria for selecting file formats for long-term preservation
• The following criteria should be considered by data
creators when selecting file formats:
• Open standards
• Ubiquity
• Stability
• Metadata Support
• Feature Set
• Interoperability
• Viability
• The following additional criteria should be considered
for migration:
• Authenticity
• Processability
• Presentation
from: Adrian Brown:
THE NATIONAL
ARCHIVES ,2003t

9. W. Kuchinke (2009)
Open Standards
• Requirement of access to detailed technical information
about the file formats in which records are preserved
• Formats for which the technical specification has been
made available in the public domain are therefore
recommended
• This is invariably the case with open standards, such as
JPEG
• The developers of proprietary formats may also publish
their specifications, either freely (for example, PDF)
• The advantages of some open formats come at the cost
of some loss in structure, context, and functionality
(e.g. ASCII), or the preservation of format at the cost of
processability (e.g. PDF); proprietary formats frequently
support features of their creating software which open
formats do not

10. W. Kuchinke (2009)
Ubiquity
• The format specification should be stable and not
subject to constant or major changes
• New versions of the format should also be backwards
compatible

11. W. Kuchinke (2009)
Stability
Whenever a new
technology comes
up, support for the
older technologies
fades out
• the format specification should be stable and not
subject to constant or major changes
• New versions of the format should also be backwards
compatible
● Open vs. proprietary formats for files and documents
Longevity media
is likely to be
threatened by
obsolescence
before its useful
life is over

12. W. Kuchinke (2009)
Metadata Support
This metadata may
be generated
automatically
• Some file formats make provision for the inclusion of
metadata
• This metadata can have value both during the
subsequent active use of the data and for long-term
preservation (e.g. technical characteristics of the data).
Guidance on the
Provision of
Metadata:
https://www.w3.org/2013/d
wbp/wiki/Guidance_on_the
_Provision_of_Metadata

13. W. Kuchinke (2009)
Feature Set
Example for a full
set of archiving
features: PDF/A
• Formats should support a full range of features and
functionality required for their designated purpose or
business process
• But: the more complex the format, the more costly it
will be to manage and preserve
PDF/A is an ISO-
standardized version
of the Portable
Document Format
(PDF) specialized
especially for the use
in archiving and long-
term preservation of
electronic documents

14. W. Kuchinke (2009)
Interoperability
Includes digital archive
management
standards and IT
solutions, data
exchange in a format
that ensures
interoperability
• Ability to exchange electronic records
• Formats should support a wide range of software or be
platform-independent
• This feature also tends to support the long-term
sustainability of the data by facilitating the migration of
the data from one technical environment to another
Example: AXF. The
archive eXchange
format.Interchange
and Interoperability
for Operational
Storage and Long-
Term Preservation

15. W. Kuchinke (2009)
Viability
Within a vibrant
technical environment,
there is a demand for
long-term viable, open
definition file formats
for archiving
• Some formats provide error-detection facilities, to allow
detection of file corruption which may have occurred
during transmission
• For example, the PNG format incorporates byte
sequences to check for three specific types of error
which could be introduced
Example: hybrid UFF-
DSA format and
CDISC ODM format

16. W. Kuchinke (2009)
Authenticity
Maintaining and
proving authenticity in
the transition from data
and document
producer to the archive
• The format must preserve the content (data and
structure) of the record, and any inherent contextual,
provenance, referencing and fixity information
• ISO 15489: An authentic record is one that can be
proven
• Built-in tests for authenticity
• Known provenance and context
1. Authenticity during
the transfer process
2. Ensuring and
maintaining
authenticity of digital
records preserved in
archive

17. W. Kuchinke (2009)
Presentation
Maintaining and
proving authenticity in
the transition from data
and document
producer to the archive
• The formatting and presentation of a digital record may
have significant information value
• Authenticity of an electronic record may require
preservation of its original ‘look and feel’
• For example, fonts, colours and layout)
Preserving the "look
and feel" of a
document is difficult
for texts, but it is even
more difficult in the
multimedia
environment

18. W. Kuchinke (2009)
Processability
Ensure automatic
processability of the
data, especially during
audits
• Certain data must retain their processability to have
any reuse value
• Even though the requirements of authenticity demand
that the archived version must not be altered through
reprocessing.

