1
00:00:09,075 --> 00:00:10,143
Hello everyone.

2
00:00:10,143 --> 00:00:14,981
My name is Marjorie Good, or Marge Good,
I go by.

3
00:00:14,981 --> 00:00:20,854
I am a nurse consultant with the Division
of Cancer Prevention within NCI.

4
00:00:20,854 --> 00:00:26,292
And in that role, I'm a nurse consultant
working with NCI grant-funded,

5
00:00:26,292 --> 00:00:31,731
community-based research programs of which
there are about 46 across the U.S.

6
00:00:31,731 --> 00:00:34,467
now and one in Puerto Rico.

7
00:00:34,467 --> 00:00:40,340
And I actually came from a community-based
program, which has really helped me

8
00:00:40,340 --> 00:00:46,246
with this new role to help new sites
that are coming on board.

9
00:00:46,880 --> 00:00:49,582
I was with the program in Wichita,

10
00:00:49,582 --> 00:00:53,486
Kansas, actually,
and did that work for over 20 years,

11
00:00:53,486 --> 00:00:57,757
managed the research program there,
which was quite busy and active.

12
00:00:57,757 --> 00:01:01,227
We actually averaged
usually between 600 and 800 patients

13
00:01:01,227 --> 00:01:04,330
enrolled in clinical trials
in a year's time.

14
00:01:04,330 --> 00:01:05,899
So we were busy.

15
00:01:05,899 --> 00:01:10,537
Was heavily involved in clinical research
and with that I came away

16
00:01:10,537 --> 00:01:16,376
with a lot of knowledge that I like
to share with as many as possible.

17
00:01:16,376 --> 00:01:20,013
So today is one of those efforts
to let you know

18
00:01:20,013 --> 00:01:24,684
a little bit more about data management
and the development of case report forms.

19
00:01:24,684 --> 00:01:29,689
And I know that there's a varying
degree of expertise, as well as job roles.

20
00:01:29,689 --> 00:01:32,992
I think that we'll be participating
in this training, so

21
00:01:32,992 --> 00:01:37,664
I kind of tried to base this around
that -- with that in mind.

22
00:01:37,997 --> 00:01:41,668
So some of it may seem
really basic to others

23
00:01:41,668 --> 00:01:45,672
and maybe a little bit above standards
for others, as well.

24
00:01:45,672 --> 00:01:49,142
So I'm happy to take questions at the end.

25
00:01:49,142 --> 00:01:54,447
So today I hope we can discuss the
importance of the proper data collection

26
00:01:54,447 --> 00:01:58,451
efforts, identify the types of data
collected for clinical trials, list

27
00:01:58,451 --> 00:02:03,590
potential sources of the -- can be used
for source documents for that data

28
00:02:03,590 --> 00:02:07,594
collection, name key factors to consider
during case report form development.

29
00:02:07,594 --> 00:02:11,631
Discuss things that we need
-- really I think are important

30
00:02:11,631 --> 00:02:13,800
to consider when you're developing CRFs.

31
00:02:13,800 --> 00:02:17,604
And I add a little more information
with this one from my last presentation

32
00:02:17,604 --> 00:02:20,540
as a request of some of the comments
after the presentation.

33
00:02:20,540 --> 00:02:23,776
So hopefully that additional information
will be informative for you, as well.

34
00:02:23,776 --> 00:02:27,547
I also want to then cover what constitutes
a poorly designed case report form.

35
00:02:27,547 --> 00:02:31,351
So hopefully it'll give you -- inform
you not to go down that road.

36
00:02:32,318 --> 00:02:33,553
Then we want

37
00:02:33,553 --> 00:02:37,624
to talk about data management itself,
which is quite large.

38
00:02:37,624 --> 00:02:42,962
And within that, it's going
-- I like to review the audit process

39
00:02:42,962 --> 00:02:47,433
as themselves, describe adverse
event reporting, and then describe some of

40
00:02:47,433 --> 00:02:51,304
-- very lightly on regulatory
requirements for data collection.

41
00:02:51,304 --> 00:02:55,208
So the use of data is very,
very important.

