Have you ever been mistaken for someone else or received another person's order at a restaurant? These are fairly everyday occurrences. My wife and I took a trip for our wedding anniversary to an island resort in the Caribbean and an interesting thing occurred when we were buying souvenirs in the gift shop. When they asked for my name, I learned that I would have to present my passport because there was another Charles Christian at the resort. Interestingly enough, the request for my passport was not for verification of my identity at that point, but so they could capture the number on my passport and associate it with my purchases.
We in the health information technology field have long sought that holy grail of an application/system that will magically prevent duplicate patient records and provide for positive patient identification. We've installed numerous master patient index applications and even enterprise master patient index applications. However, it is to no avail; we still are creating duplicate patient records every day.
When we were concerned with only managing billing information, the patient's name wasn't that much of an issue because we had the insurance group number and the policy number. In those days, the patient's medical records could be easily merged by moving the paper records from one folder to another.
Now that we have electronic health records, the automation has made it much easier to create duplicates and harder to merge the records; at least while the patient has an active encounter/admission, in some cases.
I've long been aware that the lack of positive patient identification is really a patient-safety issue, more than an operational one. Most of my conversations about physician concerns with EHRs these days (in the inpatient setting) are not related to physicians using the EHR system, but about them being expected to successfully manage their patients with incomplete information. I believe that physicians see the value of having ready access to the patient's clinical information; however, they also are more than ready to point out that all of the patient's records need to be available and accurate.
Most of the healthcare chief information officers I've spoken with agree that the lack of a reliable and standard method to accurately identity the patient accessing their healthcare system is a very expensive problem. Many have spent large sums of money cleaning up their internal master patient index files, only to have to repeat the process some years later.
I've read several papers on the work some organizations have completed in the area of using patient demographic information to establish positive patient matching. Many have shown a 95% to 98% accuracy rate. This might sound impressive; however, if we were talking about our banking records, would we be satisfied with this level of accuracy? Probably not.
As we move further down the road in creating an electronic fabric of clinical information, the desire is to be able to share that information between the various levels of care. This sharing of information will have a very positive impact on helping to lower the total cost of healthcare in many ways. However, unless we can know with certainty that we are sharing the appropriate information for the correct patient, we very well could be creating opportunities for errors at an increasing rate.
We as an industry, in partnership with our friends at the Office of the National Coordinator for Health Information Technology and the CMS, need to move this to the top of the list of actions that will greatly remove some of the friction in health information exchange and create a safer place to care for our patients.
Charles Christian is vice president and chief information officer at St. Francis Hospital in Columbus, Ga. He is also the 2015 board chairman of the College of Healthcare Information Management Executives.