19. W. Kuchinke (2009)
Methods of digital preservation
Digital preservation =
formal activity to
ensure that digital
information of
continuing value
remains accessible
and usable
• Technology preservation
• Refreshing
• Migration and reformatting
• Emulation
• Data archaeology
• Output to analogue media
Methods of
preserving data
include refreshing
data and transferring
it to fresh media

20. W. Kuchinke (2009)
Apply traditional archiving processes to eArchive
• Define, identify and document the source organization
and its processes
• Identify information to be archived
• Identify and protect vital records
• Define archive rules and retrieval processes
• Identify ownership and stewardship responsibilities
• Define access rights
• Processes
• Define comprehensive lifecycle
• Define and implement retention schedules
• Requires some processes standardization
For digital archives,
an innovative use of
information
technology is
necessary

21. What is unique in eArchiving
●
eArchiving is essentially identical to traditional archiving with a few
exceptions:
– The concept of “original” and “copy” is irrelevant
– The records can not be inspected by humans without complex
tools
– In addition to retaining media and knowledge on how to interpret
the records, information on digital formats must also be retained
●
Consequences:
– The records are much more vulnerable to loss, since formats have
very short life cycles

22. Archive Life Cycle
Create /
capture
Describe /
Metadata
Use /
manage
Change into
long-term
preservation
DestroyReuse /
audit
Archive

23. W. Kuchinke (2009)
Two paths
• The "Universal Virtual Computer" (UVC) solution
• Preserves the full environment for the future
• Requires understanding of old systems to
understand the data
• Viable for short to medium term
• The "OAIS" solution
• Continuous maintenance of the data to allow
interpretetion with modern systems
• Does not necessarily preserve the full original
environment and context
• Viable for medium to long term
• Supports the UVC solution
ISO 14721:2012
OAIS (Open Archival
Information System)
UVC-based
preservation is an
archival strategy for
handling the
preservation of digital
objects

24. Data and documents
the main dichotomy

25. W. Kuchinke (2009)
Data storage
• The storage of any data within research projects should
be based on two principles (preservation policy):
• digital storage media are inherently unreliable unless
they are stored appropriately
• all file formats and physical storage media will
ultimately become obsolete
• Necessary for long-term preservation of digital data
(from: UK Data Archive - UKDA)
Data storage is the
recording (storing) of
information (data) in a
storage medium.

26. W. Kuchinke (2009)
Data and file formats
• Acceptable data formats for preservation
• Wherever possible, data should be stored in
formats which meet long-term readability
requirements
• In general this means non-proprietary formats or
formats based on open standards
• However, some proprietary formats, like Microsoft's
Rich Text Format and Excel format, are widely used
Your data
stewardship practices
is being dictated by
the types of data you
are working with
Recommended data
formats:
For texts: XML,
PDF/A, HTML, Plain
Text
For databases: XML,
CSV, TAB

27. W. Kuchinke (2009)
Storage media
• The accessibility of any data is dependent on the
quality of the storage media on which they are stored
• The availability of the data-reading equipment
• All optical media (CDs and DVDs) are subject to physical
degradation, and best practice suggests that, wherever
possible, data files should be copied from CDs and
DVDs to a new CD or DVD, between two and five years
after they were first created
• Tapes tend to become demagnetized over time, which
can lead to data loss
• Additionally, it is good practice to check, at regular
intervals, the data files on these discs.
• Magnetic media, like hard drives or tapes, are also
subject to physical degradation
• Best practice suggests a similar migration strategy
It refers to a physical
device or component
in a computing
system that receives
and retains
information relating to
applications and
users
Example: DVD-R
Archival Grade

28. W. Kuchinke (2009)
Supporting Technologies and Applications
• COLD (Computer Output to Laser Disk)
• ERM (Enterprise Reports Management
• OCR/ICR (Optical/Intelligent Character Recognition)
• RAID (Redundant Array of Inexpensive Disks)
• WORM (Write Once Read Many storage)
• Bar code recognition
• Voice Recognition
• Automatic Classification Technologies
• Manual Indexing
• Digital Signature
Also important:
Migration strategy

