term
stringlengths 2
142
⌀ | definition
stringlengths 9
116k
⌀ | sources
listlengths 1
1
⌀ | source
stringclasses 48
values | type
stringclasses 5
values | file
stringclasses 26
values | sheet
stringclasses 3
values | text
stringclasses 212
values |
|---|---|---|---|---|---|---|---|
Page 174
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PC
|
Identifier Two-character abbreviation for the domain most relevant to the observation.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 176
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 178
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
All Rows
|
PCTPTREF is a text value of the description of a “zero” time. It should be meaningful. If there are multiple PK profiles being generated, the zero time for each will be different (e.g., a different dose such as "first dose", "second dose"). In this example it is required to make values of PCTPTNUM and PCTPT unique (See Section 4.1.4.10). Rows 1-4, 12: Sponsor has standardized values of “<0.01” to “0” for this study. Rows 5, 6, 19, 20, 25, 26, 29, and 30: Specimen properties (volume and pH) are submitted as separate rows. Rows 3-6, 17-20, 23-30: The elapsed times for urine samples are based upon the elapsed time (from the reference time point, PCTPTREF) for the end of the specimen collection period. Elapsed time values that are the same for urine and plasma samples have been assigned the same value for PCTPT. For the urine samples, the value in PCEVLINT describes the planned evaluation (or collection) interval relative to the time point. The actual evaluation interval can be determined by determining PCENDTC - PCDTC. Row STUDYID DOMAIN USUBJID PCSEQ PCGRPID PCREFID PCTESTCD
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PCCAT
|
PCSPEC PCORRES PCORRESU PCSTRESCPCSTRESNPCSTRESU 1
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
ANALYTE PLASMA
|
) CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
URINE
|
2.4 ng/mL 2.4 2.4 ng/mL 29
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DAY 11
|
11 2001-02-11T14:03 2001-02-11T20:10 11 12H 3 Day 11 Dose 2001-02-11T08:00
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PREDOSE
|
0 Day 11 Dose 2001-02-11T08:00 -PT15M 15 (cont) 0.10 3
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PT6H
|
-PT6H 21 (cont) 0.10 3
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DAY 2
|
2 2001-02-02T08:00 2 24H 3 Day 1 Dose 2001-02-01T08:00
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PT12H
|
-PT6H 27 (cont) 2.00 4
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DAY 12
|
12 2001-02-12T08:00 12 24H 4 Day 11 Dose 2001-02-11T08:00
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PT24H
|
-P12H 31 (cont) 0.10 4
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 180
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DRUG A PARENT
|
Half-life of 2nd exp phase 4.50
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DRUG A METABOLITE
|
THALF_2 Half-life of 2nd exp phase 2.25
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Clearance
|
0.88 L/h (Additional variables in dataset example on following page- > ) CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DAY 8
|
2 2001-02-08T18:30 8 600 min 12
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 182
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
RELATING DATASETS
|
If all time-point concentrations in PC are used to calculate all parameters for all subjects, then the relationship between the two datasets can documented as shown in the table below. Note that incorporating the name of the analyte and the day of the collection into the value of --GRPID (or some equivalent method for assigning different values of --GRPID for all the combinations of analytes and reference time points) is necessary when there is more than one reference time point (PCRFTDTC and PPRFTDTC, which are the same for related records), and more than one analyte (PCTESTCD, copied into PPCAT to indicate the analyte with which the parameters are associated), since these variables are part of the natural key (see Section 3.2.1.1) for both datasets. In this case, --GRPID is a surrogate key (Section 3.2.1.1) used for the relationship.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
RELATING RECORDS
|
Four possible examples of different types of relationships between PC and PP records for one drug (DRUG X in this case) are described. For all of these, the actual PC and PP data are the same. The only variables whose values change across the examples are the sponsor-defined PCGRPID and PPGRPID. As in the case for relating datasets above (Section 6.3.10.5.1), --GRPID values must take into account all the combinations of analytes and reference time points, since both are part of the natural key (see Section 3.2.1.1) for both datasets. To conserve space, the PC and PP domains appear only once, but with four --GRPID columns, one for each of the examples. Note that a submission dataset would contain only one --GRPID column with a set of values such as those shown in one of the four columns in the PC and PP datasets, or values defined by the sponsor. The example specifics are as follows: Example 1: All PK time-point-concentration values in the PC dataset are used to calculate all the PK parameters in the PP dataset for both Days 1 and 8 for one subject. Pharmacokinetic Concentrations (PC) Dataset For All Examples Pharmacokinetic Parameters (PP) Dataset For All Examples RELREC Example 1. All PC records used to calculate all PK parameters • Method A (Many to many, using PCGRPID and PPGRPID) • Method B (One to many, using PCSEQ and PPGRPID) • Method C (Many to one, using PCGRPID and PPSEQ) • Method D (One to one, using PCSEQ and PPSEQ) Example 2: Two PC values were excluded from the calculation of all PK parameters for the Day 1 data. Day 8 values are related as per Example 1. RELREC Example 2: Only some records in PC used to calculate all PK parameters • Method A (Many to many, using PCGRPID and PPGRPID) • Method B (One to many, using PCSEQ and PPGRPID) • Method C (Many to one, using PCGRPID and PPSEQ) • Method D (One to one, using PCSEQ and PPSEQ) CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 184
