All content on this Wiki is non-binding and any individual opinions expressed should not be considered indicative of the policies or positions of CDISC or any other organization.
Page tree
Skip to end of metadata
Go to start of metadata

Information for Reviewers

Text in black is part of the template used to create this document and is not under review. Please review the blue text; this text represents the changes made to the template that are specific to this QRS instrument. Also please review the example, which we are unable to put in blue text at this time.

Title

Patient Global Impressions (PGI) Questionnaires Supplement to the Study Data Tabulation Model Implementation Guide for Human Clinical Trials

QRS Long Name

Patient Global Impressions

QRS Short Name

PGI

TeamMDD Standards Team and CDISC Questionnaires, Ratings, and Scales (QRS) Sub-team
Version1.0
StatusDRAFT
Date2018-08-21
NotesThis supplement is intended to be used with other CDISC User Guides for specific Therapeutic/Disease Areas and follows the CDISC Study Data Tabulation Model Implementation Guide for Human Clinical trials.


Revision History

DateVersion
2018-mm-dd1.0 Final
2018-01-201.0 Draft

© 2018 Clinical Data Interchange Standards Consortium, Inc. All rights reserved. 

1 Introduction

This document describes the CDISC implementation of the PGI instruments, which include the Patient Global Impression of Severity (PGI-S), the Patient Global Impression of Change (PGI-C), and the Patient Global Impression of Improvement (PGI-I). These instruments are commonly administered in clinical trials and clinical practices. However, no single (i.e., standardized) version of the PGI-S, PGI-C, or PGI-I currently exists. This is because, for each of these 3 instruments: (1) the specific disease/condition, signs related to the disease/condition, function, and/or functional impact of interest is typically referenced in the wording of the global item and (2) the response option values on these instruments have not been standardized to date. Indeed, it would be extremely difficult to develop standardized versions of these instruments. Therefore, the implementation described in this document is based on the precedent of (1) sponsors using the PGI-S, PGI-C, and PGI-I as distinct instruments; (2) the specific disease/condition, signs related to the disease/condition, function, and/or functional impact of interest typically being referenced in the wording of the global item; and (3) sponsors determining (a) the exact wording used in the global item and (b) the values of both the text and numeric components of the response options. Instead of creating a separate Questionnaire (QS) Supplement for each of the 3 PGI instruments, the approach taken here was to cover all 3 instruments within 1 QS Supplement entitled “Patient Global Impression.”

The PGI instruments were initially developed based on the original Clinical Global Impression (CGI) instrument described in the ECDEU Assessment Manual for Psychopharmacology. As previously noted, both (1) the exact wording used in the global item and (2) the values of both the text and numeric components of the response options historically have been generated by sponsors for each of the 3 PGI instruments as implemented in clinical trials or clinical practices. Therefore, no single, standard Case Report Form (CRF) was developed by CDISC for any of the 3 PGI instruments.

The representation of data collected using the PGI instruments provided in this document is based on the Study Data Tabulation Model Implementation Guide (SDTMIG) QS domain table, which can be found at the CDISC website at https://www.cdisc.org/standards/foundational/sdtmigThe implementation details for the PGI-S, PGI-C, and PGI-I are meant to be used in conjunction with the SDTMIG but are recorded separately because these instruments may be used in a variety of therapeutic areas. CDISC documentation for all questionnaires, ratings, and scales can be found on the CDISC web site at https://www.cdisc.org/foundational/qrs.

The CDISC Intellectual Property Policy can be found on the CDISC web site at: https://www.cdisc.org/about/bylaws.

Reference: Guy W, ed. ECDEU Assessment Manual for Psychopharmacology. Rockville, MD: U.S. Department of Health, Education, and Welfare;1976.

1.1 Representations and Warranties, Limitations of Liability, and Disclaimers

This document is a supplement to the SDTMIG for Human Clinical Trials and is covered under Appendix F of that document, which describes representations, warranties, limitations of liability, and disclaimers. Please see Appendix F of the SDTMIG for a complete version of this material.

2 Copyright Status

The PGI instruments are all in the public domain.  Indeed, there may be many implementations of each PGI instrument in the public domain.  CDISC has included the PGI instruments of severity, change, and improvement as part of CDISC data standards.  This means that CDISC has developed documentation for the PGI instruments that consists of: (1) standard values (controlled terminology) for QSTESTCD, QSTEST, and QSCAT; (2) a standardized database structure; and (3) some example response option values for each instrument. For the reasons discussed in Section 1, Introduction, CDISC has not developed a standard CRF for any of the 3 PGI instruments to annotate with the CDISC SDTMIG submission values.

Note: CDISC controlled terminology is maintained by National Cancer Institute's Enterprise Vocabulary Services (EVS).  The most recent version should be accessed through the CDISC website: https://www.cdisc.org/standards/semantics/terminology.

