1
00:00:12,620 --> 00:00:13,850
>> Christine Gordon:
Hello, my name

2
00:00:13,850 --> 00:00:15,150
is Christine Gordon
and welcome

3
00:00:15,150 --> 00:00:17,300
to an overview
of clinical data management.

4
00:00:18,780 --> 00:00:21,610
An important aspect of
the clinical research lifecycle

5
00:00:21,610 --> 00:00:23,160
is the collection, processing,

6
00:00:23,160 --> 00:00:25,610
storage and sharing 
of data for analysis,

7
00:00:25,610 --> 00:00:27,340
which is the core
of data management.

8
00:00:27,340 --> 00:00:30,140
Today we will focus on clinical
data management practices,

9
00:00:30,140 --> 00:00:31,950
the need to apply
rigorous processes

10
00:00:31,950 --> 00:00:33,849
and procedures
to the handling of data

11
00:00:34,380 --> 00:00:36,620
often requires research
teams and sponsors

12
00:00:36,620 --> 00:00:38,780
to develop a data management plan.

13
00:00:38,780 --> 00:00:40,520
Let's start by looking at
some of the activities

14
00:00:40,520 --> 00:00:41,880
associated with data management

15
00:00:41,880 --> 00:00:43,900
and who is involved
with these activities.

16
00:00:43,900 --> 00:00:45,170
In this first session,

17
00:00:45,170 --> 00:00:46,980
we will discuss what
clinical data management

18
00:00:46,980 --> 00:00:49,960
is and cover what clinical data
management activities are,

19
00:00:50,760 --> 00:00:52,810
as well as cover
what a data management plan

20
00:00:52,810 --> 00:00:54,410
is and its components.

21
00:00:56,510 --> 00:00:58,600
So what is 
clinical data management?

22
00:00:58,600 --> 00:01:00,570
It is a 
multidisciplinary activity

23
00:01:00,570 --> 00:01:02,210
that includes investigators,

24
00:01:02,210 --> 00:01:04,380
research nurses,
study coordinators,

25
00:01:04,380 --> 00:01:06,900
clinical data managers,
database programmers,

26
00:01:06,900 --> 00:01:09,720
bio statisticians,
monitors and support personnel.

27
00:01:10,430 --> 00:01:12,190
Clinical data management
is the activities

28
00:01:12,190 --> 00:01:13,980
involving the 
handling of information

29
00:01:13,980 --> 00:01:16,680
outlined in the protocol
to be collected and analyzed.

30
00:01:18,260 --> 00:01:20,660
A variety of activities make 
up clinical data management

31
00:01:20,660 --> 00:01:22,890
at the site,
as well as for the sponsor.

32
00:01:22,890 --> 00:01:25,660
And these can include data
acquisition or collection,

33
00:01:26,170 --> 00:01:29,320
data abstraction or extraction,
data processing

34
00:01:29,320 --> 00:01:32,770
and coding, data analysis,
data transmission,

35
00:01:32,770 --> 00:01:34,230
data storage and security,

36
00:01:34,230 --> 00:01:36,580
data privacy,
and data quality assurance.

37
00:01:39,000 --> 00:01:42,110
So what is the data
being collected used for?

38
00:01:42,110 --> 00:01:44,650
It may be used to support
the analysis of the primary

39
00:01:44,650 --> 00:01:47,080
or secondary objectives
of the protocol,

40
00:01:47,080 --> 00:01:49,770
safety reporting,
regulatory reporting,

41
00:01:49,770 --> 00:01:52,940
new drug applications,
support for labeling claims,

42
00:01:52,940 --> 00:01:55,440
publications, or even
planning future protocols.

43
00:01:57,840 --> 00:02:00,170
As you know, good clinical
practice guidelines

44
00:02:00,170 --> 00:02:02,620
are an international ethical
and scientific quality standard

45
00:02:02,620 --> 00:02:04,000
for the design,

46
00:02:04,000 --> 00:02:06,610
conduct and record
of research involving humans.