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
ABC-123-0001
|
12 DY1_DRGX DY1_DRGX DY1_DRGX_A DY1DRGX_D 123-0001-12
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DRUG X
|
First dose of study drug, where drug is DRUG X 2 weeks after start of Element
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PCELTM
|
1 (cont) 9 9 ug/mL 1.00 1
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Day 1 Dose
|
2001-02-01T08:30 PT10H 13 (cont) 10.0 10.0 ug/mL 1.00 2
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Day 8 Dose
|
2001-02-08T08:30 PT8H20M 24 (cont) 11.0 11.0 ug/mL 1.00 2
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 186
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
5 (cont)
|
T1/2, 2 Half-life of 2nd exp phase
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
12 (cont)
|
T1/2, 2 Half-life of 2nd exp phase
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DY8DRGX
|
2 * RELID 1 indicates records with PCSEQ values of 1-12 are related to records with PPGRPID = DY1DRGX. * RELID 2 indicates records with PCSEQ values of 13-24 are related to records with PPGRPID = DY8DRGX. Method C (Many to one, using PCGRPID and PPSEQ)
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PPSEQ
|
7 4 The Day 8 relationships are the same as those shown in Example 1. * RELID 1 indicates records with PCSEQ values of 1-4 and 6-12 are related to the record with a PPSEQ value of 1. * RELID 2 indicates records with PCSEQ values of 1-5 and 7-12 are related to the record with a PPSEQ value of 2. * RELID 3 indicates records with PCSEQ values of 1-10 are related to the record with a PPSEQ value of 3. * RELID 4 indicates records with PCSEQ values of 1-12 are related to the records with PPSEQ values of 4-7.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 188
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
DY1DRGX
|
1 The Day 8 relationships are the same as those shown in Example 1. * RELID 1 indicates records with PCSEQ values of 1-7 and 10-12 are related to records with PPGRPID = DY1DRGX. * RELID 2 indicates records with PCSEQ values of 13-24 are related to records with PPGRPID = DY8DRGX. Method C (Many to one, using PCGRPID and PPSEQ)
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 190
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PPGRPID
|
DY1DRGX_HALF 2 The Day 8 relationships are the same as those shown in Example 1. * RELID 1 indicates records with PCSEQ values of 1-12 are related to records with PPGRPID = DY1DRGX_A * RELID 2 indicates records with PCSEQ values of 1-7 and 10-12 are related to records with PPGRPID = DY1DRGX_HALF. Method C (Many to one, using PCGRPID and PPSEQ)
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 192
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
PCGRPID
|
DY1DRGX_D 4 The Day 8 relationships are the same as those shown in Example 1. * Same RELID of "1" Indicates that Tmax used records with PCGRPID values DY1DRGX_A, DY1DRGX_C, and DY1DRGX_D. * Same RELID of "2" Indicates that Cmax used records with PCGRPID values DY1DRGX_A, DY1DRGX_B, and DY1DRGX_D. * Same RELID of "3" Indicates that AUC used PCGRPID values DY1DRGX_A, DY1DRGX_B, and DY1DRGX_C. * Same RELID of "4" Indicates that all other parameters (PPGRPID = OTHER) used all PC time points: PCGRPID values DY1DRGX_A, DY1DRGX_B, DY1DRGX_C, and DY1DRGX_D. Note that in the above RELREC table, the single records in rows 1, 3, 5, 7, and 9, represented by their --GRPIDs (TMAX, DY1DRGX_C, CMAX, DY1DRGX_B, AUC) could have been referenced by their SEQ values, since both identify the records sufficiently. At least two other hybrid approaches would have been acceptable as well: using PPSEQ and PCGRPIDs whenever possible, and using PPGRPID and PCSEQ values whenever possible. Method D below uses only SEQ values. CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 194
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 196
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
CFOBJ
|
CFORRES CFORRESU CFSTRESC CFSTRESCU VISIT
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
LEFT ARM
|
Perm SDTMIG 2.4.3, SDTMIG 4.1.1.9
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 198
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
VESICLES
|
. 5. Records in CF may or may not relate to records in an Events domain, such as AE or CE. If there are related event records, then RELREC may be used to document the relationships. a. If the clinical event or condition being evaluated does not represent a reportable adverse event, and if no data about the event as a whole (e.g., start date, end date) is collected, then there is no requirement that there be an event record for the clinical event or condition. b. If the clinical event or condition being assessed represents a reportable adverse event, as defined/agreed in the particular study, then it must be recorded in the AE domain, since all adverse events must be recorded in the AE domain. 6. Although no examples for clinical findings related to interventions have been provided, data on findings about interventions are expected to arise and may be stored in CF. 7. The following qualifiers should generally not be used in CF: --BODSYS, --MODIFY, --DTHREL, --SEV. CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Date of Collection
|
Did you have the following?