29. W. Kuchinke (2009)
Supporting Techniques
• Digital Records Management
• Hierarchical Storage Technologies
• Data Warehousing
• Media Management
• Media Migration
• Data Migration Schedules
• Automated Format Transformation Tools
• System Utilities
• Secure data destruction
• PDF (Portable Document Format)
• Standardization
Preservation is a set
of activities aimed at
prolonging the life of
a record, book, or
data while making as
few changes as
possible

30. Evaluation of properties of the most important text
based file: txt, tiff, html, xml, open document, doc
(Mircosoft), PDF
The archival properties of
seven common formats
Rating=high (orange), middle
(yellow), low (green)

31. W. Kuchinke (2009)
Physical conditions
• Most researchers do not need to maintain data storage
systems
• However, areas and rooms designated for storage of
digital or non-digital data should be suitable for the
purpose for which they are being used.
• The conditions under which data are stored significantly
affect their longevity.
• Optical media, like CDs and DVDs, are vulnerable to:
poor handling; changes in temperature; changes in
relative humidity; air quality; and lighting conditions.
• Magnetic media, like hard drives, are equally sensitive
to their physical environment
Factors to consider
when using an
archiving room: fire,
water, emission, theft,
pollution

32. Recommended Data Formats for Preservation
Purposes in the Florida Digital Archive
Formats for text files

33. W. Kuchinke (2009)
Archival lifespan of CD and DVD
● Cyanine-based CD-R discs have about 70 years (based on
accelerated aging tests) (TDK)
● Under recommended storage conditions, CD-R, DVD-R, and
DVD+R discs should have a life expectancy of 100 to 200 years
● Best quality CD-ROM media would be suitable for storing
information for 50 until 100 years (National Media Laboratory
NML) under normal conditions

34. W. Kuchinke (2009)
Reasons for good or poor CD-R permanence
● Plastic (polycarbonate substrate) is oxygen permeable,
aluminum corrodes when exposed to oxygen and silver corrodes
or tarnishes when exposed to sulfides in the air. This could
happen in as little as two years with poor CD-R’s.
● The Dyes used in the recording layer are light sensitive and will
react to ambient light and fade over time. Quality CD-R’s use a
dye that resists fading.
● Humidly may seep through a poor lacquer coating.
● A scratch on the base side can be repaired, but a scratch on the
lacquer side renders the CD useless. Quality CD-R’s have a
thick protective coating to resist scratches

35. W. Kuchinke (2009)
Data security
● Data security is the practice of protecting digital information from
unauthorized access, corruption, or theft throughout its entire
lifecycle
● The core elements of data security are confidentiality, integrity,
and availability
● Don't configure archive storage in a way that could potentially
become a single point of failure
● Data security is the protection of any data from unauthorized
access, use, change, disclosure and destruction
Don't treat all of your
data the same with
regards to archival

36. W. Kuchinke (2009)
Confidentiality
● Risks of identification of personal information are typically
contained by the anonymisation of data and the provision of
access through a dedicated rights management framework
● Legally, data which contain personal information must be treated
with more care than data which do not.
● Signed consent forms that usually contain identifying information
should be stored separately from the data
Recognize and
discuss legal issues
relating to privacy
and confidentiality
issues in general and
for digital archives in
particular

37. Recommended Data Formats for Preservation
Purposes in the Florida Digital Archive
Formats for text files

38. W. Kuchinke (2009)
Common causes of archive object loss
● People:
● Theft, damage, corruption
● Natural disasters
● Fire, flood, earthquake
● Erosion
● Rot, loss of meaning
● Mismanagement
● Loss of authenticity, lost in a shuffle

39. W. Kuchinke (2009)
Archiving planing process
Define motivation for archiving
Need for GCP
compliance
Identify objects to archive TMF
Determine how long to keep 10 years
Determine when objects will not be needed End of clinical tudy
For clinical trials

40. W. Kuchinke (2009)
Archiving planing process
Determine how to dispose of objects destruction
Determine who has access to archive Archive representative
Determine form of archive objects ODM file
Determine place of archive Archiving room