42
00:02:55,208 --> 00:02:59,279
Data analysis and reporting,
the actual capturing of the data

43
00:02:59,279 --> 00:03:02,549
is what informs
the analysis and the reporting,

44
00:03:02,549 --> 00:03:06,219
the publication that goes on
after the study's completed.

45
00:03:07,053 --> 00:03:11,124
It's a method of tracking the patient
or the subjects.

46
00:03:11,124 --> 00:03:13,560
It's used for FDA safety monitoring.

47
00:03:13,560 --> 00:03:16,396
It's used in new drug application
submissions.

48
00:03:16,396 --> 00:03:18,431
It supports those labeling claims

49
00:03:18,431 --> 00:03:22,502
for new drugs
that have gone through the FDA process.

50
00:03:22,502 --> 00:03:25,371
It's reviewed by data safety
monitoring boards

51
00:03:25,371 --> 00:03:28,608
to assess the trial safety and ongoing
activity.

52
00:03:28,608 --> 00:03:31,477
It's used in publications
and medical journals,

53
00:03:31,477 --> 00:03:35,548
and it definitely helps
to inform future development of trials.

54
00:03:37,250 --> 00:03:39,085
So the actual data management reporting.

55
00:03:39,085 --> 00:03:42,722
It's the outcomes of the clinical trial
are dependent on this data

56
00:03:42,722 --> 00:03:45,758
that is collected,
and it needs to be collected accurately,

57
00:03:45,758 --> 00:03:47,260
and in a timely manner.

58
00:03:47,260 --> 00:03:48,795
It needs to be verifiable.

59
00:03:48,795 --> 00:03:52,432
So those are those things we're going
kind of talk about today.

60
00:03:52,432 --> 00:03:55,802
Data must also reflect the actual aims
of the clinical trial,

61
00:03:56,769 --> 00:03:58,972
and it must comply with regulatory
agencies.

62
00:03:58,972 --> 00:04:01,207
And really, all of this is key.

63
00:04:01,207 --> 00:04:04,978
You're going to get all this information
if you have adequately designed

64
00:04:04,978 --> 00:04:05,912
case report forms.

65
00:04:05,912 --> 00:04:09,382
So basically, there's this old adage
that I hear still frequently.

66
00:04:09,382 --> 00:04:10,984
It's garbage in, garbage out.

67
00:04:10,984 --> 00:04:13,486
If you don't have good case report forms,

68
00:04:13,486 --> 00:04:17,590
you're not going to get the data
and the information that you need.

69
00:04:18,358 --> 00:04:21,894
The research teams play
a really vital role in this process.

70
00:04:21,894 --> 00:04:26,065
They can be made up of such levels
I've listed on this slide.

71
00:04:26,065 --> 00:04:28,001
The principle investigator,
clinical research nurse,

72
00:04:28,001 --> 00:04:31,204
or CRA, the data manager, data base
administrator, and statistician.

73
00:04:31,204 --> 00:04:33,473
All these people play a key role.

74
00:04:33,473 --> 00:04:37,977
However, there are other people
that can play a role in it, as well,

75
00:04:37,977 --> 00:04:42,048
depending on the type of study
that you're trying to conduct.

76
00:04:42,048 --> 00:04:43,182
If it's --

77
00:04:43,182 --> 00:04:47,287
you also may need to include radiation
oncologist, nutritionist, case workers.

78
00:04:47,287 --> 00:04:51,758
Again, it depends on your study
that you're trying to develop.

79
00:04:51,758 --> 00:04:55,094
Part of the main thing to remember here

80
00:04:55,094 --> 00:04:58,831
is the principle investigator really does
play a key role.

81
00:04:58,831 --> 00:05:01,801
He's responsible
-- he or she is responsible

82
00:05:01,801 --> 00:05:05,538
for all aspects of the trial,
including the ethical conduct

83
00:05:05,538 --> 00:05:09,609
of it, of the research, the eligibility,
protocol adherence, adverse event

84
00:05:09,609 --> 00:05:12,979
reporting, and the reporting of timely
and adequate data.

85
00:05:13,546 --> 00:05:16,382
It's a lot to carry, but it's --

86
00:05:16,382 --> 00:05:20,453
and PI is principle investigators
need be aware of these roles.

