GAO highlights need to better match patient records
- By Ben Moscovitch
- Jan 17, 2019
A federal report released Jan. 15 underscores the need to improve how patients’ electronic health records are linked among the various doctors’ offices and hospitals where they receive care. Better matching of these records would enhance the quality of that care and help reduce costs.
The Government Accountability Office (GAO) report makes clear that problems with patient matching persist and examines strategies to address them. The 21st Century Cures Act, a wide-ranging health bill enacted in late 2016, required the agency’s review of current practices and ways the federal government could improve match rates.
In its health information technology work, The Pew Charitable Trusts has identified two key opportunities to improve matching: standardizing the way that demographic data, such as addresses, are used by electronic health record (EHR) systems; and examining how biometrics, such as fingerprint or iris scans, can help link patient records. The GAO report highlights both.
Over the past decade, most hospitals and doctors’ offices have transitioned from paper records to EHRs, which boost efficiency and give patients and medical professionals more complete information on which to base decisions. But a key element of successfully exchanging this information is making sure that the available records correctly refer to a specific individual. Current patient match rates between organizations, however, can be as low as 50 percent. That results in patients and clinicians not having the information they need, which introduces safety risks and increases costs.
There can be many reasons for incorrect matches, from typos or missing information to differences in how EHR systems format data. Whatever the reason, mismatches can be dangerous to patients, for example, when doctors don’t have relevant test results or medication allergy information, or when mismatches contribute to patients receiving care that isn’t right for them. Incorrect matches also can be expensive to remedy.
Solving this problem requires progress on multiple fronts and agreement from a range of stakeholders, including EHR developers, medical professionals, hospitals, and government agencies.
Government can take a key near-term step by specifying standards for how demographic data are documented. Patients’ addresses, for example, can be recorded in different ways, such as by abbreviating the street suffix (e.g. listing “Blvd.” instead of “Boulevard).” Differences in how EHRs store that information can make it harder to link records.
Not yet published research led by Indiana University’s Regenstrief Institute suggests that standardizing certain demographic data—particularly addresses through use of the U.S. Postal Service format—could meaningfully improve match rates. Additionally, using elements such as email addresses, which are increasingly recorded in a patient’s profile, would add another data point to help ensure that matches are accurate.
Given the clear benefits of improved matching, the Office of the National Coordinator for Health Information Technology (ONC), the agency within the Department of Health and Human Services that oversees EHRs, should identify standards for using demographic data for matching, and additional data elements to use.
Increasing the use of biometrics represents a greater challenge, but one that could lead to even larger improvements in match rates. This is because biometrics are virtually unique to each individual when combined with demographic data.
In a series of focus groups Pew conducted with patients in 2017 and 2018, biometrics were the most popular option for improving match rates. Many participants noted they already use fingerprints or facial scans to unlock smartphones and move through security checkpoints at airports. Given the increasing prevalence of these technologies, people recognized the potential of biometrics to assist with record matches, especially when patients may be unconscious or unable to identify themselves.
To use biometrics for matching, developers would need to agree on a framework that allows the use of different technologies, while protecting patient privacy and the security of patient information. At the same time, providers may need to invest in technology and consider the impact on registration workflows. Pew is conducting research to help determine whether and how biometrics can help solve matching problems.
The GAO report highlights the importance of improving patient matching to ensure quality care. To help achieve that goal, ONC should take immediate action to standardize patient data, while the private sector can begin to lay the groundwork for the use of biometrics to better link electronic medical records. Taken together, these steps could help ensure that patients get the care they need when they need it.
Ben Moscovitch is project director, health information technology, with the Pew Charitable Trusts.