47
00:02:07,150 --> 00:02:09,990
GCP is composed
of 13 core principles,

48
00:02:09,990 --> 00:02:11,630
two of which apply to data,

49
00:02:11,630 --> 00:02:14,370
specifically all
clinical trial information

50
00:02:14,370 --> 00:02:16,320
should be recorded,
handled and stored

51
00:02:16,320 --> 00:02:18,760
in a way that allows
for accurate reporting,

52
00:02:18,760 --> 00:02:20,440
interpretation and verification.

53
00:02:21,060 --> 00:02:23,070
Also, the 
confidentiality of records

54
00:02:23,070 --> 00:02:25,620
that could identify 
subjects should be protected,

55
00:02:25,620 --> 00:02:27,940
respecting the privacy
and confidentiality rules

56
00:02:27,940 --> 00:02:30,750
in accordance with applicable
regulatory requirements.

57
00:02:33,020 --> 00:02:36,640
The FDA Part 11 regulations
address electronic data

58
00:02:36,640 --> 00:02:37,980
and e-signatures.

59
00:02:37,980 --> 00:02:39,930
These regulations
apply to all data

60
00:02:39,930 --> 00:02:42,350
residing both
at the institution site

61
00:02:42,350 --> 00:02:45,790
and at the sponsor site.
Create an electronic record

62
00:02:45,790 --> 00:02:47,520
that will be 
submitted to the FDA.

63
00:02:47,520 --> 00:02:51,200
The scope of these regulations
include validation of databases,

64
00:02:51,200 --> 00:02:53,530
audit trail for 
corrections in the database,

65
00:02:53,530 --> 00:02:56,010
accounting for legacy 
systems and databases,

66
00:02:56,010 --> 00:02:58,440
copies of records
and record retention.

67
00:03:00,590 --> 00:03:03,480
Implementation of new
technologies and systems

68
00:03:03,480 --> 00:03:04,720
to capture, store

69
00:03:04,720 --> 00:03:07,520
and manage data has led to
a growing awareness for the need

70
00:03:07,520 --> 00:03:10,330
to document the process
to ensure the security,

71
00:03:10,330 --> 00:03:13,460
integrity and quality of data
along its lifecycle.

72
00:03:13,460 --> 00:03:16,620
As a result,
the development, implementation

73
00:03:16,620 --> 00:03:18,550
and maintenance
of data management plans,

74
00:03:18,550 --> 00:03:20,210
also known as DMPs,

75
00:03:20,210 --> 00:03:23,100
has grown as good practice
for conducting research

76
00:03:23,100 --> 00:03:26,000
and is commonplace in most
organizations that handle data.

77
00:03:26,720 --> 00:03:29,290
The importance of planning
and documenting data management

78
00:03:29,290 --> 00:03:31,160
has led some 
organizations to require

79
00:03:31,160 --> 00:03:34,210
DMPs as a condition
for funding of grants,

80
00:03:34,210 --> 00:03:37,310
and often sponsors or auditors
will ask to review this document

81
00:03:37,310 --> 00:03:39,870
along with standard
operating procedures.

82
00:03:41,190 --> 00:03:43,260
In addition to regulations
and requirements,

83
00:03:43,260 --> 00:03:45,530
the key to successful study
is how early in the.

84
00:03:45,530 --> 00:03:49,010
Process data planning starts.
The quality of the planning

85
00:03:49,010 --> 00:03:51,600
and the effectiveness
of the implementation.

86
00:03:51,600 --> 00:03:54,730
Data management plans are living
paper or electronic records

87
00:03:54,730 --> 00:03:56,950
that document the 
processes and procedures

88
00:03:56,950 --> 00:03:59,840
to promote consistent, efficient
and effective data management

89
00:03:59,840 --> 00:04:01,610
practices on
an individual study.

90
00:04:02,560 --> 00:04:04,630
The reason why it's important
to provide documentation

91
00:04:04,630 --> 00:04:08,570
on a study level rather than
a department or program level

92
00:04:08,570 --> 00:04:11,710
is because each study has unique
project and data requirements

93
00:04:11,710 --> 00:04:14,640
that should be outlined
and serve as a record of what

94
00:04:14,640 --> 00:04:16,600
and how the data with handled.

