2. Unraveling the Current Information Tangle
Insurance
Companies
Physicians
Specialty
Providers
Hospitals & Clinics
Patients
& Families
Lab tests &
XRAYs
Medications
Public Health
2
* One Directional
* Not Real-Time
* Lacks Standards
3. Desired Solution
Health Information Exchange
Insurance
Companies
Physicians
Specialty
Providers
Hospitals & Clinics
Patients
& Families
Lab tests &
XRAYs
Medications
Public Health
HIE
3
Standards govern:
Form, Content and Response
5. 5
SOM HIE
Doctors EHRs
SUB-STATE HIE
Michigan HIE Model
Early 2012
SOM HIE Phase I
* Implement interim solution for HL7 messages
* Direct Connect to SOM HIE
* Send immunization records to MCIR
* Compliance with CDC Grant
* Send hospital reportable labs to MDSS
* Ability to send message acknowledgements
6. 6
SOM HIE
Medicaid
Doctors &
Community
Providers EHRs
SUB-STATE HIEs
Michigan HIE Model
Current
Doctors &
Local PH Dept EHRs
SOM HIE Phase II
* Implement production solution
* Establish MiHIN – SOM HIE connection
* Continued Infrastructure Development
* Gather new functional requirements
7. 7
Medicaid
DW
MSSS
SOM HIE Phase III
*MiHIN – SOM HIE – connection via
VPN
*Next SOM Use Cases
* MSSS
* MCIR forecast/history
* State Labs (StarLIMS)
*HPD – Health Provider Directory
* Not Limited to Licensed Providers
Michigan HIE Model
Planning
State Labs
StarLIMS
8. 8
Medicaid
DW
MSSS
*Match person-level
records across data sets
*Link data from multiple
data sources for an
individual person
*Include data sets beyond
HIE
*Phased implementation
approach
SOM Master Person Index
State Labs
StarLIMS
9. *
*Affordable Care Act and other healthcare
initiatives are driving data sharing
*Comprehensive view of individuals in multiple
programs making it possible to
*effectively assess and analyze healthcare
program data
*make better and faster decisions
*manage and measure programs
*reduce costs
*improve outcomes
9
11. *
*Completed March 2012
*Deployed the Initiate Citizen Hub
*Implemented Initiate Inspector for data
resolution
*Replaced Data Warehouse UCI
*UCI originally implemented 2001
*Configured single input and output streams
combining 17 data sources
*Initial conversion included 38 million records
11
12. 12
MPI Phase II
SSHIE
- PID
DC -
PID
SSO
- PID
SSO -
DDE
* Establish real-time interface
between MCIR and MPI
* Data Warehouse to accept
MCIR updates from MPI
* Establish process to add
additional MDCH systems
MPI Future Phases
* MDSS and StarLIMS
* other SOM systems
* Household identification
13. 13
Step 1: Optimizes data for statistical comparisons
– Normalizes & compacts data, creates derived data layer,
source data remains intact
– Phonetic equivalences, tokenization, nicknames, etc.
Step 2: Finds all the potential matches
– Casts a wide net – all matches on current or historical attributes, prevents misses
– Partial matches, reversals, anonymous values, etc.
Step 3: Scores accurately via probabilistic statistics
– Compares attributes one-by-one and produces
a weighted score (likelihood ratio)
– Frequency weights specific to your business
– Edit distance, proximity of match
– Allows custom deterministic rules
Step 4: Custom threshold settings
– Single or dual threshold models
– Link, don’t link, don’t know – “learns”
from manual input
– Manage cost/quality trade-offs
– Manage the linkages, workflow review
Manual Review
Lowest
Possible
Score
Highest
Possible
Score
Don’t
Link Link
Lowest
Threshold
Upper
Threshold
Should be linked
Should not be linked
*
14. *
14
Alerts data stewards to potential data quality
and relationship issues delivering the ability
to collaboratively inspect data, visualize
relationships and resolve issues
Inspect and resolve data
quality issues
Resolve duplicates within same
source
Manually link records across multiple
systems
Visualize and manage relationships
Centralize and leverage data across the
enterprise…with a single tool
IBM® INITIATE® INSPECTORIBM® INITIATE® INSPECTOR
Issue
Resolution
Hierarchy/
Relationship
Management
Central
Data Store
15. 15
CONSUMING
SOURCES
DATA
SOURCES
Source 1 Source 2 Source ‘n’Source 3
System 1 System 2 System 3
Catherine Lamb
763543Dr Kath J. Jones 1:N4456 15/06/1970 9263462232 Sussex St 0415266721
763543Dr Kate Lamb 2:2736 Female 02-9263-4622
763543Mrs. K. Jones 3:S7846 15/06/1970 9263-4622Level 1, 32 Sussex Rd +61415266721
763543n:97662 15/06/2006 Female 92630-6000 0415-266-721
60558
60558
60558
Local ID: EID:Name: DOB: Sex: Home Phone:Address 1: Mobile:Zip Code:
763543Dr Kath J. Jones 15/06/1970 Female 02 9263 4622 0415 266 721Level 1, 32 Sussex St 60558
System ‘n’
*
16. *
*Administrators
*Assign tasks to data stewards
*Resolve tasks
*Link records
*Identify target record
*Assign trusted source
*Data Stewards
*Review potential linkages
*Resolve data integrity issues
*Ensure source-record quality
*Monitor and maintain data accuracy
16
17. *
*Limited to authorized users (data stewards)
and to specific data/levels of information
*Identifier information is not shared
*Data can be linked without being visible
*Oversight by MDCH Security Office
17
18. *
*Collaboration with MiHIN for development of
Health Provider Directory
*Leverage development to expand to Provider
Index
*Include provider/organizations beyond HIE
18