41. W. Kuchinke (2009)
Archiving planing process
Determine operational processes Archiving SOP
Determine administrative processes Archiving SOP
Determine change processes Archiving SOP

42. W. Kuchinke (2009)
Archiving planing process
EDC
data management
system op DB
Remains
in the
system
EDC
data management
system op DB arch DB
EDC
data management
system op DB
EDC
data management
system op DB
archiving
system
archive
room

43. The Trial Master File (TMF) the
central archival object
Analysis of TMFs

44. W. Kuchinke (2009)
What is the TMF
The TMF should be
held at the
coordinating site
(usually the Study
Coordinating Centre.
For multi-site trials,
copies of relevant
documents should be
kept at each
participating site in an
Investigator Site File
(ISF).
• A trial master file contains the essential documents for
a clinical trial that may be subject to regulatory agency
oversight (audits)
• In the European Union (EU), TMFs have a different
definition and set of requirements than in the US
• The EU Commission's Directive 2005/28/EC 63 Chapter
4 states 'the trial master file shall consist of essential
documents, which enable both the conduct of a clinical
trial and the quality of the data produced to be
evaluated
• A Trial Master File (TMF) should be set up at the
beginning of a trial
• The essential documents that make up the file should
be kept in a secure but accessible manner.

45. W. Kuchinke (2009)
Archivierung von Investigator Initiated Trials (ISS)
Investigator initiated
studies (IIS) are
clinical studies initiated
and managed by non-
pharmaceutical
company researchers,
like individual
investigators or
institutions. The
researcher
(investigator) is
responsible for all
legal and regulatory
responsibilities of the
trial sponsor.
• Focus is on IIS
• Implementation of all regulations of GCP can be a
problem for the academic field
• Often no awareness for the difficulty of archiving exists
• Considerable resources for paper based studies are
necessary (e.g. archiving room, archiving
representative, security measures against fire and
burglary)
• University clinics are assuming sponsorship for
academic studies (IIT) (responsibility for archiving)
• Archiving stands for: the safe-keeping also for longer
periods of data and documents and keeping them
accessible, including source data, lab reported, medical
record, site file, identification lists, pseudonymisation
lists, …

46. W. Kuchinke (2009)
Analysis of TMFs
Investigator initiated
studies (IIS) are
clinical studies initiated
and managed by non-
pharmaceutical
company researchers,
like individual
investigators or
institutions. The
researcher
(investigator) is
responsible for all
legal and regulatory
responsibilities of the
trial sponsor.
• An analysis of different TMFs was carried
• The result was that the following types of documents
had often been generated in clinical studies:
• forms, permits, applications, test plans, SOPs
(Standard Operating Procedures), instructions,
labels, receipts, curriculum vitae, calculations, plans,
documentation, correspondence, protocols, reports,
lists, CRFs (Case Report Forms), invoices / receipts,
certificates, contracts, notes / memos, manuals.
• Three representative studies were selected, a small,
medium and large clinical study
• The corresponding TMFs owere examined in detail
• The large and medium-sized studies use electronic data
capture (eCRF), while patient data in the small study
are recorded on paper CRFs

47. W. Kuchinke (2009)
Analysis of TMFs - Categories
Investigator initiated
studies (IIS) are
clinical studies initiated
and managed by non-
pharmaceutical
company researchers,
like individual
investigators or
institutions. The
researcher
(investigator) is
responsible for all
legal and regulatory
responsibilities of the
trial sponsor.
• The variety of documents in a TMF made it necessary to
introduce further classification criteria
• Since paper-based documents are also used and stored,
criteria such as the paper format, number of pages,
type of stapling, etc. has to be considered
• The presence of a signature played an important role
• In the case of electronic documents, the Signature Act
distinguishes between different forms of electronic
signatures
• simple electronic signature
• advanced electronic signature
• qualified electronic signature
• In general, simple signatures can be carried out from
the corresponding EDC (Electronic Data Capture)
programs when filling out the CRFs using a password