95
00:04:16,600 --> 00:04:19,720
The DMP should be designed to
meet the needs of various types

96
00:04:19,720 --> 00:04:23,040
of trials, patient registries,
or other therapeutic areas.

97
00:04:25,090 --> 00:04:28,620
The complexities and nature of
a study can influence the DMP,

98
00:04:28,620 --> 00:04:31,230
but there are minimum standards
recommended by professional

99
00:04:31,230 --> 00:04:34,340
clinical data management
societies such as SCDM,

100
00:04:34,340 --> 00:04:37,430
Society of Clinical Data
Management, and AMIA,

101
00:04:37,430 --> 00:04:40,310
the American Medical Informatics
Association data management,

102
00:04:40,310 --> 00:04:43,770
working groups,
and these should include: 

103
00:04:43,770 --> 00:04:46,560
1) Having a plan in place
prior to the first participant

104
00:04:46,560 --> 00:04:48,200
being enrolled.

105
00:04:48,200 --> 00:04:50,910
2) Assuring that the plan is in
compliance with regulations

106
00:04:50,910 --> 00:04:53,960
and oversight agencies.

107
00:04:53,960 --> 00:04:57,380
3) Identifying and defining 
personnel  and roles involved 

108
00:04:57,380 --> 00:05:00,230
in decision making, data collection, 
data handling and data quality control.

109
00:05:00,230 --> 00:05:02,440
And 4)
assuring data management

110
00:05:02,440 --> 00:05:03,780
processes are described

111
00:05:03,780 --> 00:05:07,410
and defined from study
start to database lock

112
00:05:07,410 --> 00:05:09,040
and steady closure.

113
00:05:09,040 --> 00:05:11,380
It is important that the plan
is current in version

114
00:05:11,380 --> 00:05:14,450
to document process
changes took place.

115
00:05:17,570 --> 00:05:20,020
The type of research,
be it a prospective,

116
00:05:20,020 --> 00:05:22,830
observation, or registry
or therapeutic area

117
00:05:22,830 --> 00:05:25,840
can influence the content
or design of a DMP

118
00:05:25,840 --> 00:05:27,790
because of the variations
in requirements,

119
00:05:27,790 --> 00:05:30,040
types of data and methods
used in collection.

120
00:05:30,580 --> 00:05:33,130
Usually the scope
of a DMP specifies

121
00:05:33,130 --> 00:05:35,140
who is involved
with the data handling,

122
00:05:35,140 --> 00:05:37,520
what the deliverables tasks
and processes are,

123
00:05:37,520 --> 00:05:38,990
and how they are carried out.

124
00:05:38,990 --> 00:05:41,390
Let's look at the different
components of a DMP.

125
00:05:42,650 --> 00:05:45,220
Roles and responsibilities,
maintaining a current list

126
00:05:45,220 --> 00:05:47,550
of all trained 
study team members

127
00:05:47,550 --> 00:05:50,230
who are working with the data
on the study is important.

128
00:05:50,230 --> 00:05:52,980
According to FDA regulations
for electronic records,

129
00:05:52,980 --> 00:05:55,520
access must be limited
to authorized individuals.

130
00:05:56,110 --> 00:05:58,920
Some guidelines recommend
maintaining a list of user roles

131
00:05:58,920 --> 00:06:00,940
and access
to security safeguards

132
00:06:00,940 --> 00:06:03,380
which includes a plan
for removing access.

133
00:06:03,380 --> 00:06:06,160
Organization organizations
may manage this

134
00:06:06,160 --> 00:06:07,630
with a list
of all individuals

135
00:06:07,630 --> 00:06:10,990
who have access to the database
and the dates of their access.

136
00:06:10,990 --> 00:06:12,710
Though is not required
as part of a DMP,

137
00:06:12,710 --> 00:06:14,400
it is strongly recommended,

138
00:06:14,400 --> 00:06:17,120
and some regulatory agencies
require that there is a record

139
00:06:17,120 --> 00:06:18,540
that individuals have the trained

140
00:06:18,540 --> 00:06:21,440
or the skills documented to
perform the task on the trial.

141
00:06:22,500 --> 00:06:25,710
Description of data collected,
some organizations

142
00:06:25,710 --> 00:06:28,030
have a metadata file,
a data dictionary,

143
00:06:28,030 --> 00:06:30,970
or a document listing data
collected for a study.

