How Data Fields Differ and How HIOs Can Address Resulting Matching Issues
As technology evolves, so too must the industry utilizing it. For healthcare, this means keeping up with a continuous movement of data and patient information that is often sensitive and private in nature.
Herein lies the problem. As health IT has evolved, healthcare has struggled mightily to maintain data integrity, as information flows from one system to another. Incomplete patient data is a byproduct of this exchange and is increasingly detrimental to patient safety.
You may be surprised to learn that the solution to this historically intractable problem doesn’t solely rely on technology. Rather, it requires a human component.
The first step toward resolution is creating an understanding among HIOs that linking records electronically is achieved primarily via matching demographic data fields from the patient records. The problem is that provider organizations are often using different levels of standards, a situation exacerbated when they try to address how data is entered into or captured by the system.
For example, HL7 has a separate field for first, middle, and last name and suffix. Some EHR systems store data separately, while others don’t. If there is only one field for the patient’s name, then you’re at the mercy of how the registrar enters the name. Often, there is inconsistency in how data is entered, with no data integrity team there to double-check for mistakes.
Research Just Associates conducted in conjunction with The College of St. Scholastica revealed that 60 percent of duplicates have two or more data discrepancies. Because this number is so high, it is unlikely that data received by an HIE/HIO will line up correctly. In addition, if an organization isn’t using a sophisticated algorithm to detect potential duplicates, the likelihood of the HIE receiving incomplete or inaccurate data is even greater. As we become more reliant on technology, it is ironic that the most important element of the data quality solution is the presence of a human, specifically a data manager.
Organizations with a data manager available to review incoming data by source are better equipped to more accurately identify their challenges, make additions to incomplete data and modify their data strategy. Beyond that, it is imperative that facilities and organizations implement a sophisticated patient matching algorithm to help identify duplicates before they pollute their database. Finally, cleaning the data on an ongoing basis, reporting errors back to the source, and creating system rules prevent dirty data at the source.
No technology or data policy can capture and correct 100% of human errors, which is why a data management position is essential to any successful data quality strategy. This, in conjunction with established policies and procedures – for example standard naming conventions or search routines – for front-end and back-end staff to follow, is foundational for the overall data integrity process.