48. W. Kuchinke (2009)
Categories of documents in TMF
Document
personal reference documental reference GCP relevance
Personally Pseudonymouse nonee Original Print out Essential Optional
Reference
Study
Organisation
Patient
Originator
Sponsor
Investigator
Ethics
Commission
Pharmacist
Leading
Investigator
statistician
Type of
document
Form
Certificate
Approvals
Investigator's
Brochure
Clinical Study
Protocol
Case Report
Forms (CRF)
Medium
Text
Table
Graphic
Photo
Type of Sig.
Signature
Sig. by
Investigator
Pharmacist
Leading
Investigator
Insurer
Patient
Authority
handwritten digital
Document
creation
Paper digital

49. Hendley (1998) has analysed a
large number of document
categories and the
corresponding formats.
Types of documents in
general
Document Categories (after Hendley)
data (alphanumeric)
structured text (alphanumeric, MarkUp, XML,...)
Office documents (alphanumeric text, graphic)
design documents (raster- and vector graphic)
presentations
pictures / photos
sound
video
interactive multimedia

50. Hendley (1998) has analysed a
large number of document
categories and the
corresponding formats.
Types of documents in
general
Document Categories (after Hendley)
data (alphanumeric)
structured text (alphanumeric, MarkUp, XML,...)
Office documents (alphanumeric text, graphic)
design documents (raster- and vector graphic)
presentations
pictures / photos
sound
video
interactive multimedia

51. The standard IEC 61355-1
Classification and designation
of documents for plants,
systems and equipment
describes rules and
guidelines for the uniform
classification and
identification of documents
based on their characteristic
content of information.
ISO document types Document types (ISO EN 61355)
lists
reports
correspondence
training documents
calculations
contractual documents
order- / delivery documents
data sheets
explanatory documents
guidance / handbooks
diagrams
QM reports
order documents / SOPs

52. W. Kuchinke (2009)
TMF analysis - example

53. W. Kuchinke (2009)
Composition of different TMFs
Considerable
differences were found
in the TMFs of
different clinical trials
• The volume of site-specific documentation is also
strongly influenced by
• the number of patients recruited
• the number of adverse event reports and the
number of queries
• The survey showed that using the 15 categories one
can distinguish about 190 different types of documents
that are kept in a TMF, about half can have a signature
• Many documents are generated by the study
members electronically and then printed
• Their most important formats are: PDF, doc, TIFF
(scanned documents), OpenDoc
• The study database is stored as XML, access, SAS,
CSV, PDF. Documents may contain handwritten
notes, even stickers were used, which impedes
digitalisation.
●
190 different types
of documents
●
about half of them
can have a signature

54. W. Kuchinke (2009)
Composition of different TMFs
• TMF contains: documents which individually and
collectively permit evaluation of the conduct of a
clinical trial and the quality of the data produced are
defined as essential documents according to
CPMP/ICH/135/95
• They are filed in a way that facilitates the management
of the clinical trial, audit and inspection (Trial Master
File) and comprise diverse documents like:
• insurance statements, signed protocol and
amendments, informed consent forms, advertisements
for subject recruitment, signed agreements between
involved parties, approval of ethics committee,
composition of ethics committee, shipping records, CVs
of investigators, normal ranges for laboratory
procedures, etc.
●
ICH E6 (R2) Good
clinical practice. This
document addresses
the good clinical
practice, an
international ethical
and scientific quality
standard. Replaced
Note for Guidance
on Good Clinical
Practice
(CPMP/ICH/135/95)

55. W. Kuchinke (2009)
Composition of different TMFs
• To analyze the documentation of TMFs 15 categories for
classification were developed (person reference, GCP
relevance, document generation, document format,
signature, medium, etc.). TMFs of three clinical trials
were analyzed:
• a small trial (4 sites, 40 patients, 34 pages/CRF,
paper-based CRF),
• a medium-sized trial (7 sites, 40 patients, 65 pages/
CRF, eCRF)
• a big trial (17 sites, 142 patients, 23 pages/CRF,
eCRF).
• Although, the essential study documents are clearly
defined by GCP requirements, depending on the nature
of the clinical study, the different parts of the TMFs can
vary greatly in size. Of the six different parts of TMFs
the site-specific part grows with the number of
participating study centres.
TMFs of three
clinical trials were
analyzed
Different parts of the
TMFs can vary
greatly in size