144
00:06:30,970 --> 00:06:32,260
This will include any data

145
00:06:32,260 --> 00:06:34,330
that is generated or 
received from sources

146
00:06:34,330 --> 00:06:37,160
other than paper or 
electronic data capture forms

147
00:06:37,160 --> 00:06:38,880
completed by the 
principal investigator,

148
00:06:38,880 --> 00:06:41,580
study coordinator, data manager,
or study participant.

149
00:06:42,780 --> 00:06:45,590
The list of standards
or terminology dictionaries.

150
00:06:45,590 --> 00:06:47,910
The use of standard
terminology or dictionaries

151
00:06:47,910 --> 00:06:50,300
usually is in reference
to coding of medications

152
00:06:50,300 --> 00:06:52,870
and adverse events
and user research submission

153
00:06:52,870 --> 00:06:56,320
exchange standards, researchers,
sponsors and regulators

154
00:06:56,320 --> 00:06:58,230
want to know what 
standards and terminologies

155
00:06:58,230 --> 00:06:59,890
are used along
with the versions.

156
00:07:00,690 --> 00:07:03,360
For a long term trial,
versions may change

157
00:07:03,360 --> 00:07:05,740
and it is important
to document all versions.

158
00:07:05,740 --> 00:07:07,180
Because versions
impact the data,

159
00:07:07,180 --> 00:07:08,460
there should be a description

160
00:07:08,460 --> 00:07:10,720
of how the version
was implemented for the study

161
00:07:10,720 --> 00:07:13,770
if the entire database
was versioned up

162
00:07:13,770 --> 00:07:15,490
or only data after certain date.

163
00:07:16,040 --> 00:07:18,640
If using an auto encoder,
it is important to document

164
00:07:18,640 --> 00:07:21,250
the workflow of
how manually coded items

165
00:07:21,250 --> 00:07:23,340
are handled and approved.

166
00:07:23,340 --> 00:07:25,920
How data is collected,
processed, and stored.

167
00:07:26,530 --> 00:07:28,840
Planning how data is 
collected will help ensure

168
00:07:28,840 --> 00:07:30,820
it is collected
in a timely manner.

169
00:07:30,820 --> 00:07:34,450
Documenting if data is captured
using Electronic Data Capture,

170
00:07:34,450 --> 00:07:37,730
EDC, paper,
or directly from a participant

171
00:07:37,730 --> 00:07:40,390
or even downloaded
from a device is important.

172
00:07:40,390 --> 00:07:43,440
For organizations using 
paper to collect data,

173
00:07:43,440 --> 00:07:46,490
data will ultimately be entered
into a sponsor's database,

174
00:07:46,490 --> 00:07:48,760
so determining if single 
entry or double data entry

175
00:07:48,760 --> 00:07:50,680
will be used is important.

176
00:07:50,680 --> 00:07:53,230
Double data entry is entering
the same data twice

177
00:07:53,230 --> 00:07:55,100
to ensure data accuracy.

178
00:07:55,100 --> 00:07:56,940
With the increase
of electronic tools

179
00:07:56,940 --> 00:07:59,470
it should be documented
what sources are electronic.

180
00:07:59,470 --> 00:08:02,570
For example some studies
allow participants to enter data

181
00:08:02,570 --> 00:08:04,550
directly into a tablet

182
00:08:04,550 --> 00:08:07,250
that is automatically
uploaded into a database.

183
00:08:07,250 --> 00:08:10,710
In this case, the data uploaded
is the source data,

184
00:08:10,710 --> 00:08:12,500
and that should be documented.

185
00:08:12,500 --> 00:08:15,160
Listing data transfers
and how that data is processed

186
00:08:15,160 --> 00:08:18,280
or where it is stored once
it arrives should be described.

187
00:08:19,720 --> 00:08:21,420
Data handling rules.

188
00:08:21,420 --> 00:08:23,300
There should be
written procedures on how data

189
00:08:23,300 --> 00:08:25,820
is entered and cleaned 
for data provenance.

