Clinical data warehouse

From Citizendium
Jump to navigation Jump to search
This article is developing and not approved.
Main Article
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
This editable Main Article is under development and subject to a disclaimer.

A clinical data warehouse or CDW is a facility that houses all electronic data collected at a clinical center [1]. For any modern clinical institute, it is necessary to separate operational data from informational data by creating a clinical data warehouse [2]. A growing number of technologies for integrating and performing structured analyses of data from disparate sources are competing to win the day for healthcare organisations [3].

The TrialDB clinical study data management system of Nadkarni et al was the first to use multiple EAV (Entity-Attribute-Value model or Object-Attribute-Value Model or Open Schema) tables, one for each database management system (DBMS) data type. The EAV/CR (with Classes and Relationships) framework, designed primarily by Luis Marenco and Prakash Nadkarni [4], overlaid the principles of object-orientation on to EAV; it also built on Slezak's object table approach. Both TrialDB and EAV/CR are open-source, though they are built on Microsoft technologies rather than Java/Linux. In Oracle Clinical, a package for the management of clinical trials data; the clinical data component is stored in an EAV structure where all values are coerced to the varchar2 data type. Here a major DBMS vendor has decided that, despite all the difficulties and hazards of working with EAV, some circumstances make its use unavoidable. [5]

The solution to warehousing the ever-changing structures of clinical data will require (a) a single, multi-trial repository based upon abstract rather than directly representational data models, and (b) adaptive, meta data-driven ETL (extract, transform and load) programming [6].

A virtual laboratory is a managed, networked collection of near patient and traditional laboratory instruments. It is Taylor et al's thesis [7] that dramatic changes in the practice of laboratory medicine will emerge. Information from point-of-care devices and laboratories will be treated in a unified manner and all devices will become components of integrated internet connected virtual laboratories. Instruments will include devices located in traditional laboratories, such as complex multi-channel analyzers and specialty laboratory tests instruments, along with point-of-care devices located at a variety of sites including clinics and at patients’ homes. Virtual laboratories will combine the immediacy offered by point-of-care devices with services offered by traditional laboratories such as quality assessment, quality control, data capture, and the dissemination of results to patients and clinicians.

JANUS [8] is a standards-based clinical data repository that utilizes the open source data model, Janus. Janus was created by FDA in partnership with IBM under a CRADA. This repository provides a data collection and analysis warehouse for clinical trial data submitted for protocols (what was supposed to happen) as well as clinical outcomes data (what did happen - events, interventions, etc.).


  1. Data Warehouse. Retrieved on 2008-02-06.
  2. Clinical Data Warehousing - Clinfowiki. Retrieved on 2008-02-06.
  3. How business intelligence is making healthcare smarter — NHS Connecting for Health. Retrieved on 2008-01-30.
  4. The EAV/CR Data Model. Retrieved on 2008-02-06.
  5. Ask Tom "Query on design". Retrieved on 2008-02-06.
  6. Optimal Data Architecture for Clinical Data Warehouses. Retrieved on 2008-02-06.
  7. Design of an Integrated Clinical Data Warehouse. Retrieved on 2008-01-30.
  8. Portal - Janus. Retrieved on 2008-02-06.