Loading...
Please wait, while we are loading the content...
Similar Documents
Best Practices for Working with CDISC Metadata in the SAS ® Data Integration Server
| Content Provider | Semantic Scholar |
|---|---|
| Author | Kilhullen, Michael |
| Copyright Year | 2007 |
| Abstract | Over the past few years, SAS has demonstrated how the SAS Metadata Server can be used to implement and manage CDISC metadata and facilitate a metadata driven approach to standardizing clinical data. In this paper, we examine best practices for using SAS Data Integration Studio to execute and manage key CDISC concepts such as controlled terminology, value level metadata, normalization of data, importing and exporting XML documents, and producing the CRT-DDS. Within the context of these topics, we will also examine considerations for setting up and managing study metadata, writing efficient transformation processes, leveraging metadata to answer key business questions, and effective use of change management. INTRODUCTION This paper will present best practices as a series of topics that contain basic information, considerations, and real world examples from SAS pharmaceutical solution developers and consultants, and through customer feedback. The topics presented are based on a common understanding about how the metadata is organized and used for a typical clinical data integration project. This does not imply that this is how you must implement a metadata server. When considering business objectives and workflow, we sometimes find that some of these recommendations are less helpful. For example, customers who start with data that is already standardized, or customers who only have a handful of SAS programmers, will look at these recommendations from a different perspective. What is more important is that you consider the reasoning behind these recommendations, decide what is best for your business processes, and then standardize the way you intend to use the metadata server. For example, the Clinical Data Integration Service provided by SAS Professional Services and Delivery is built on top of the “clinical metadata framework”. The framework, which implements and leverages the best practices presented in this paper, defines how clinical data and standards are represented in the metadata, how consultants can more easily access the metadata to build customizations, plug-ins and reports, and how metadata as a whole can satisfy common business needs such as change management, archiving, and integration with other solutions such as SAS Drug Development. ORGANIZE YOUR METADATA Metadata describes your physical world in a way that makes sense for how users will interact with it to produce deliverables. It does not require you change how or where your store your physical data and can leverage your existing SAS environments and programs. The metadata server only stores enough information about where everything is, what it looks like, who can access it, etc., to make it easier for you to focus on what you need to do rather there where you need to go to do it. In order to efficiently integrate with business workflow and provide metadata capabilities across your organization, the metadata server maintains one or more metadata repositories. A metadata repository is simply a collection of metadata. All metadata servers contain at least one repositoryFoundation. This repository contains mostly IT level metadata, such as server configurations, user and group information, standard metadata templates, etc. As users begin to interact with the metadata server, they can optionally create other repositories. The two types of repositories supported by the SAS Metadata Server are Custom and Project repositories. Custom repositories leverage all the information defined in Foundation, but allow you to keep your business level metadata physically separate and distinct. From an administrative perspective, it means that you can act on the repository independently to take advantage of Metadata Server capabilities such as replication, promotion, and backups. From a user perspective, it provides more focus. When you work with a custom repository, you only see content, libraries, and folders that are meaningful to that repository. Project repositories are used when working under change management. They are typically assigned to an individual, and stored securely on the metadata server. As a user checks-out objects from a custom repository, the metadata is copied into the project repository. There, metadata can be added, edited, and/or deleted. Changes made by the user are not applied to the custom repository until the contents of the project repository are checked in. Foundation |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://www.lexjansen.com/pharmasug/2007/sas/SA03.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