190
00:08:25,820 --> 00:08:28,350
To ensure that changes
or self-evident corrections

191
00:08:28,350 --> 00:08:30,140
on case report forms
are documented,

192
00:08:30,140 --> 00:08:32,670
necessary and endorsed
by the investigator.

193
00:08:32,670 --> 00:08:35,320
Examples include conversions
of units of measure,

194
00:08:35,320 --> 00:08:38,370
guidelines for manual queries
or corrections of misspellings.

195
00:08:39,640 --> 00:08:41,420
Data sharing or access.

196
00:08:41,420 --> 00:08:43,850
This component is
fairly new to DMPs,

197
00:08:43,850 --> 00:08:45,480
but is becoming
increasingly more common

198
00:08:45,480 --> 00:08:48,190
because of regulations
requiring data sharing plans

199
00:08:48,190 --> 00:08:51,150
to expedite translational
science and increase knowledge.

200
00:08:53,800 --> 00:08:56,650
Planning data sharing
and access early in the study

201
00:08:56,650 --> 00:08:59,180
can make certain there's
compliance to regulations

202
00:08:59,180 --> 00:09:01,030
and establish 
resources and processes

203
00:09:01,030 --> 00:09:03,310
that will enable
effective data sharing.

204
00:09:03,310 --> 00:09:05,740
The plan should help ensure data
is made available publicly

205
00:09:05,740 --> 00:09:06,970
in a timely manner

206
00:09:06,970 --> 00:09:09,970
and describe the methods used
to provide access to the data.

207
00:09:10,820 --> 00:09:14,290
More extensive DMPs may include
the following components:

208
00:09:14,290 --> 00:09:17,140
form design, edit checks,
system validations

209
00:09:17,140 --> 00:09:20,560
and testing, data flow diagrams,
reporting metrics,

210
00:09:20,560 --> 00:09:24,530
risk analysis, data exchange,
auditing and quality control.

211
00:09:27,300 --> 00:09:30,470
In this session, we learned that
clinical data management

212
00:09:30,470 --> 00:09:32,360
is a multi-disciplinary activity

213
00:09:32,360 --> 00:09:33,910
involving the 
handling of information

214
00:09:33,910 --> 00:09:36,960
to be outlined in the protocol
to be collected and analyzed.

215
00:09:36,960 --> 00:09:38,500
Clinical data
management activities

216
00:09:38,500 --> 00:09:41,480
include data acquisition
and collection,

217
00:09:41,480 --> 00:09:44,990
data abstraction and extraction,
data processing encoding,

218
00:09:44,990 --> 00:09:47,060
data analysis,
data transmission,

219
00:09:47,060 --> 00:09:49,810
data storage and security,
data privacy,

220
00:09:49,810 --> 00:09:51,560
data quality assurance.

221
00:09:51,560 --> 00:09:53,650
Data management plans
are often required

222
00:09:53,650 --> 00:09:55,640
and are living paper
or electronic records

223
00:09:55,640 --> 00:09:57,910
that document the 
processes and procedures

224
00:09:57,910 --> 00:09:59,790
needed to promote
consistent, efficient

225
00:09:59,790 --> 00:10:02,740
and effective data
management practices for study.

226
00:10:02,740 --> 00:10:05,040
At a minimum the 
components of a DMP

227
00:10:05,040 --> 00:10:07,200
should include the roles
and responsibilities

228
00:10:07,200 --> 00:10:10,420
that all study team
members a list of standards

229
00:10:10,420 --> 00:10:14,260
or terminology dictionaries,
how data is collected processed

230
00:10:14,260 --> 00:10:17,960
and stored, data handling rules,
data sharing or access.

231
00:10:18,560 --> 00:10:20,640
A DMP may be more extensive

232
00:10:20,640 --> 00:10:22,490
and include other 
components as well.

233
00:10:24,750 --> 00:10:27,700
Here are two questions to check
your understanding so far.

234
00:10:27,700 --> 00:10:29,730
Can you answer
what activities are part

235
00:10:29,730 --> 00:10:31,650
of a clinical data management

236
00:10:31,650 --> 00:10:33,540
and what is a data
management plan?

237
00:10:36,140 --> 00:10:37,740
Thank you for your time.