KEYWORDS:
Data quality, deduplication, connect
BACKGROUND:
The New York City Department of Health and Mental Hygiene (NYC DOHMH) is deploying a Master Client Index (MCI). The MCI will provide a department-wide hub that will enable any system to accurately match and identify any person being entered into a participating system. The Citywide Immunization Registry (CIR) is one of two systems participating in the initial deployment of the MCI. The MCI uses ChoiceMaker 2 from ChoiceMaker Technologies (CMT) as its core record matching/deduplication engine.
OBJECTIVE:
Allow easy deployment of advanced record matching technology.
METHOD:
DOHMH participation was initially required in three areas:
1. Marking individual pairs of records as a “match”, a “non-match”, or “unsure”;
2. Supervising a scientific test of ChoiceMaker 2 on data unseen by CMT
3. Reviewing records that ChoiceMaker marked as “unsure” in production
New tools and techniques developed by CMT have considerably reduced deployment effort.
RESULT:
Step 1 is no longer necessary. CMT now only requires human judgments in steps 2 and 3. The addition of a “Communicable Disease Surveillance System” to the MCI is requiring considerably less client effort.
In recent tests, ChoiceMaker identified 35,000 duplicate records in a previously not deduplicated birth cohort. Since DOHMH staffers average 150 records per day when reviewing possible duplicates, the MCI’s automatic deduplication has saved DOHMH at least 233 person-days per birth-cohort.
CONCLUSION:
NYC’s success using advanced record matching techniques can be duplicated at other registries which have smaller quality assurance staffs.
LEARNING OBJECTIVES:
Participants will learn the steps that NYC DOHMH took to deploy the MCI and how technological advances enable new systems to be deployed with modest effort.
Web Page: www.choicemaker.com/press%20releases/nyc_doh_contract.htm
Back to Winning the Match Game: Improving Deduplication Processes
Back to The 2003 Immunization Registry Conference (October 27-29, 2003)