How other countries’ struggles with patient identifiers can help the U.S.
Discussion of a national patient identifier (or lack thereof) has run rampant in healthcare over the last two decades. The conversation continued recently with an op-ed article that appeared in The New England Journal of Medicine calling on Congress to lift its ban on using federal funds to establish a patient identifier in the U.S.
Encouraging greater efforts at the federal level, the authors, who include John Halamka from Beth Israel Deaconess Medical Center, point to the successful implementation of unique patient identifiers in other countries that have benefited from this core piece of infrastructure. Although both England and Scotland have a mandated number, which allows data sources to be connected, it alone is not enough to achieve total integration across health and social care services.
For example, older health system applications still cannot connect in real time to query for the National Health Service (NHS) number in England. Furthermore, systems continue to have multiple NHS numbers applied to one patient, demographic inaccuracies, and incorrect NHS numbers applied to individuals.
In fact, even with the NHS number in place, healthcare leaders have discovered that an external enterprise master patient index (EMPI) solution is needed to help manage these identifiers. The EMPI is used by multiple NHS trusts and enables social care data, which often does not include a NHS number, to be married together with healthcare data. It also resolves cases in which patients are issued more than one NHS number.
In Scotland, the Community Health Index number will be leveraging an EMPI to replace the mainframe-based Community Health Identifier. Here, the EMPI provides quality control over the core patient information by providing duplicate detection services, enforcing data capture workflows and enabling all systems to easily connect to a demographic lookup service using the latest messaging standards. In addition, the EMPI will maintain relationship data so that the association between a patient and a general practitioner in Scotland can be centrally managed.
Ultimately, the EMPI acts as an additional layer that helps to deal with cases where the national identifier cannot be relied on. While this is a relatively small fraction of patients, it is still large enough that it cannot be ignored. Additionally, the EMPI exposes very sophisticated patient search capabilities for all connected systems to ensure that when a NHS or CHI number is not presented, there is a high likelihood the patient will be found through other attributes, thus, preventing the fragmentation of data.
Since the U.S. healthcare system is much larger in scope and far more complex than that of the U.K. and Scotland, we can expect the difficulties in implementing a unique patient identifier to be proportionally greater. The premature end of CHIME’s National Patient ID Challenge last year should be a sobering industry reminder that even private efforts to develop a universal solution have faltered.
That is not to say that pursuing a national patient identifier isn’t a laudable goal. But we shouldn’t assume that a single identifier will be some kind of final solution. Rather, we should treat a unique identifier as just another strong indicator of identity that, in conjunction with other demographics as well as biometric data, can assist with matching. A national patient identifier would be an important piece of the overall identity puzzle, but it would not be definitive.
Ultimately, the real challenge of identity management and piecing together a longitudinal health record has to do with integration. More specifically, it revolves around managing the flow of demographics and associated identifiers dispersed across multiple systems.
Because these systems often have little reason to communicate with one another, and because they store their data through fragmented architecture, an excessive proliferation of identifiers occurs. The result is unreliable demographic information, triggering further harm in data synchronization and integrity.
As healthcare digitization proliferates to a more mobile and connected reality, it is equally critical to consider use of other identification mechanisms to ensure that patients are accurately identified at the point of care, and that their demographic information is accurate and up-to-date. Biometric identification, for example, in conjunction with the use of smartphone technology, can streamline the process of patient identification substantially. Convenient biometric solutions such as facial recognition affords individuals management of their own data via their smart phone and can ensure that the process of hospital registration is as simple, accurate, and reliable as scanning a boarding pass for a flight.
Instead of looking at a national patient identifier as the definitive answer for solving patient identification issues, the nation should instead focus on a pragmatic, multi-faceted approach to improving patient matching—one that moves beyond a reliance on probabilistic algorithms. Existing technologies can be supplemented with others that rely on demographic data with advances being made in biometrics as well as the ubiquitous use of smartphones. This combination of technology enables the cameras of commodity smartphones to capture accurate and secure non-reversible biometric signatures without the complexity of expensive and specialized equipment, like palm vein scanners.
As we begin to embrace these and other technological innovations into the patient-matching equation, we will increase our ability to accurately identify a patient at any time, in any care setting, and set the stage for the collaborative and informed decision-making needed for value-based care and population health management success.