87
00:05:20,453 --> 00:05:24,524
They also determine the response
to therapy and intervention and determines

88
00:05:24,524 --> 00:05:25,925
whether the adverse event

89
00:05:25,925 --> 00:05:30,530
may be attributed the actual agent or drug
that's used in a trial.

90
00:05:30,530 --> 00:05:35,835
Each member of the team, however, does
play a key -- a role in conducting

91
00:05:35,835 --> 00:05:37,236
research ethically,
maintaining confidentiality,

92
00:05:37,236 --> 00:05:40,773
and following the code of federal
regulations, and complying HIPPA.

93
00:05:42,075 --> 00:05:42,809
So in

94
00:05:42,809 --> 00:05:45,945
order
to do that, the training is essential.

95
00:05:45,945 --> 00:05:48,581
The protocols often considered a roadmap.

96
00:05:48,581 --> 00:05:52,585
It's a document that includes information
and hopefully it's written clearly

97
00:05:52,585 --> 00:05:56,189
so everyone understands that,
but it provides info regarding why

98
00:05:56,189 --> 00:06:00,159
this study's been done,
the hypothesis behind it, the aims in

99
00:06:00,159 --> 00:06:04,530
how the study is going to be conducted
to meet those aims.

100
00:06:05,331 --> 00:06:09,168
The protocol should contain sufficient
detail so that there is uniformity

101
00:06:09,168 --> 00:06:12,338
in the selection
and treatment of patients, and uniform

102
00:06:12,338 --> 00:06:13,906
data collection, and submission.

103
00:06:13,906 --> 00:06:17,210
So things to consider
whey you're developing the protocol.

104
00:06:17,210 --> 00:06:20,380
What data elements
are you going to be collecting?

105
00:06:20,380 --> 00:06:24,584
What is the design and the content
of the data collection instruments

106
00:06:24,584 --> 00:06:28,054
then that need to be considered
to capture that data?

107
00:06:28,621 --> 00:06:34,193
And then what type of computer database
are you going to be utilizing?

108
00:06:34,193 --> 00:06:38,164
Common data elements is a term
that's often used.

109
00:06:38,164 --> 00:06:42,535
And it's actually quite important
in developing your case report forms.

110
00:06:42,535 --> 00:06:46,939
They're elements that have been determined
to be identical between projects

111
00:06:46,939 --> 00:06:50,910
and context, and the examples
might be name, age, gender.

112
00:06:50,910 --> 00:06:54,113
There's not too many ways to define those.

113
00:06:54,113 --> 00:06:57,283
They tend to be
fairly common across studies.

114
00:06:58,251 --> 00:07:00,286
So they're
considered common data elements.

115
00:07:00,286 --> 00:07:05,691
And I'll go into a little bit more detail
as we go further into this discussion,

116
00:07:05,691 --> 00:07:08,728
but they really do help
to facilitate the understanding

117
00:07:08,728 --> 00:07:11,097
and sharing of that information
across studies.

118
00:07:11,097 --> 00:07:15,501
So if someone else wants replicate,
or do a similar type of study,

119
00:07:15,501 --> 00:07:19,906
would want to use the same data elements,
and that would then inform

120
00:07:19,906 --> 00:07:23,309
maybe a future meta-analysis
that someone would like to do.

121
00:07:23,643 --> 00:07:27,547
Meta-analysis are much easier to do
when there are common

122
00:07:27,547 --> 00:07:30,049
data elements being used between trials.

123
00:07:30,049 --> 00:07:34,520
So at study entry,
some data elements that could be captured,

124
00:07:34,520 --> 00:07:38,391
or the demographic data, eligibility
criteria, family history, patient history.

125
00:07:38,391 --> 00:07:44,230
All of these are pieces of information
that are collected at the time of study

126
00:07:44,230 --> 00:07:48,901
entry and each of them may have multiple
data elements within them.

127
00:07:49,669 --> 00:07:53,673
So on study, each category,
again, can contain multiple data elements

128
00:07:53,673 --> 00:07:57,677
and they may consider capturing
the treatment, the assessments of labs,

129
00:07:57,677 --> 00:08:01,681
radiology, things that have been done
in the interim, during treatment.

