Translated and updated from the original presentation at the seminar of “Openness and the Future of Healthcare IS” by Service Factory, Aalto University School of Business at 18 th of March 2014. OPEN DATA - Critical Capability in Healthcare Information Production Processes Sami Laine, Doctoral Student Department of Computer Science and Engineering, Aalto University Email: [email protected]
26
Embed
Open Data - Critical Capability for Open Healthcare
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Translated and updated from the original presentation at the seminar of “Openness and the Future of Healthcare IS” by Service Factory, Aalto University School of Business at 18th of March 2014.
OPEN DATA - Critical Capability in Healthcare Information Production Processes
Sami Laine, Doctoral StudentDepartment of Computer Science and Engineering, Aalto UniversityEmail: [email protected]
Benchmarking claimed significant productivity differences in neurology
specialty
Pirkanmaa Hospital District
Hospital District of Southwest
Finland
1,120,71
Laine, S., Niemi, E. (2013), “Transparency of Hospital Productivity Benchmarking in Two Finnish Hospital Districts”, In the Proceedings of the 29th annual Patient Classification Systems International (PCSI) Conference , Helsinki, Finland.
For example, fragmentation bias rewards splitting and heterogeneity
Hospital District
Hospital District
32
DRG X
Episode A
DRG X
DRG A
DRG B
DRG C
Less productio
n but more
health for same money!
More production but less
health for same money!
Laine, S., Niemi, E. (2013), “Transparency of Hospital Productivity Benchmarking in Two Finnish Hospital Districts”, In the Proceedings of the 29th annual Patient Classification Systems International (PCSI) Conference , Helsinki, Finland.
IPP has three problem themes: Human Errors, Software Features and Obscurity
Data Warehous
e
Scripts produce external reports
Data Warehous
e
Data Entry Errors
Application Feature Bias
Architecture Bias
Scripting Error
Interpretation Mismatch
Laine, S., Niemi, E. (2013), “Transparency of Hospital Productivity Benchmarking in Two Finnish Hospital Districts”, In the Proceedings of the 29th annual Patient Classification Systems International (PCSI) Conference , Helsinki, Finland.
IPP has three problem themes: Human Errors, Software Features and Obscurity
Data Warehous
e
Scripts produce external reports
Data Warehous
e
Data Entry Errors
Application Feature Bias
Architecture Bias
Scripting Error
Scripting Error
Obscurity Obscurity
Obscurity
Obscurity
Laine, S., Niemi, E. (2013), “Transparency of Hospital Productivity Benchmarking in Two Finnish Hospital Districts”, In the Proceedings of the 29th annual Patient Classification Systems International (PCSI) Conference , Helsinki, Finland.
Inconsistent figures have been produced about the same issue at the same time…
Ambulatory Procedures in
Administrative Reports
Ambulatory Procedures in Operation
Room Reports
15687
10726
Which one is correct
?Laine, S. (2012), "APC-SIMULATOR: Demonstrating the Effects of Technical and Semantic Errors in the Accuracy of Hospital Reporting" In the Proceedings of the 17th International Conference on Information Quality (ICIQ), Paris, France.
Inaccuracies and semantic mismatches exist in healthcare data
“Patient bills”“Manually duplicated
codes”
“Planned procedures”
15687
10726
Both are more accurate than
expected but only for a specific
purpose.
”Actually PerformedAmbulatory Procedures”
- X %+ Y %
There exists semantic
mismatches and error rates between contexts for good
reasons.Laine, S. (2012), "APC-SIMULATOR: Demonstrating the Effects of Technical and Semantic Errors in the Accuracy of Hospital Reporting" In the Proceedings of the 17th International Conference on Information Quality (ICIQ), Paris, France.
Open Data means also Open Methods and Open Processes
Local Data Warehous
e
National Hospital
Benchmarking
Scripts produce external data sets
National Data
Registry
Scripts produce public reports
DATA METHODSOPEN DATA
Scripts produce external data sets
Scientific Data Set
Scripts produce external analysis
PROCESS
Information Production Process of Researchers
Information Production Process of Health Service Providers
Information Production Process of National Organizations
The problem is often the
obscurity of actual data
creation situations
Another problem is earlier
information production processes
Medical Imaging System
Electronic Patient Record
Scriptscollect data
Scriptscollect data
Management Reports
Scripts produce internal reports
Research Results
DATA SUPPLY DATA MANUFACTURING DATA CONSUMPTION
IPP capable for openness can be secured or opened at any phase
Local Data Warehous
e
National Hospital
Benchmarking
Scripts produce external data sets
National Data
Registry
Scripts produce public reports
PRIVATEOPENNES
S
Scripts produce external data sets
Scientific Data Set
Scripts produce external analysis
Information Production Process of Researchers
Information Production Process of Health Service Providers
Information Production Process of National Organizations
What aggregation level can be opened to
confidential or even public use?