56. W. Kuchinke (2009)
Composition of different TMFs
• Our analysis showed that there were not so much
differences in the total amount of paper documents
produced by the three exemplary trials:
• 5224 pages for the small trial, 5062 pages for the
medium trial, and 4213 pages for the big trial
• but additional factors which contributed strongly to
the generation of study documents
• In the number of documents of the small trial about
1400 pages of paper CRFs are included, both other
trials used eCRFs
• Another factor is the working habit of investigators and
other participants
• The parts “correspondence” and “drafts, protocols,
other documents” were the smallest in the big study
with 133 pages and 265 pages and the biggest in the
medium study with 452 pages and 1467 pages
A large trial doesn‘t
mean automatically
a large TMF

57. W. Kuchinke (2009)
Analysis result: page numbers of three TMFs

58. W. Kuchinke (2009)
Result of analysis
• Example studies
• M (7 sites, 40 patients, 65 pages CRF),
• K (4 sites, 40 patients, 34 pages CRF),
• G (17 sites, 142 patients, 23 pages CRF)
• Sum (number of pages), 4213(G), 5224(K), 5062 (M)
• K: contains about 1400 pages of completed CRFs (Case
Report Forms).
• Both other studies have eCRF, in each case burned on
CD
• The number of CRFs can constitute a large part of the
study documentation, because it increases with the
number of patients. One can assume that a CRF amount
to about 40 - 100 pages
The small study K:
contained about
1400 pages of
completed CRFs

59. W. Kuchinke (2009)
Archiving of „Investigator Initiated Trials“
• Implementation of all regulations for GCP is a problem
for the academic field
• Often no awareness for the difficulty of archiving exists
• Considerable resources for paper based studies are
necessary (e.g. archiving room, archiving
representative, security measures against fire and
burglary)
• University clinics are assuming sponsorship for
academic studies (IIT) (responsibility for archiving)
• Archiving stands for: the safe-keeping also for longer
periods of data and documents and keeping them
accessible, including source data, lab report, medical
record, site file, identification lists, pseudonymisation
lists, …
In Investigator-
Initiated Trials (also
called non-
commercial trials),
the leading
investigator usually
establishes only one
file for collecting all
study documents,
the TMF.

60. W. Kuchinke (2009)
Archiving of „Investigator Initiated Trials“
• Implementation of all regulations for GCP is a problem
for the academic field
• Often no awareness for the difficulty of archiving exists
• Documentation is still mainly paper based
• Considerable resources for paper based studies are
necessary (e.g. archiving room, archiving
representative, security measures against fire and
burglary)
In Investigator-
Initiated Trials (also
called non-
commercial trials),
the leading
investigator usually
establishes only one
file for collecting all
study documents,
the TMF.
Archiving stands for: the
safe-keeping also for
longer periods of data
and documents and
keeping them accessible,
including source data, lab
report, medical record,
site file, identification
lists, pseudonymisation
lists, …

61. W. Kuchinke (2009)
Electronic signatures
• Electronic signatures may be created with increasing
levels of security and validity
• Advanced electronic signature
• The signatory can be uniquely identified and linked to the
signature
• The signatory must have sole control of the private key that was
used to create the electronic signature
• Qualified electronic signature
• It is difficult to challenge the authorship of a document signed
with a qualified electronic signature
• The signature is non-repudiable
• Using a digital certificate, which has been encrypted through a
security signature-creating device
• It is authenticated by a qualified trust service provider.
In Investigator-
Initiated Trials (also
called non-
commercial trials),
the leading
investigator usually
establishes only one
file for collecting all
study documents,
the TMF.