130
00:08:01,681 --> 00:08:04,951
What kind of concomitant meds were taken,
adverse events,

131
00:08:04,951 --> 00:08:06,018
hospitalizations, treatment response?

132
00:08:06,018 --> 00:08:10,389
And then there are patient reported
outcomes, forms that may be completed,

133
00:08:10,389 --> 00:08:14,026
as well, such as diaries
and quality of life questionnaires.

134
00:08:14,360 --> 00:08:17,363
These are things you want to consider

135
00:08:17,363 --> 00:08:20,600
as you're developing your case report
forms.

136
00:08:20,600 --> 00:08:25,972
Source documents are basically any
document where the data is first recorded.

137
00:08:25,972 --> 00:08:28,975
It's where the information for the forms

138
00:08:28,975 --> 00:08:31,978
-- case report forms is captured from.

139
00:08:31,978 --> 00:08:35,848
They confirm the protocol adherence,
they confirm and validate

140
00:08:35,848 --> 00:08:39,719
the data being submitted
is actually there and real.

141
00:08:39,719 --> 00:08:44,457
It also serves audit trail,
allowing the recreation of the study.

142
00:08:44,457 --> 00:08:49,195
And it also confirms the actual existence
of the study participants

143
00:08:50,129 --> 00:08:55,668
I think you have or will be having
your lecture on scientific conduct

144
00:08:55,668 --> 00:09:00,339
and hopefully within that there was
some examples shared regarding fraud.

145
00:09:00,339 --> 00:09:02,475
And it actually does happen.

146
00:09:02,475 --> 00:09:04,610
Some investigators make up patients.

147
00:09:04,610 --> 00:09:08,014
And doing an audit
or actually having source

148
00:09:08,014 --> 00:09:11,417
documents
helps to track and find those issues.

149
00:09:11,417 --> 00:09:15,688
So source document examples might be
simply is the hospital

150
00:09:15,688 --> 00:09:20,393
records, the clinic and office charts,
and may include various reports.

151
00:09:20,393 --> 00:09:25,064
The actual progress notes, including
physicians as well as nurse notes.

152
00:09:26,899 --> 00:09:27,333
Other

153
00:09:27,333 --> 00:09:31,671
examples might be letters from referring
physicians, actual radiological films,

154
00:09:31,671 --> 00:09:32,738
tumor measurements,

155
00:09:32,738 --> 00:09:33,873
participant diaries, participant

156
00:09:33,873 --> 00:09:37,076
interviews, and pharmacy
dispensing records, and photographs.

157
00:09:37,076 --> 00:09:42,481
So actual data abstraction then is the
anything that basically is considered

158
00:09:42,481 --> 00:09:48,521
in this, is anything recorded on a CRF
should be in the source document.

159
00:09:48,521 --> 00:09:51,557
If there's something recorded on the CRF,

160
00:09:51,557 --> 00:09:55,861
it has to be tracked
back to a source document.

161
00:09:55,861 --> 00:09:58,898
And a common statement or common
post-it-note,

162
00:09:58,898 --> 00:10:02,802
IR post,
or I would put around oftentimes for

163
00:10:02,802 --> 00:10:06,272
my staff was basically to keep in mind

164
00:10:06,272 --> 00:10:09,275
if it's not written, it didn't happen.

165
00:10:09,275 --> 00:10:12,311
That's how the auditors will see that.

166
00:10:12,311 --> 00:10:16,849
It may have actually happened,
you wrote it on a CRF,

167
00:10:16,849 --> 00:10:21,554
but if it's not written in the source,
it's actually not replicated

168
00:10:21,554 --> 00:10:25,491
or found in the source document,
then they're basically going

169
00:10:25,491 --> 00:10:28,060
to consider it
falsified information, possibly.

170
00:10:28,060 --> 00:10:32,164
Any correction or change should be dated,
initialed, and explained,

171
00:10:32,164 --> 00:10:36,502
and should not -- the original information
should not be obscured.

172
00:10:36,502 --> 00:10:41,207
And this applies to the CRF
as well as the source document.

173
00:10:41,207 --> 00:10:43,175
Only provide the requested data.