“Truly Ideologically Open Data”
CONFIDENTIAL
OPEN
Medical Imaging System
Electronic Patient Record
Scriptscollect data
Scriptscollect data
Management Reports
Scripts produce internal reports
Research Results
DATA SUPPLY DATA MANUFACTURING DATA CONSUMPTION
Interfaces, Processes and Methods can be always made open
Local Data Warehous
e
National Hospital
Benchmarking
Scripts produce external data sets
National Data
Registry
Scripts produce public reports
DATA MODELS!APPLICATION INTERFACES!
TOOLS!SCRIPTS!
ALL OPEN!
Scripts produce external data sets
Scientific Data Set
Scripts produce external analysis
PROCESSES
WORKFLOWS
Information Production Process of Researchers
Information Production Process of Health Service Providers
Information Production Process of National Organizations
These should be OPEN at least for INTERNAL USE!
Medical Imaging System
Electronic Patient Record
Scriptscollect data
Scriptscollect data
Management Reports
Scripts produce internal reports
Research Results
DATA SUPPLY DATA MANUFACTURING DATA CONSUMPTION
Open Data supports Evidence-Based Management, Open Government and Open Science
Local Data Warehous
e
National Hospital
Benchmarking
Scripts produce external data sets
National Data
Registry
Scripts produce public reports
DATA METHODS OPEN X!
Scripts produce external data sets
Scientific Data Set
Scripts produce external analysis
PROCESSES
OPEN SCIENCE
EVIDENCE-BASED MANAGEMENT
OPEN GOVERNMENT
Medical Imaging System
Electronic Patient Record
Scriptscollect data
Scriptscollect data
Management Reports
Scripts produce internal reports
Research Results
DATA SUPPLY DATA MANUFACTURING DATA CONSUMPTION
Closed Data leads to traditional habits and beliefs, insider power politics and flawed science.
Local Data Warehous
e
National Hospital
Benchmarking
Scripts produce external data sets
National Data
Registry
Scripts produce public reports
DATA METHODSClosed & Obscure
Scripts produce external data sets
Scientific Data Set
Scripts produce external analysis
PROCESSES
OBSCURITY
OBSCURITY
OBSCURITY
Flawed Science
Habits and Beliefs
Insider Politics
Distributing Data
Typically Open Data Product Data Organization Data Service Data Administrative Data Customer
Instructions
Potential but challenging Open Data Distributing Sensitive
Data by using External Authorization
Distributing Sensitive Data in Aggregated Form
Acquiring Data
04/11/2023Laitoksen nimi 18
Typically Open Data Geographical data Aggregated
demographical data Aggregated statistics News and social
media feeds Reference data
Potential but challenging Open Data Full-scale master data
(i.e. customers, services etc)
External service transactions (i.e. Health services, transportation services etc)
Sensitive individual level demographical data
All Information Production Processes should have Open Data capability
Laitoksen nimi
Open Data should not be ”ideological product” but ”practical capability” that is utilized everywhere
Open Data as capability means Open Data, Open Methods and Open Processes For internal use For external users
However, The level of Openness of Data Sets is controlled according their content requirements.
Data SetsData ModelsApplication Interfaces
ToolsWorkflowsDocuments
Open Data Capability
All processes
Every workflow
Prevents Vendor-lock-in• Open standards and Application Interfaces• Open collaboration between stakeholders and use cases
Improves Transparency• To own healthcare service production• To own information management and software systems
Supports Quality Control• Own internal activity (patient services or medical device
maintenance)• Service providers (e.g. software or logistics)• External Benchmarking (e.g. international productivity
benchmarking)
Open Information Production brings Transparency and Interoperability to Software Systems and Healthcare
Application for Strategic Research Opening at 11.6.2012
Project time schedule 1.9.2012-31.8.2014
QUALIDAT
QUALITY OF DATA FOR VALID DECISIONS
Supporting Diverse Uses and UsersNieminen, Marko (prof.)Rossi, Matti (prof.)Borgman, JukkaKaipio, JohannaLaine, SamiMahlamäki, KatrineNiemi, Erkkahttp://qualidat.aalto.fi/
Scientific publications about open data and open science
Nosek, B. A., & Bar-Anan, Y. (2012), Scientific Utopia: I. Opening scientific communication. Psychological Inquiry, 23, pp. 217–243.
Nosek, B. A., Spies, J. R. and Motyl, M. (2012), Scientific Utopia: II. Restructuring Incentives and Practices to Promote Truth Over Publishability. Perspectives on Psychological Science, 7(6), pp. 615-631.