You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 11 Next »

The following steps are recommended to support development of CRFs using standards in this guide:


OrderStepImplementation
1

Identify data collection requirements.

Determine requirements for data collection from the protocol and by reviewing internally maintained libraries of standardized CRFs when applicable.

2

Identify collection domains and fields.

Review collection standards in this guide to identify data collection domains and fields which align with collection requirements. As much as possible, domains in this guide will be used to collect data in a manner that will be effective for data collection. Develop the data collection tools using domains in this guide first. Create additional fields in alignment with guidance in Section x.x,  


3


During the development of CDASH-conformant collection instruments (e.g., CRFs, eCOA screens), the SDTMIG domain to which the collected data is to be mapped must be determined. The choice of the SDTMIG domain to use does not depend upon the mode of transmission, the methodology used to generate the data, the medium used to store the data, the person who recorded the data, or the subject described by the data. The SDTMIG domain to be used affects what CDASH variable names, question texts, prompts, controlled terminology, and so on, to use. CDASH suggests a format to be presented to those entering the data, but it does not dictate any data structure in which to store the collected data (often referred to as a data management operational database).


The key steps to developing CRFs using CDASH are:


  1. Example 1: A study has meal-consumption diary data captured via a subject-completed PRO. Another study also captures meal-consumption data, but the subject takes a photo of the food prior to and after the meal, and sends the photos to a third party, which determines food consumption. Even though captured in different ways, the data from both studies will map into the SDTMIG Meal Data (ML) domain.

    Example 2: A study has subjects' blood samples sent to a central lab, which analyzes the samples and sends results to the sponsor via electronic data transfer. In a second study, the samples are analyzed locally and results are captured on a CRF. The laboratory results from both studies are stored in the SDTMIG Laboratory Test Results (LB) domain.

    CDASH recommends that dates be collected in an unambiguous format and suggests using the DD-MON-YYYY format. This defines the format to be presented to those entering the data, but does not define the electronic format in which to store the data. One system may store each date as a character field; another may store them as numeric values (e.g., an SAS date); and yet another as 3 separate fields formatted as day, month, and year. Each of these is a legitimate way to store the data collected.


  2. Using the root variables and other CDASH metadata in the CDASH Model, add any additional variables that are needed to meet the requirements of data collection. Follow CDISC Variable Naming Fragment (see Appendix B, Glossary and Abbreviations) conventions, and CDASH root variable-naming conventions where they exist (e.g., --DAT for dates, --TIM for times, --YN for prompts, as described in the CDASH Model). 
    Example: Replace "--" with the 2-character domain code that matches the other variables in the same domain. For example, to add the --LOC variable to a Medical History CRF, the domain code is MH, so the variable would become MHLOC in that domain.


  3. The Question Text and Prompt columns in the CDASH Model metadata provide different variations in the recommended text for asking the question on a CRF. For each question, the sponsor may elect to either use the Question Text or the Prompt on the CRF. Some text is presented using brackets [ ], parentheses ( ), and/or incorporating forward slashes. These different formats are used to indicate how the Question Text or Prompt may be modified by the sponsor.


    1. Example: The CDASH variable --PERF, from the CDASH Model, has the following Question Text and Prompt.

      Question Text:  [Were any/Was the] [--TEST/ topic] [measurement(s)/test(s)/examination(s)/specimen(s)/sample(s)] [performed/collected]?

       Prompt:  [--TEST/Topic] [Measurement(s)/Test(s)/Examination(s)/Specimen(s)/Sample(s)] [Performed/Collected]?

      The sponsor wants to add a question to a CRF that asks whether a lab specimen was collected, using a Yes/No response.

      The sponsor selects the CDASH variable --PERF and adds the appropriate domain code. LBPERF  

      Use either the Prompt or the Question Text on the CRF.

      Question Text: Was the laboratory specimen collected?

      • In the first set of brackets, the text option "Was the" is selected, as the study required only 1 lab test to be performed. [Were any/Was the
      • In the second set of brackets, the text used is "laboratory," which is the topic of interest. [--TEST/Topic (laboratory)]
      • In the third set of brackets, the text option "specimen," without the optional "s," is selected. [measurement(s)/test(s)/examination(s)/specimen(s)/sample(s)]
      • In the fourth set of brackets, the text option "collected" is selected. [performed/collected]

      Prompt: Laboratory Specimen Collected

      •  In the first set of brackets, the text used is the topic of interest (i.e., laboratory). [--TEST/Topic (Laboratory)
      • In the second set of brackets, the text option "specimen," without the optional "s," is selected. [Measurement(s)/Test(s)/Examination(s)/Specimen(s)/Sample(s)]
      • In the third set of brackets, the text option "collected" is selected. [Performed/Collected]


  4. Create custom domains based on 1 of the General Observation Classes in the CDASH Model. See Section 3.4, How to Create New Data Collection Fields When No CDASHIG Field Has Been Defined, for more information.

CDASHIG metadata table attributes provide building blocks for the development of a CRF and the underlying database or other data-collection structure.

Additional information on developing CRFs can be found in Section 2.3.1, Overview of Example CRFs.



  • No labels