Home  |   Subscribe  |   Resources  |   Reprints  |   Writers' Guidelines

June/July 2019

Industry Perspectives: Data Standardization Can Improve Patient Matching
By Daniel Cidon
For The Record
Vol. 31 No. 6 P. 28

Roughly $35 billion in total taxpayer subsidies have been spent on digitizing patient records. But despite this massive effort and investment in EHRs, the output hasn’t yet translated into a world-class health care system.

To truly support the nation’s transition to EHRs and embrace the vision of a highly interoperable, clinically integrated system, patients must be accurately and consistently matched to their data.

While it’s pretty clear that current EHR functionalities for patient matching are not working as well as they should, a new paper, “Evaluating the Effect of Data Standardization and Validation on Patient Matching Accuracy,” recently published in the Journal of the American Medical Informatics Association (JAMIA), reveals that standardizing and validating certain patient demographics in EHRs can improve match rates. Specifically, standardizing patient addresses using the US Postal Service’s format.

Linkage Problem Widespread
While EHRs have become commonplace, the disjointed, competitive nature of IT systems contributes to a proliferation of duplicate and incomplete patient records. Because disparate systems capture patient demographic elements in different ways, data contained in EHRs are notoriously inconsistent.

For instance, some providers or EHR systems will include hyphens, apostrophes, and suffixes in last names while others don’t. This makes it a challenge to link patient information coming from more than one provider—or even data that are siloed within a facility.

The JAMIA paper, led by researchers at Regenstrief Institute, Indiana University, and The Pew Charitable Trusts, estimates that 30 billion Health Level 7 messages containing protected health information are transmitted annually from different software applications. All of these need to be matched to a record.

A prior Pew study reported EHR matching rates within facilities as low as 80%, meaning that 1 out of 5 patients may not be completely matched to his or her record. When exchanging records outside the organization, match rates can be much worse—as low as 50%—even when the providers are running an EHR from the same vendor.

Addresses, Last Names Are Key
At its core, health care is an industry rooted in helping individuals live better, longer, healthier lives. Each year, as many as 440,000 Americans die from preventable errors in hospitals. Duplicate records and poor patient identification, which are enormously costly and place patient safety at risk, contribute to the problem.

As a result, the industry relies on patient matching algorithms. However, the outcome is only as good as the quality and consistency of the data.

The JAMIA paper found that patient matching algorithms performed better when using standardized data—most notably, addresses culled from the Postal Service’s certified standardization formats. This method led to better match rates, with some improving by as much as 3%.

In addition, when standardized rules were applied to both address and last name, match rates rose as much as 8%.

In general, a consistently formatted address can help health care organizations proactively manage, detect, and eliminate data quality issues and greatly improve match rates.

Data standards can help with such consistency, especially those that enable the clear and disambiguated data needed for use in automated patient matching solutions. A high-quality patient address, along with the patient’s birthday, is a powerful tool to gain accurate algorithm-based patient matching. The addition of an accurate phone number provides an even greater match rate.

On the Government’s Radar
The JAMIA paper made concrete recommendations to facilitate standardization, suggesting that EHR developers adopt standardization rules and prioritize incorporating address standardization functionality into their patient registration products.

Significantly, the researchers also called on the federal government to step in and take action, suggesting that the Office of the National Coordinator for Health Information Technology (ONC) encourage the use of additional data, such as e-mail addresses, that are routinely captured in EHRs.

The timing of the JAMIA paper is noteworthy; inadequate patient record matching is beginning to be recognized as a major impediment to data sharing, a priority for the government and a requirement of the 21st Century Cures Act.

While the Cures Act doesn’t specifically address patient matching, the Centers for Medicare & Medicaid Services and ONC took the somewhat unusual step of including a request for information on the issue in its proposed rules, noting that patient matching is a critical component to interoperability and the nation’s HIT infrastructure. The organizations plan to use the information to issue regulations or to provide future guidance.

HIT Tools Can Maximize Matching
The paper’s recommendation to prioritize the standardization of patient addresses is an important first step. Nevertheless, HIT can be further deployed to capture additional data, boost the patient matching algorithm’s power, and increase match rates.

Hospitals and health systems that fail to run an efficient enterprise master patient index (EMPI) platform are operating EHR systems fraught with duplicates and inaccurate patient information.

According to a 2018 survey by Black Book Research, hospitals without an EMPI in place for managing patient identification reported duplicate record rates of 18% within their organization and 24% when exchanging records out of network. This occurs because MPIs within EHRs were designed for a single vendor-based environment and lack the sophisticated algorithms for linking data across various sources and settings of care. When sent downstream, duplicate and disjointed records trigger further harm, leading to medical errors, skewed analytics and reporting, denied claims, and increased costs from repetitive tests and procedures.

Leveraging an EMPI is an industry best practice essential for promoting interoperability and helping evolving health care enterprises map an individual’s entire care journey. As a vendor-neutral, centralized platform for enabling timely, bidirectional access to patient information, an EMPI is a critical technology tool in ensuring data flow freely and accurately from provider to provider.

EMPIs that leverage location intelligence and geocoding to verify addresses allow organizations to standardize and authenticate these data in real time, thereby avoiding risks associated with duplicate record creation and identity fraud. Seamless integration of location intelligence technology enables front-end, real-time address capture as well as back-end batch address verification and enhancement.

Type-ahead, or autofill, technology verifies addresses with each keystroke to prevent errors from entering the system. EMPIs with address verification technology can reduce address entry time by as much as 78%, prevent data entry errors at the point of capture by more than 20%, and enable batch processing of more than 3 million address records per hour.

No Silver Bullet
EHR-level requirements can only go so far; ultimately, the core identity of a patient and basic associated demographics should not be in the control of any single system. Rather, this information must be externalized from such insulated applications to maintain accuracy and consistency across all connected systems within the delivery network.

Linking patient records takes considerable effort. It is easier and more effective when computer programs do the work automatically. In turn, the software’s success rates are boosted further when the data being matched are standardized.

Ultimately, in today’s transformative digital health care landscape, health organizations are going to have to view patient identification as a data governance and interoperability challenge that requires a combination of people, processes, and technology to minimize errors and improve outcomes. The ability to capture, record, and share accurate demographic data elements across systems is essential. Standards can play an important role.

Every patient record has a real life behind it with real-world consequences if that record is incomplete or inaccurate. This critical endeavor takes time, commitment, and collaboration between providers, hospitals, health care systems, and the patients they serve.

— Daniel Cidon is chief technology officer of NextGate, a health care enterprise identification company.