Assessment of Multiple Data Sources Results in Increased Data Quality Standards in IIS

Background:
The Kansas Immunization Registry (KSWebIZ) has implemented several interfaces with different data sources and as the number of client and vaccination records has increased, so has the number of duplicates (patient and vaccine level). As data quality decreased, the need for re-evaluation of data quality protocols increased.

Objectives:
Improve data quality by documenting data flow; establishing business rules for data prioritization; and developing tools for data de-duplication.

Methods:
• Examined the different types of data sources connected to KSWebIZ and evaluated the type of connection (i.e. direct entry, batch load, two-way interface); the timeliness of exchange (i.e. as the patient was seen, nightly, weekly); and the type of data (i.e. legacy loads vs. real-time, billing vs. encounter dates, vital statistics).
• Prioritized data based on source and quality.
• Developed business rules for de-duplication at both the vaccine and patient level.
• Created tools to assist both manual and electronic interventions.

Results:
• Developed matrices for both patient and vaccination de-duplication for the different sources of data to determine which record to keep in each scenario.
• Developed data quality protocol to be conducted before and after the provider site was enrolled for pre-certification of source and continuous data quality performance reviews.
• New reports were developed to filter vaccination duplicates on a 10 day window within the same antigen group.
• Enhanced patient duplicate report so that providers could access a list of patients in their jurisdiction that were identified as possible duplicates.

Conclusions:
Evaluation of data sources connected to the IIS can create universal data quality standards and tools to improve system integrity.