Index - Major Sections
Home

**InHCc HMIS**

Site Map
 Health Economic and Reform

Benefits

Discussion

Data and Data Analysis

Health Management

Product and Services
References
Team

_______________

Index - Same Level Subject


 

Index - Child Subjects

Introduction

The Data Warehouse is the repository of information from the Clinical System. Whereas the Clinical system is the on-line real time transactional database, the data warehouse is the "read only" access to the historical information. 

Purpose

A data warehouse stores stable, verified data values.

  • A transaction database helps people carry out activities, and a data warehouse helps people make decisions. For example, a transaction database might show which clinics have a medical specialist so that a client can be referred. A data warehouse, on the other hand might show the historical pattern of how many requests for a medical specialist took place and in what areas so that a manager can decide to adjust staffing in the future.
  • A transaction database is volatile; its information constantly changes as new clients are seen. A data warehouse is stable; its information is updated at standard intervals--perhaps monthly, weekly, or even hourly--and, in an ideal world, an update would add values for the new time period only without changing values previously stored in the warehouse.
  • A transaction database focuses on the details; the physician making the referral doesn't want to know the average number of visits per month to each medial specialist. A data warehouse focuses on high-level aggregates: the manager updating staffing doesn't care which specialist is not at the present moment occupied. The implication of this difference is that the key values in a data warehouse must be numeric values that can be summarized.

Characteristics

A Data Warehouse can be viewed as a Information Database. In additional to the characteristics of the Information Database it also has the following characteristics:

  • Organized by subject 

  • Data is time-variant. It contains historical data.

  • Data is integrated from multiple data sources

  • Data is no-volatile. The individual data is never updated or deleted.  

  • Data is added in batches

  • Data is optimized for read only queries and not for data entry.

  • May requires mapping of sources to the database structure. 

Advantages

  • Does not disrupt legacy systems. Legacy systems can still interact with the data warehouse without having to change any underlining procedures.

  • Ability to pull data from any data source

Organization

There are two basic concepts on how to organize a database

  • Organized for the use as operational data

  • Organized for the use as informational data

There are differences between the types of databases.

Reasons to split of Operational data from Informational data (Innom, 1996)

  • The data serving operational needs is physically different data from that serving informational or analytic needs.

  • The supporting technology for operational processing is fundamentally different from the technology used to support informational or analytical needs.

  • The user community for operational data is different from the one served by informational or analytical data.

  • The processing characteristics for the operational environment and the informational environment are fundamentally different.

Operational Analytical
Individual transactions Masses amount of data
Frequent Updates Read Only
Quick response time Response time not as important
Normalized data Denormalized Data
Traditionally operational data Traditionally Research data

A data warehouse is not as susceptible to system failures as a transaction processing system. If the system crashes and data is lost, the data warehouse can be reloaded from its sources.

Reasons to separate the operational from the data warehouse (Innom, 1996, page-27)

Relationship to the Relational Databases

A data warehouse almost never lets the user update or add records to the database…therefore the main benefit of the normalization of the database for the relational database does not apply to the data warehouse.

Flexibility comes at a price. Performs poorly because of the necessary to create joins.

Relationship to Object-oriented databases

The elimination of the table-row-column structures of relational data bases and substituted directly into the database engine, the concepts of classes and subclasses (or types and subtypes), objects, properties, and methods.

Relationship to Multidimensional Databases

Multidimensional databases throw out the idea of the conventional relations database.

To understand multidimensional databases, we must first look at the types of analytical analysis that they perform.

Relationship to Online Analytically Processing (OLAP)

OLTP applications typical access only a few tables during a given short lived transaction that can either added data or extract data from the data warehouse.

OLAP applications typical access many tables during a given long lived transaction that only reads the data.

 

Back to Top