174
00:10:43,175 --> 00:10:47,013
Some research staff
I've had like to go on and write

175
00:10:47,013 --> 00:10:50,816
a book sometimes, or a lot of information
that's not necessary.

176
00:10:50,816 --> 00:10:53,953
Just -- the CRF should be designed
well enough

177
00:10:53,953 --> 00:10:57,790
to capture the information if they want
and what is needed.

178
00:10:57,790 --> 00:11:00,226
And use standard
medical terminology, as well.

179
00:11:00,226 --> 00:11:04,030
And as far as the corrections
and the data abstraction, too,

180
00:11:04,030 --> 00:11:07,867
the thing to keep in mind is
I've had occasionally staff

181
00:11:07,867 --> 00:11:10,970
that would use post-it-notes
kind of as a reminder

182
00:11:10,970 --> 00:11:14,440
for themselves to go back in
and make that note.

183
00:11:14,440 --> 00:11:15,841
Then they forget or

184
00:11:15,841 --> 00:11:19,679
they forget to take the post-it-note out
when the auditors appear.

185
00:11:19,679 --> 00:11:22,214
And so, it doesn't look very good.

186
00:11:22,214 --> 00:11:25,451
So post-it-notes
should be not used in these cases.

187
00:11:25,451 --> 00:11:26,919
So again, data abstraction.

188
00:11:26,919 --> 00:11:30,890
The auditors when they come in
should be able to reconstruct

189
00:11:30,890 --> 00:11:34,493
that patient's study course
by piecing together all the data

190
00:11:34,493 --> 00:11:37,029
that's obtained
from those source documents.

191
00:11:37,029 --> 00:11:39,565
So the methods of data collection.

192
00:11:39,565 --> 00:11:43,569
Just to give you kind of a pictorial view,
very simple,

193
00:11:43,569 --> 00:11:47,173
of a possible electronic captured form,
as well as paper.

194
00:11:47,473 --> 00:11:52,278
Those are the two main types of forms
that are currently in use.

195
00:11:52,278 --> 00:11:55,214
A case report
form is basically a preprinted,

196
00:11:55,214 --> 00:11:59,251
if it's printed, form,
developed by the sponsor or the PI

197
00:11:59,251 --> 00:12:04,423
to determine the data elements that are
going to be outlined in the protocol.

198
00:12:04,423 --> 00:12:07,727
It basically translates the protocol
specific activities into data.

199
00:12:07,727 --> 00:12:09,562
They ensure there's actual standardization

200
00:12:09,562 --> 00:12:12,898
and consistency of the data
captured across multiple sites,

201
00:12:13,799 --> 00:12:16,302
so the data that's collected in

202
00:12:16,302 --> 00:12:20,039
Kansas is the same data
that's collected in Colorado.

203
00:12:20,039 --> 00:12:20,473
Design.

204
00:12:20,473 --> 00:12:24,210
The protocol report case
report forms are hopefully designed

205
00:12:24,210 --> 00:12:27,580
to impact the quality of the data
collected.

206
00:12:27,580 --> 00:12:30,916
And basically, proper designed CRF
actually streamline audits

207
00:12:30,916 --> 00:12:36,122
make them easier and also help
with the data analysis and reporting.

208
00:12:36,122 --> 00:12:42,194
So I put together a bit of a pictorial
based on a previous presentation

209
00:12:42,194 --> 00:12:45,931
of the actual protocol,
which describes the study conduct.

210
00:12:45,931 --> 00:12:49,702
It goes through the purpose
objectives and operational aspects.

211
00:12:49,702 --> 00:12:54,173
And then there's the analysis plan, which
is part of the protocol document, also.

212
00:12:54,173 --> 00:12:58,310
But the data analysis then describes
the data that's going to be collected,

213
00:12:58,310 --> 00:13:03,082
key endpoints that are going to be looked
for, as well as the statistical methods

214
00:13:03,082 --> 00:13:06,285
that are going to be used to analysis
the data.

215
00:13:06,285 --> 00:13:08,821
The case report form supports
those two things.

216
00:13:08,821 --> 00:13:11,390
That's how vital a case report form is.

217
00:13:11,690 --> 00:13:15,795
So, I think this view of it says a lot.