62. W. Kuchinke (2009)
Components of a database archiving solution
Important documents
with signatures
Documents
Electronic documents
Archive of legally
important paper
documents with
signature
Archive
repository
digitalisation
eSig

63. The Archiving Challenge
Archiving of data data and documents together
until a digital archive is prevalent

64. W. Kuchinke (2009)
Digital archiving
• Digital archiving is using a repository of digital material
that has to be kept for a longer period of time
• It stores collections of digital information such as
documents, video, pictures, etc in a digital format with
the intention of providing long-term access to the
information
• A digital archive can be an elaborate collection with a
multi-tiered storage system or located on a hard drive.
• Long-term preservation of appropriate materials should
be based on “evidential, legal, contextual or information
value”
• GCP requirements also apply to digital archives
• The main component is the archiving of the clinical trial
database

65. W. Kuchinke (2009)
Components of a database archiving solution
Database archiving
means the removing
of selected data
objects from an
operational
database that are
not referenced any
more and to place
them in an archive
data store, where
they can be
accessed if needed.
Archive application
data gathering
Archive application
design
Archive
data extraction
Archive data
management
Archive access
Archive administration
From: Olson, Database archiving, 2009
Archive
Data store
Archive
repository
Archive repository is
a place in which
records with archival
value are
permanently
preserved.

66. W. Kuchinke (2009)
Components of a database archiving solution
Database archiving
means the removing
of selected data
objects from an
operational
database that are
not referenced any
more and to place
them in an archive
data store, where
they can be
accessed if needed.
Archive application
data gathering
Archive application
design
Archive
data extraction
Archive data
management
Archive access
Archive administration
From: Olson, Database archiving, 2009
Archive
Data store
Archive
repository
Archive repository is
a place in which
records with archival
value are
permanently
preserved.

67. From C. Dollar: 2001 PDA
Good Electronic Records
Management Conference
The three main components of a
document record
Content
• The “Data” (as entered, pre-filled, etc.)
Structure
• The Proxy for Comprehending the Record (form,
format, such as the signature, logo, graphics, etc.)
Context
• circumstances; the “what” or “why” of the transaction
( contract, approval , etc.)

68. From C. Dollar: 2001 PDA
Good Electronic Records
Management Conference
The three main components of a
document record
Content
• The “Data” (as entered, pre-filled, etc.)
Structure
• The Proxy for Comprehending the Record (form,
format, such as the signature, logo, graphics, etc.)
Context
• circumstances; the “what” or “why” of the transaction
( contract, approval , etc.)

69. Lifecycle of data
From: Olson:
Database Archiving,
MK, 2009
discard
operational
phase
active data
reference
phase
semiactive
data
archive
phase
inactive
data
create

70. Data Life Cycle Management
From: Data vs.
Document – What’s
the Difference? C.
Dollar: 2001 PDA
Good Electronic
Records Management
Conference
Retention Metadata
Receipt
Active
Use
Semi-
active/
Creation
DisposalArchival
Storage
Pre-archival
Start Date
Archive
Date
Purge
Date
Period
= n yrs.

71. Integrated Clinical Data Archive
From C. Dollar: 2001
PDA Good Electronic
Records Management
Conference
●
Conceptual Model
Data
Digital
Tracking and preservation
control application
Identification data
Retention data
Preservation data
Location of archived
information set
Audit Trail
Hard
copy
Management Metadata
Archived Clinical Record
Package
Metadata

72. W. Kuchinke (2009)
Electronic Data Records Management
• Regulatory Requirements
• FDA - 21 CFR Part 11: Effective 8/20/97: PART 11
ELECTRONIC RECORDS; ELECTRONIC SIGNATURES
• Electronic documents AND data must be retained where
used as “official records” during the drug development
process, including electronic audit trails.
https://www.accessdat
a.fda.gov/scripts/cdrh/
cfdocs/cfcfr/CFRSearc
h.cfm?CFRPart=11

73. W. Kuchinke (2009)
Persistence of Electronic Clinical Trial Data
• Periodic transfer of data fo new storage media
• Capture and storage of metadata
• Interoperability through technology neutral format(s)
• Likely to be costly

74. W. Kuchinke (2009)
Persistence of Electronic Clinical Trial Data
• Periodic transfer of data fo new storage media
• Capture and storage of metadata
• Interoperability through technology neutral format(s)
• Likely to be costly

75. W. Kuchinke (2009)
Persistence of Electronic Clinical Trial Data
• Periodic transfer of data fo new storage media
• Capture and storage of metadata
• Interoperability through technology neutral format(s)
• Likely to be costly

76. Archive dichotomy - Two types of archiving
documents database
Clinical study
CRF
Study protocol
Contracts
Insurance
Signatur
Scannen
eCRF
e-Study protocol
TMF

77. W. Kuchinke (2009)
Database archiving
• Archiving of data objects
• Data objects: a unit of information that someone in the
future will seek from the archive (represents a business
event or object)
• Creation: data object is created at a specific point in
time
• Operational: data object is used in executing other
business transactions (e.g. data is used to create
other data objects)
• Reference: a data object enters the reference phase
when it has no potential to participate in other
transactions
• Not expected to be referenced: there are no business
processes that will need to use this data
• Data archiving is the practice of identifying data that is
no longer active and moving it out of production
systems into long-term storage systems.

