Legacy data migration offers challenges, opportunities
CHICAGO—Although health IT professionals agree that the utilization of data standards can enhance business operations, allow interoperability of systems and improve patient care, adoption of standards can be challenging due to legacy data and a lack of consensus as to which standards should be implemented, according to Geraldine Wade, MD, principal and managing director of Atlanta-based Clinical Informatics Consulting.

Wade, former director of medical informatics and information management at Hewlett-Packard, offered her thoughts on enterprise standards selection in a narrated electronic poster presentation at the 2009 Healthcare Information and Management Systems Society (HIMSS) conference.

She advocated a four-step approach for selecting standards.

"First, one must analyze the use case or the problem to be solved," she said. "Next, look at the big picture of enterprise goals, industry trends and adopted standards. Then, evaluate your existing resources such as people, processes and tools. Lastly, develop a migration plan to transform legacy data to standards."

In considering the goals of the enterprise, as well as what is happening in the industry, one should consider using standards that have been adopted and are widely implementable, Wade noted. She cited examples such as SNOMED CT for laboratory result contents and anatomy, HL7 for demographics, and ICD-9 for billing and financial systems.

When evaluating the enterprise resources available for implementation, all stakeholders in the initiative must be engaged in order to ensure a successful transition, Wade said. In addition to assessing the available tools within an organization, the health IT administrator also may want to consider bringing in outside help for a project.

"Utilize domain experts that can help define the goals and help evaluate the data sources," Wade offered.

The migration plan for legacy data to a standard is an opportunity to give added value to existing data by allowing its interoperability with other systems in the enterprise. Wade suggested that one choose "high value" data--such as those associated with costly diseases--to migrate first. She emphasized that data migration involve a thorough understanding of data mapping and recommended the engagement of domain experts for this process.

"Standards are critical components for healthcare data representation and exchange," she said. "They enable data aggregation, data sharing and interoperability between systems. They have different functional roles, are adopted for specific domain areas in healthcare and their implementation can improve patient outcomes and lead to new discoveries."