Reference: No official reference document.

3 The QS Domain Model

3.1 Assumptions for QS Domain Model

All assumptions and business rules described in the SDTMIG QS domain are applicable to this supplement. Additional assumptions specific to the PGI are listed below.

The PGI is a multiple-choice patient-reported questionnaire that may used to assess an underlying construct, or Concept of Interest (COI; e.g., symptom, function, functional impact, medical condition). It consists of 3 separate items for severity, change and improvement, each rated on a Likert scale based on the user's choice of responses to the items.

  1. There currently exist both patient and clinician versions of the global impression instruments (PGI and CGI, respectively). Patient and clinician assessments of severity, change, and improvement are to be kept separate with respect to how these data are captured in clinical trials and then submitted to the FDA or other regulatory authority for review. A separate supplement exists for the CGI instruments of severity, change, and improvement. 
  2. The PGI and CGI instruments are sometimes referred to by other names:
    1. The PGI instruments are sometimes also referred to as the “Patient Global Assessment” or “PGA” instruments.
    2. The CGI instruments are sometimes also referred to as the “Clinical Global Assessment” or “CGA” instruments.
  3. CDISC standards will not be developed for synonymous instruments (e.g., for the Patient Global Assessment or Clinical Global Assessment instruments) because these instruments do not substantially differ from the analogous PGI and CGI instruments. Individuals wishing to use the PGA instruments, CGA instruments, or some other synonymous instruments are directed to CDISC standards developed for the PGI and CGI instruments.
  4. The 3 PGI instruments (i.e., PGI-S, PGI-C, PGI-I) are mutually distinct measures, even when evaluating the same disease/condition within the same clinical study.
  5. Assessments made based on a single point in time should be done using the PGI-S. Assessments made in comparison to earlier time points should be done using either the PGI-C or PGI-I.
  6. Given (a) the use of the PGI instruments in a wide variety of therapeutic areas and (b) that the wording of the global item is often disease/condition-specific, this supplement aims to define data standards that are perhaps more general or broader than usual and can be used to report data generated by any version of the PGI instruments implemented in clinical studies.  Specifically:
    1. QSSCAT (Subcategory for Question) will be used to capture the underlying construct, or COI (e.g., signs related to a specific disease/condition, function, functional impact, medical condition), being measured.
    2.  Although general conventions exist for both the numeric and text components of the response options of all 3 PGI instruments, the actual values of the response options for each of these instruments may vary and are to be determined by the sponsor. However, regardless of the exact values of the response options:
      1. Response option text is to be stored in QSORRES (Finding in Original Units).
      2. The numeric component of a response option is to be stored in QSSTRESC (Character Result/Finding in Standard Format) and QSSTRESN (Numeric Finding in Standard Units) as character and numeric values, respectively.  Example response option values for QSSCAT, QSORRES, QSSTRESC, and QSSTRESN are provided in Section 4, SDTM Mapping Strategy. Please note that the examples provided in this document are not in any way intended to be prescriptive but rather are provided for illustrative purposes.

    3. Note that an instrument is sometimes referred to as the “Patient Global Impression of Improvement,” or PGI-I, when in fact the instrument is assessing change in either direction and is therefore actually an implementation of the “Patient Global Impression of Change,” or PGI-C. In true implementations of the PGI-I, the response options only allow patients to report either no change or improvement in the disease/condition, functions, functional impacts, or other aspects of the patient’s experience of the disease/condition.
  7. In the final dataset submitted to the FDA or other regulatory authority for review, records for QSCAT, QSTESTCD, QSTEST, QSORRES, QSSTRESC, and QSSTRESN are to be provided for each study subject at each assessment time point for each PGI instrument administered. Records are to be provided for both missing and non-missing data (assessments).
  8. The CDISC variable QSMETHOD will be used to capture scale type (e.g., Likert scale, Verbal Rating Scales). 
    1. Note that instruments using the Visual Analog Scale (VAS) and Numeric Rating Scales (NRS) are not covered in the current PGI supplement.
  9. The time period of evaluation for the PGI is populated in the QSEVLINT field in ISO 8601 format (or QSEVINTX, as appropriate) when the evaluation interval can be precisely described as duration. The evaluation interval for the PGI is determined based on the sponsor's stated evaluation interval.

  10. Terminology
    1. QSCAT, QSTESTCD and QSTEST are approved CDISC controlled terminology.

    2. An example list of response value sets for QSORRES, QSSTRESC, and QSSTRESN fields are provided in Section 4. 

3.2 Example for the PGI QS Domain Model

The PGI example below shows the terminology used to implement the instrument in the QS domain. This example shows the data for one subject collected at visits 1 and 2 for a PGI instrument. The example uses CDISC controlled terminology for QSTESTCD, QSTEST, and QSCAT. All original results (i.e., responses) are represented with preferred terminology in QSORRES. This result is then transformed into the standard numeric score in QSSTRESN and a character representation of the standard numeric score in QSSTRESC.