78. W. Kuchinke (2009)
Database archiving
• Need for archiving of data objects
• Data objects is a unit of information that in the future
will be searched for in the archive
• Creation: data object is created at a specific point in
time
• Operational: data object is used in executing other
business transactions
• Reference: a data object enters the reference phase
when it has no potential to participate in other
transactions
• Not expected to be referenced: there are no business
processes that will need to use this data
• Data archiving is the practice of identifying data that is
no longer active and moving it out of production
systems into a long-term storage system
A Data object
represents a
container for data
values, a place
where data values
may be stored and
later retrieved

79. W. Kuchinke (2009)
Database archiving
• Archiving: begins when the data object is not expected
to participate in operational transactions and are not
expected to be referenced
• This is the case when the data object leaves the data
management system after the end of a clinical trial
• Discard: removal of the data object from the electronic
systems

80. W. Kuchinke (2009)
Two forms of storage of clinical trials data
• Archiving of all essential documents in the Trial Master
File (TMF) or the Investigator Site File (ITF)
• Mostly paper
• Archiving of the study database
• Mostly in a proprietary format, or as XML, CSV
• A CSV file typically stores tabular data (numbers and
text) in plain text, in which case each line will have
the same number of fields.
• eCRFs as XML or PDF

81. W. Kuchinke (2009)
Two forms of storage of clinical trials data
• The problem of electronic archiving of clinical studies
joins the spheres of document oriented archival, as can
be found in clinical care (e.g. archiving of electronic
patient records, EPR) with the archiving of research
databases (life science and medical research)
• Unfortunately, from both areas, solutions for the
archiving of clinical trials can be found only to a limited
degree, because in clinical research specific conditions
prevail
• On the one hand only the necessity for a rare access
to the archived trial data due to audits by authorities
exists
• On the other hand there is the need to store a
multitude of different data and documents and to
observe in this context strict laws and regulations
(e.g. AMG, GCP, FDA)

82. W. Kuchinke (2009)
Archiving as part of „Document Life Cycle“
●
Create The making of an analog or digital document
●
Capture Converting or digitizing physical or analog
documents into digital format
●
Index Cataloging process or creating metadata via manual or
automatic methods
●
Manage Managing and storing documents for rapid
accessibility
●
Access Searching processes to find documents using
classification, meta-data, fulltext or other search technologies
●
Retrieve Viewing documents from search results
●
Administer Managing users, resources, content types and
structures
●
Share and Collaborate Sharing, collaborating and modifying
in a work group via real time and non-real time methods
●
Archiving for long-term duration

83. W. Kuchinke (2009)
Access to archived material The archived
material must be
provided for the
cases of audits and
inspections be
authorities

84. W. Kuchinke (2009)
At the end
Ensure automatic
processability of the
data, especially during
audits
• Data storage is different from data archiving
• Data sharing must be handled before the archiving of data
• Preparation of data archiving should begin already at the
data generation stage
• A completely digital TMF should be created, eliminating any
need for paper storage
• For data archiving the most important requirement is that
data is accessible and readable even after very long time
spans of 20 and more years, when the underlying technology
will have completely changed
• Thus, archiving must be done independently from
technological developments
• Qualified electronic signatures may be a way to guarantee
that documents are not corrupted after such long time spans

85. Thank You!
Contact:
Wolfgang Kuchinke, Heinrich-Heine University, Duesseldorf, Germany
wolfgang.kuchinke@uni-duesseldorf.de
wokuchinke@outlook.de
More information about archiving of clinical trials: http://www.tmf-ev.de
This presentation contains
additional explanatory
material for the workshop