Iris recognition, palm-vein, fingerprinting: Which biometric is best for healthcare?
When University Health Care System was deciding which biometric technology to deploy in conjunction with its Epic EHR, the Augusta, Georgia, hospital opted for iris recognition.
“It’s more accurate than fingerprint scanning and vein mapping, and a plus is that it’s a clean process because the patient never has to touch anything, which is a big issue in healthcare because of infection control,” said University Health Care revenue cycle director George Ann Phillips (pictured). “The camera sits on a tripod and it is voice-activated and tells a patient exactly how to position their head right before it snaps a photo.”
Iris recognition is just one in an emerging field of biometric technologies — others include palm-vein, fingerprinting and facial recognition — for identifying and authenticating patients.
While the different forms of biometrics each have pros and cons from various perspectives, some healthcare executives have reached a conclusion applicable to them all: If patients don’t like a particular technology, they will refuse to use it — and then what’s a health system to do?
Iris recognition earns accolades
Iris recognition is the most accurate biometric technology available in the market today, according to the study “5-year Opportunity Analysis of the Global Iris Recognition Market,” from research and consulting firm Frost and Sullivan. The technology scans more than 250 signature points to collect data from a source while competing systems scan fewer than 100, and iris recognition technology will make headway in the healthcare market over the long term as technology costs decline, the study predicted.
The accuracy of iris recognition, the unchanging state of the iris throughout a person’s lifetime, and the technology’s contactless and non-intrusive nature make a strong case for iris recognition in healthcare, the Frost and Sullivan study added.
But while Frost and Sullivan has carefully analyzed different biometrics technologies to come up with its conclusions, EHR maker McKesson, for example, researched the various biometrics technologies when it was looking for a company to partner with for a patient authentication system.
Palm vein biometrics
At the end of its quest McKesson ultimately decided on the PatientSecure system from Imprivata that scans a patient’s palm-vein, which is underneath the skin in the hand.
“We did the research, and we believed the palm-vein was right for McKesson for ensuring the security of the patient while being something that patients would not walk away from, like fingerprint,” said Debbie Fox, vice president and solution line manager at McKesson Enterprise Information Solutions. McKesson embeds PatientSecure into its patient registration systems.
Overall, Imprivata reports that PatientSecure is used in 70 health systems encompassing around 400 hospitals and thousands of clinics.
“The palm-vein pattern is unique to every person, and it does not change throughout your life,” Fox added. “It is a very simple solution to use, it is not invasive, and it is very accurate. And it is not used by law enforcement, so it does not have that stigma associated with it, like a fingerprint does.”
The law enforcement stigma is not to be underestimated — the average American very commonly associates fingerprints with criminality and, more significantly, with Big Brother. The stigma can cause some patients to decline to use the technology.
Patients decline for other reasons, too
There are other reasons why patients decline to use some biometrics technologies. Some of them not as potentially adverse as Big Brother, but no less troubling to a patient.
“We built prototypes with different biometric modalities and rolled them into the waiting rooms of a partner health system and waited for patient feedback,” said David Wiener, general manager of the PatientSecure products group at Imprivata. “When it came to modalities that involve pictures, well, in a healthcare setting, you are not necessarily in the best condition, and a lot of patients did not want their pictures taken, so the participation rate dropped.”
Iris recognition, for example, while non-invasive, requires a photograph be taken of a patient to identity the iris. (Iris recognition should not be confused with retinal scanning, an invasive technology.)
“External modalities like fingerprint and iris turned patients off, while with palm-vein, patients were more apt to just put their hand down for the simple scan,” Wiener said. “The decisive point for us was identifying unconscious patients, that clinched it in the healthcare setting. The clinicians we worked with had difficulties prying open the eyes of an unconscious patient in the ER, and dealing with a non-focusing eye. Palm-vein worked the best.”
Fox added that when McKesson began selling PatientSecure, the technology’s ability to easily authenticate an unconscious patient became a big selling point. And Wiener contended that the accuracy rate of palm-vein technology is as good as that of iris recognition, though he added that the finer distinctions in biometrics accuracy are moot if patients refuse to use a technology.
University Health Care System started its patient authentication program with the RightPatient iris technology in June 2015, implementing it in one area at that time so staff and patients could get used to the new technology. The health system registered 1,000 patients in the first month and today is up to 42,000 patients.
“The accuracy rate is stated at 99.99 percent,” Phillips added. “Thus far we have found that to be true.”
When it comes to biometrics technologies for patient authentication, in the end, it’s a matter of what works best for individual healthcare organizations.
“You have to think about use cases. Biometrics are really good when they are convenient. For instance, we’re seeing fingerprint biometrics really picking up on smartphones, and that’s because there it is convenient,” said Adrian Sanabria, senior security analyst at IT consulting firm 451 Research. “But any biometrics technology I might recommend today would likely change as technology gets better at doing different things.”
Which means that hospitals need to do more than pick one and stick with it forever. Rather, executives need to track the evolutionary changes underway in the various biometric technologies to ensure they’re using the best one at any given time.
Indeed, University Health Care System’s Phillips predicted that while iris recognition meets their needs now, facial recognition technology could come into its own.
“It seems like the wave of the future,” Phillips said. “Eventually I can see facial recognition, where people can simply take a selfie and register through their mobile phones.”