KEYWORDS:
Data quality, de-duplication, data matching
BACKGROUND:
The Utah Statewide Immunization Information System (USIIS) accepts immunization data from multiple sources. The number of possible duplicate records in USIIS has been increasing since 1998 when significant data loads began. There are currently 906,888 patient records in USIIS, which contains 6.2% of possible duplicates.
OBJECTIVE(S):
To explain an improved data quality plan which includes a more efficient electronic matching process and a plan for more efficient data collection.
METHOD(S):
Match criteria will be expanded to include address, phone number, mother’s first name, guardian’s name, in addition to first name, last name, middle name, date of birth, mother’s maiden name, and social security number that are currently used. While the current matching procedure heavily relies on “exact” match of two records, the new procedure will exclude punctuation such as hyphens, apostrophes, periods and spaces to increase the chance of match. A “fuzzy match” process will be used to locate the candidate match that would have otherwise been missed. Quarterly newsletters, monthly e-mails, web site information, semi-annual meetings, and data quality incentive awards will be utilized to increase the users’ awareness of data problems and prevent problems at the source.
RESULT(S):
The new load/match procedure will be completed in July of 2002. Tests will be conducted to document improvement in the match process. Increased user communication is currently being implemented and data quality improvement will be quantified in the beginning of 2003.
CONCLUSIONS(S):
Including more parameters in the load/match procedure and increasing user training/communication will decrease the possibility of duplicate records.
LEARNING OBJECTIVES:
Understand processes that can be implemented to increase data quality.
Handout (.ppt format, 554.0 kb)
Back to Protect: Data Quality — Part I
Back to Contributed Papers
Back to The 2002 Immunization Registry Conference of CDC