Data warehousing as a healthcare business solution

Healthcare Financial Management, Feb, 1998 by Ronald Scheese

Despite making significant investments in computer technology

to process, accumulate, and store massive amounts of patient information,

healthcare organizations still lack the tools to leverage the data stored

in various information systems throughout the organization. Investments in

healthcare information systems (HISs) to date have not helped end users to

better understand their overall business and compete more effectively in a

rapidly changing environment.

An increasing number of healthcare organizations are realizing the strategic

benefit of using data warehouses to tap the data stored in their systems.

This emerging information system technology trend is so significant that

experts estimate that every healthcare organization will have a data

warehouse tool in place within five years. a

A data warehouse is a single information repository for a massive collection

of operational and financial data, arranged to organize and deliver the data

to individuals across the enterprise who can then easily run queries,

produce reports, and perform analyses for strategic and operational decision

making.

One benefit of a data warehouse is that managers can use it to improve their

operational performance. If operational managers work in an environment in

which day-today decision making is tightly controlled, a data warehouse can

provide information to support decisions. If, on the other hand, managers

operate in an accountable, empowered environment, a data warehouse can

increase the managers' ability to access and analyze operational data,

thereby improving operational performance.

One critical element of a data warehouse is its ability to deliver

information to the end user. The availability of word processing and

spreadsheet tools has created increasingly self-sufficient end users who

manipulate data themselves. Implementation of a data warehouse not only

satisfies user demands for ad hoc analysis and reporting capabilities but

also substantially reduces information system staff analytical and reporting

responsibilities, thus freeing HIS resources for strategic operations.

Warehouse Data Processing

Data warehousing facilitates both production and ad hoc data processing.

HIS production processing, often described as on-line transaction processing

(OLTP), focuses on an organization's day-to-day business needs, such as

patient registration, billing, and payroll. When OLTP systems stop

operating, an organization's more routine functions are significantly

disrupted. HIS ad hoc processing, often described as on-line analytical

processing (OLAP), retrieves, analyzes, reports, and shares data from

disparate systems, vendors, and departments, such as decision support

systems, executive information systems, budgeting, and the emerging clinical

data repositories. When OLAP systems are not operating, there is no

immediate or obvious business operations concern. The analytical system,

however, is a critical tool for ensuring an organization's long-term

competitiveness and profitability.

OLAP information needs differ significantly from OLTP information needs. For

example, billing a third-party insurance carrier is a production function,

whereas determining how much and how soon a carrier will pay or how much

volume that carrier has produced for the healthcare organization is an

analytical function and contributes to an understanding of the organization.

The two processing platforms have incompatible operating design needs and

are fundamentally different applications (see Exhibit 1). To use an analogy,

the OLTP system is like a manufacturing factory operating in a production

mode, whereas the OLAP system is like a warehouse, which collects,

organizes, and groups a product for future distribution. In an information

systems environment, however, the product - information - is not tangible.

Exhibit 2 shows a complex data warehouse design in which information is

extracted from multiple systems, mapped into a single warehouse, and then

accessed and analyzed at a data-mart level. A data mart is a summary, or

departmental, warehouse that represents a component of a larger data

repository. Designs such as the one depicted in Exhibit 2 may be established

to speed the query and [TABULAR DATA FOR EXHIBIT 1 OMITTED] reporting

capabilities of very large databases, to achieve additional information

security, or to increase the ability of local facilities to perform detail

transactions at a level that might otherwise be lost in a data warehouse.

Cost Justification

Much of the healthcare industry's investment in HIS technology historically

has focused on automating transactions for operations. The cost

justification for this type of information technology spending has been

straightforward and has almost always been based on a cost/benefit analysis.

The cost justification for data warehousing technology, however, is not

straightforward. A data warehouse is concerned with the creative and

analytical aspects of a business, with a focus on applications that extract

and manipulate data. As such, there is no simple, equivalent