Retrieving Medidata Rave Study Data

Overview: Retrieving Medidata Rave Study Data
Notification of Potential Changes
Set Up the Job to Retrieve Medidata Rave Data
Run the Job, and Then Check the Results
Ensuring No Raw Data Is Lost during Transformation

Overview: Retrieving Medidata Rave Study Data

After you have created the SAS Clinical Data Integration data table definitions from Medidata Rave metadata, you can populate the SAS Clinical Data Integration study with data extracted from a Medidata Rave study. There are two transformations to retrieve study data from Medidata Rave.
Note: All study data retrieved from Medidata Rave is Unicode.
The Extract Medidata Rave Data transformation deletes all Medidata Rave data from the specified data tables in the associated SAS Clinical Data Integration study. Then, the transformation inserts the Medidata Rave study data into the specified data tables. You specify the data tables when you set up the job.
The Update from Medidata Data Rave transformation updates the Medidata Rave study data in the specified data tables with data that has changed since the study data was last retrieved. You specify the data tables when you set up the job.
If you run the Update from Medidata Data Rave transformation on a SAS Clinical Data Integration data table that has not been populated using either of these two transformations, the Update from Medidata Data Rave transformation populates the study just as the Extract Medidata Rave Data transformation does.

See Also

Create a SAS Clinical Data Integration Data Table Definition from Medidata Rave Metadata

Notification of Potential Changes

When you retrieve study data from Medidata Rave, it is possible that the Medidata Rave table metadata has changed since the data table definition was initially imported. You can choose how to handle a potential change. You can attempt to retrieve the changed data, or you can skip the retrieval of study data for any data table with Medidata Rave table metadata that might have changed. And, you can choose to notify one or more users via e-mail if the transformation detects potential table metadata changes.
The value that you specify for the e-mail address can be a single address, a mailing list, or any other value that is valid for your default e-mail client. The transformation passes the value to the e-mail client without any validation.
Note: Notification does not apply to the Medidata Rave Comments table. The structure of the Comments table cannot change across study versions. Therefore, omitting notification prevents unnecessary notifications.

Set Up the Job to Retrieve Medidata Rave Data

To set up the job to retrieve Medidata Rave data, perform the following steps:
  1. In the Folders tree, navigate to a SAS Clinical Data Integration study that is mapped to Medidata Rave, and create a job.
  2. In the Transformations tree, expand Clinical.
  3. Drag the Extract Medidata Rave Data transformation or the Update from Medidata Rave Data transformation onto the job, and then double-click the transformation.
    The Properties dialog box appears.
  4. To select a Medidata Rave data table definition, perform the following steps:
    1. Click the Tables tab, select a clinical study, and then select one or more Medidata Rave data table definitions.
      Extract Medidata Rave Data Properties dialog box — Tables tab
      Note the selected data table definitions.
      Note: You can select a Medidata Rave data table definition that contains no data.
    2. Click OK.
  5. To notify users of potential changes to the Medidata Rave table metadata, perform the following steps:
    1. Click the Notifications tab.
      Extract Medidata Rave Data Properties dialog box — Notifications tab
    2. Select the Send Email on Potential Table Metadata Mismatch check box.
    3. In the Address field, enter a valid e-mail address or e-mail group.
    4. Specify what action to take when a potential mismatch is detected, and then click OK.
  6. Confirm that there is a green check mark in the lower right corner of the transformation.
    If the green check mark is not there, click on the transformation to display a message that indicates any problems.

Run the Job, and Then Check the Results

To update data table data from Medidata Rave, perform the following steps:
  1. Run the job
    After the job runs, there should be no errors or warnings in the log. If there are, debug them, and run the job again.
  2. Save and close the job.
  3. Select the transformation log table, and open it.
    There is a record per selected table.
  4. Close the transformation log table.
  5. For each data table that was selected for processing, perform the following steps:
    1. In the Folders tree, right-click a table, and then select Open.
      The View Data page appears.
    2. Confirm that the data is correct.
    3. Compare the values in the regular data columns to the _RAW data columns to ensure that no raw data was lost.
    If the Medidata Rave data tables contain data, the tables selected for processing contain data. If the Medidata Rave data tables do not contain data, the tables selected for processing do not contain data.
    Note: It is possible that a table might not have data in it if there are no records in the data entry system.

Ensuring No Raw Data Is Lost during Transformation

When the Medidata Rave Web Services transforms raw data into regular data, some data might fail validation or transformation. Failure during validation or transformation is the result of flawed raw data in Medidata Rave. To ensure that this flawed raw data is not lost during transformation, you can examine in SAS Clinical Data Integration the raw data that was used during the transformation. The ability to examine the raw data helps you prevent submitting null data to a regulatory body when flawed data was entered in Medidata Rave.
The raw data is included in the SAS Clinical Data Integration data set so that you can examine the data and correct it.
For every _RAW data column definition in a Medidata Rave data table definition, there is a corresponding regular data column with the same name, minus the _RAW designation. For example, the IT_SEVERE_RAW data column has a corresponding regular data column named IT_SEVERE.
If the raw data was validated or transformed without error, the values in both the regular data column and the _RAW data column are identical.
If the raw data was not validated or transformed without error, the value in the regular data column is empty, and the _RAW data column contains the raw, invalid value that was in Medidata Rave.
Copyright © SAS Institute Inc. All rights reserved.