Business Intelligence and H Healthcare lth MHPW 2010, 15-16 - - PowerPoint PPT Presentation

Business Intelligence and H lth Healthcare

MHPW 2010, 15-16 February Eindhoven, the Netherlands

Information Systems Group Uzay Kaymak

Overview

Goal: discuss role of business intelligence (BI) in the healthcare in relation to information systems y Outline Outline

• Affordability of healthcare

B i i t lli

• Business intelligence
• BI in the healthcare
• Role of BI in the healthcare
• Possible directions for research

Possible directions for research

• Conclusions

Dutch Health Expenditure

Source: CBS

Growth of Dutch Health Expenditure

Source: World Health Organization

Growth of Elderly Population

Source: World Health Organization

Business Intelligence

From data to information to knowledge

• Definition:

Business intelligence is composed of methods that Business intelligence is composed of methods that enhance efficiency and facilitate decision making by integrating information and processes with the use integrating information and processes with the use

• f tools that transform data into useful and

actionable information. actionable information.

Many Success Stories

• Real-time business intelligence solution at Continental

Airlines (CA): CA uses business intelligence to gather ( ) g g and store data from the flight manifest, customer information, reservation data, departure times, gate p g locations and real-time flight data in a data warehouse. Continental’s analytical toolset combines the data, determines which passengers will miss a connecting flight and combines this information with customer data to determine whether special arrangements have to be made to best help these customers. (Wixom, 2008).

Business Intelligence Systems

Reporting and Reporting and visualization Data integration Analytics

Azvine et al. (2005)

Analytics Capability

What’s the best that can happen? What if these trends continue? What will happen next? Why is this happening? What if these trends continue? What actions are needed? Where exactly is the problem? What happened? How many, how often, where?

Source: SAS

Crisp-DM Data Mining Framework

Reporting

• Summarization
• Visualization
• Dashboards
• Monitoring
• Definition of KPI’s

BI Framework for Healthcare

Mettler & Vimarlund (2008)

Benefits of BI in Healthcare

Information

• Reduction in time and effort

Information

• Reduced amount of referrals
• Improvement in documentation
• Reduction in time and effort

spent on KPI tracking

• Time savings from integration of

Improvement in documentation and archive quality and usability

• Executive dashboard (monitoring

g g processes ( g

• f operational performance)
• Possibility of data analysis to

id tif d di t t d Costs

• Implementation of dashboards

hi h di l d ti it d identify and predict trends Effi i which display productivity and quality metrics

• Improved decision making for

Efficiency

• Improved scheduling and

responsiveness

• Improved decision making for

achieving strategic objectives responsiveness

• Decreased no show rate

Examples

• Awareness

(definition of KPI’s) (definition of KPI s)

• Benchmarking
• Benchmarking
• Advanced planning
• Advanced planning

and scheduling

Business Drivers for BI in Healthcare

Does IT Really Reduce Costs?

• Thomas et al. (2009)
• Telemedicine for ICU patients does not reduce mortality

Telemedicine for ICU patients does not reduce mortality

• r costs
• Himmelstein et al. (2010)

Computerization quantified by number of Computerization quantified by number of applications

• No evidence of savings on clinical or administrative

No evidence of savings on clinical or administrative costs

• Some evidence of slight improvement in quality of care

Some evidence of slight improvement in quality of care

Role of BI

Ad-hoc medicine Clinical decisions Process Medicine Patient pathways

• Clinical decisions
• Specialists
• Patient pathways
• Also practitioners
• Variable/flexible

Problem sol ing

• Repeated/structured

Val e addition

• Problem-solving
• DSS, expert systems
• Value addition
• WfMS, PAIS

p y

• Exploration
• Optimization

Business Intelligence Systems BioMedical Technology

Favorable Evidence for IT

• Garg et al. (2005)
• Clinical DSS improve practitioner performance

Clinical DSS improve practitioner performance

• Deveraj & Kohli (2000)
• Deveraj & Kohli (2000)
• Business process reengineering needed to reduce

costs improve quality and increase patient satisfaction costs, improve quality and increase patient satisfaction

Amarasingham et al (2009)

• Amarasingham et al. (2009)
• Computerized patient records improve profitability

potential and quality potential and quality

Digitize Your Data!

21

Business Process Redesign

Workflow Management Benefits

0 08 0.07 0.08 0.05 0.06 0.03 0.04 0.01 0.02 Without workflow With workflow

Concept maps

• Maps that show related terms in a knowledge field
• Concepts that appear in the same document are

Concepts that appear in the same document are assumed to be (semantically) related

• Co-occurrence as association measure:
• Co-occurrence as association measure:

ij

c N a

jj ii j ij

c c N a ⋅ =

• Using the association measure, concepts can be

visualized in a 2-d space, such that the closer two terms, the more related they are

Case Study: corpus selection

• Abstracts of scientific papers indexed by ISI Web of

Science

• Search query:

(“information system” OR “information systems” OR “information management” OR “information g technology” OR “information technologies” OR “e- health” OR telemedicine OR “ICT”) AND (“health ) ( care” OR healthcare)

Concept Map View

Concept Density View

Cluster Density View

Detail View Information Systems

IS Concepts

Conclusions

• Healthcare industry is going through a large

transformation due to extensive use of IT (comparable to the introduction of ERP in production industry a couple of decades ago) y p g )

• Business intelligence is a key technology to extend

Business intelligence is a key technology to extend the reach of process medicine

• Research should investigate integration of BI

systems and PAIS systems and PAIS

Acknowledgements

• Nees Jan van Eck
• Ludo Waltman
• Geert Dela Haije
• Arno van den Berg