The table below represents an example implementation of the PGI instrument dataset qs.xpt. It represents a given subject’s responses to the global questions posed in the 3 PGI instruments at 2 visits. QSSCAT is populated with the specific underlying construct (e.g., “BACK PAIN”) being evaluated. The values of QSORRES, QSSTRESC, and QSSTRESN are determined by the sponsor and are therefore represented here with values of “X”. Note that for a given construct (e.g., “BACK PAIN”), it is neither required nor typical for all 3 PGI instruments to be administered within the same clinical study.  All 3 have been included in the example below for illustrative purposes only. QSMETHOD is used to store the type of scale used.

 qs.xpt
Rows 1-2:Represent the severity test at Visits 1 and 2.
Rows 3-4:Represent the change and improvement tests based on comparison to the subject at Visit 1.

qs.xpt

RowSTUDYIDDOMAINUSUBJIDQSSEQQSTESTCDQSTESTQSCATQSSCATQSORRESQSSTRESCQSSTRESNQSLOBXFLQSMETHODVISITNUMQSDTC
1STUDYXQS2324-P00011PGI0101PGI01-SeverityPGIBACK PAINXXXXXYLIKERT SCALE 7-POINT12015-06-15
2STUDYXQS2324-P00012PGI0101PGI01-SeverityPGIBACK PAINXXXXX
LIKERT SCALE 7-POINT22015-06-22
3STUDYXQS2324-P00013PGI0102PGI01-ChangePGIBACK PAINXXXXX
LIKERT SCALE 7-POINT22015-06-22
4STUDYXQS2324-P00014PGI0103PGI01-ImprovementPGIBACK PAINXXXXX
LIKERT SCALE 7-POINT22015-06-22

4 SDTM Mapping Strategy

This section provides some examples of commonly utilized response option values for each of the 3 PGI instruments (i.e., PGI-S, PGI-C, PGI-I). Please note that these examples are not intended to be prescriptive; they are provided to help clarify some of the concepts and mapping strategies covered in Section 3, The QS Domain Model. The text of the response option selected by a given study subject at a given assessment time point is stored in the SDTM variable QSORRES, and the corresponding numeric score for the selected response option is stored as a numeric value in the SDTM variable QSSTRESN and as a character value in the SDTM variable QSSTRESC. The values of QSORRES, QSSTRESN, and QSSTRESC provided in this section are to be considered as examples only; the actual values of these variables depend on the specific instrument(s) implemented within a given clinical study.

PGI QSTESTCD and QSTEST Controlled Terminology


Instrument 

CDISC Controlled Terminology


QSCAT

QSTESTCD

QSTEST

PGI-S

PGI

PGI0101

PGI01-Severity

PGI-C

PGI

PGI0102

PGI01-Change

PGI-I

PGI

PGI0103

PGI01-Improvement

QSMETHOD
VERBAL RATING SCALE 4-POINT  
VERBAL RATING SCALE 7-POINT  
LIKERT SCALE 4-POINT
LIKERT SCALE 7-POINT








QSTESTCD = PGI0101  QSTEST = PGI01-Severity    QSMETHOD = LIKERT SCALE 7-POINT

QSORRES

QSSTRESC

QSSTRESN

Normal

1

1

Borderline

2

2

Mild

3

3

Moderate

4

4

Marked

5

5

Severe

6

6

Extreme

7

7


QSTESTCD = PGI0101  QSTEST = PGI01-Severity     QSMETHOD = LIKERT SCALE 4-POINT

QSORRES

QSSTRESC

QSSTRESN

Normal

1

1

Mild

2

2

Moderate

3

3

Severe

4

4


QSTESTCD = PGI0102  QSTEST = PGI01-Change     QSMETHOD = LIKERT SCALE 7-POINT

QSORRES

QSSTRESC

QSSTRESN

Very much improved

1

1

Much improved

2

2

Minimally improved

3

3

No change

4

4

Minimally worse

5

5

Much worse

6

6

Very much worse

7

7


QSTESTCD = PGI0102  QSTEST = PGI01-Change     QSMETHOD = LIKERT SCALE 7-POINT

QSORRES

QSSTRESC

QSSTRESN

Much worse

-3

-3

Moderately worse

-2

-2

A little worse

-1

-1

No change

0

0

A little better

1

1

Moderately better

2

2

Much better

3

3


QSTESTCD = PGI0103  QSTEST = PGI01-Improvement     QSMETHOD = LIKERT SCALE 7-POINT

QSORRES

QSSTRESC

QSSTRESN

No change                                                             

1

1

Almost the same                                                             

2

2

A little better

3

3

Somewhat better

4

4

Moderately better

5

5

Much better

6

6

A great deal better

7

7

End of Document

  • No labels