Why Record Matching Algorithms Matter

As information exchange plays an increasingly prominent role in bending the healthcare quality and cost curves, many hospital executives are wondering: what is the secret behind improving Medicare/Medicaid EHR incentive programs, ACOs and HIEs?

The answer:  the ability to access real-time clinical data wherever a patient encounter takes place to inform care decisions, avoid preventable adverse events, eliminate diagnostic and therapeutic duplications and shorten lengths of stay.

This solution, however, does not come without potential challenges.

It is imperative that individual hospitals and healthcare systems ensure the integrity of their data to guarantee success and accuracy. The accelerated adoption of EHRs and other clinical systems is creating the technical infrastructure required for data exchange, but this can create complications for data integrity.

As more systems are interfaced, significantly higher volumes of data are flowing into patient records and other clinical, administrative and financial systems. An error at any point along the way — an incorrect birth date, transposed digit in a Social Security Number, missed middle initial or misspelled name — quickly snowballs as the information feeds from one system to the next.
Fortunately, many hospitals are aware of the potential complications and are taking proactive steps to eliminate duplicate and overlaid patient records from their master patient index (MPI). But the problem is not solved once the MPI is clean. In order to maintain the long-term integrity of data, certain steps must be taken to prevent new duplicates from being created, as well to quickly identify, validate and reconcile any that sneak into the system.

Those steps must start with ensuring that record matching algorithms are capable of efficiently and accurately identifying patients. This isn’t necessarily the first thought during the health IT planning and selection process, but it should be.

Too often, facilities rely on the record matching algorithms contained within whatever clinical information system is ultimately chosen. However, these systems typically contain basic or intermediate algorithms that rely primarily on deterministic matching. The result? An unacceptably high number of false alarms (the algorithm incorrectly identifies two records as belonging to one person) and/or false negatives (the algorithm misses identification of a true duplicate).

While these will always occur, systems that utilize more advanced probabilistic algorithms focused on statistical and/or mathematical matching will deliver better rates of each. The result is fewer duplicates entering the system, driving a much cleaner MPI and ensuring that the data that is ultimately shared is — and remains — accurate.