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 200
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
SYMPTOMS
|
INVESTIGATOR GERD SYMPTOM MEASUREMENT (IF SYMPTOM OCCURRED) OCCURRED? Yes/No VOLUME (mL) Total for all episodes
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
NAUSEA
|
1 1 1 2006-02-02 10
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
XYZ-701-002
|
13 NUMEPISD Number of episodes DIARRHEA 1 1 2 2006-02-03 14
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
NOT DONE 2006-02-03
|
CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
VOMIT
|
>10 >10 1 2006-02-02 4
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 202
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Visit
|
1 2 3 4 5 6
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
AE Domain (For clarity, only selected variables are shown.)
|
Row 1 shows the record for a verbatim term of "Morning queasiness", for which the maximum severity over the course of the event was "Moderate." Row 2 shows ????
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
CF domain
|
Rows 1-4 show severity data collected at the four visits that occurred between the start and end of the AE 'Morning queasiness'. CFOBJ = NAUSEA, which is the value of AEDECOD in the associated AE record. Rows 5-6 show severity data collected at the two visits that occurred between the start and end of the AE 'Watery stools'. CFOBJ = DIARRHOEA, which is the value of AEDECOD in the associated AE record.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 204
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 206
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Domain
|
A collection of observations with a topic-specific commonality about a subject
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Rule
|
Rule that expresses the criterion in computer-executable form.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 208
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Randomization
|
Evaluation for disease progression Evaluation/consent for surgery
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
In the next diagram, the Epochs of the trial have been identified. The Epochs are represented by the rectangles
|
"behind" the blocks representing the Elements. In this simple trial, which has three Elements in each Arm, there are three Epochs. There is not always a one-to-one relationship between the Elements in the Arms of the trial and the Epochs of the trial, but such a relationship is common, particularly for blinded trials.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 210
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Blind
|
CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Drug B
|
First dose of study drug, where drug is Drug B 2 weeks after start of Element
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 212
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
P-5-10
|
CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 214
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Placebo-5mg-10mg
|
Second Treatment Epoch 4 5 5 mg 5
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Third Treatment Epoch
|
6 10 10 mg 14
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
First Treatment Epoch
|
2 5 5 mg 17
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
TA
|
CRS/No P/S CRS/No P/S Contin/
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 216
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Rescue
|
See Section 7.2.3.1 for additional discussion of when a decision point in a trial design should be considered to give rise to a new Arm. CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Screen Study Drug
|
Blind/Rescue Blind/Open A Study Drug Open Drug A
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 218
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Treatment A
|
5 days after start of
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Treatment B
|
Assigned to Rescue on basis of
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 220
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
In the diagram below, the trial is considered to have one Epoch for each cycle, with each Epoch consisting of two
|
Elements, a treatment Element and a rest Element. The choice between these alternatives is a somewhat subjective one. In oncology trials, there is a strong tradition of discussing "cycles" and the second diagram is more consistent with this tradition. However, if there were important analyses which distinguished between assessments or events in the Treatment and Rest Elements, then the first assignment of Epochs might better support those analyses.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 222
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
The logistics of dosing mean that few oncology trials are blinded, but the diagrams we have drawn might represent a
|
blinded trial. If this trial is blinded, then the diagram below, based on the one-Epoch-per-cycle alternative, shows the trial from the viewpoint of blinded participants.
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 224
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Rest
|
Last dose of previous treatment cycle + 24 hrs At least 16 days after start of Element and WBC recovered 5
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 226
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
R A Follow
|
Trt A = Treatment A R A = Rest A B = Treatment B
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 228
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
The Trial Design Matrix below corresponds to the last diagram for this trial, with one Epoch per cycle.
|
Trial Design Matrix for Example Trial 5, with one Epoch per Cycle
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Rest A
|
If disease progression, go to Follow-up Epoch 8
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Rest B
|
If disease progression, go to Follow-up Epoch 18
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 230
|
CDISC. © 2007. All rights reserved July 1025, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
The Trial Design Matrix for this trial assumes that Epochs are assigned as in the diagram above. Note that in this
|
trial, there are eight Elements in the treatment Epoch for Arm A, and six Elements in the treatment Epoch for Arm B. Trial Design Matrix for Example Trial 6
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 232
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 234
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
No surgery
|
However, it is also possible to see this study as a two-Arm trial, but with major "skip forward" arrows for some subjects, as illustrated in the following diagram. 3-5w
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Arm 2
|
CDISC, © 2007. All rights reserved
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Progression
|
Example Trial 7, RTOG 93-09: Five Arms and Four Epochs
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Page 236
|
CDISC. © 2007. All rights reserved July 25, 2007
|
[
"SDTM.pdf"
] |
SDTM.pdf
|
pdf
|
glossary.json
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.