There are simple things in our health care system that are failing us, like the lack of a unique health identifier.
It may sound technical, or even wonky, but missing a single way to identify a patient across our healthcare system is a burden for patients and our economy alike.
When we receive care from any part of the health system a record is created. These records are filled with clinical information: like what happened during the visit, our medical history, discharge instructions, lab reports, and imagery.
These records used to be in paper, with the information locked in manila file folders. In 2009, a $40B government-funded effort spurred the shift to electronic health records (EHRs). A driver for those funds was a 2005 research report by the think tank RAND that estimated “that if 90 percent of doctors and hospitals successfully adopt health information technology and use it effectively, resulting efficiencies would save $77 billion annually.”
It should have been a no-brainer. By digitizing the data, you’d be able to capitalize on the efficiencies of better coordinating care and eliminating waste. Unfortunately, for all the benefits EHRs had to offer, the investment was crippled by the same people who wrote the check. Congress funded the dollars for EHRs, but separately passed a federal ban on unique patient identifiers.
Databases are great at storing data and they excel at combining and retrieving data. They do it by using unique identifiers. Without the identifier, data cannot be joined accurately or confidently.
For example, it is much harder to retrieve the data based on something like a name. If you were a “James Smith” or a “Maria Garcia” there would be no way for find your records. There are over 38,000 James Smiths and 32,000 Maria Garcias in the United States. Let’s say you do what the healthcare industry does, which is to use a combination of name and birthday. That fails too. In Texas there is a patient database for the city of Houston. It has over 3.5 million patients in it. There are 70,000 people who share the same first name, last name, and birthday with more than one person.
On the other hand, it is very easy for a computer to retrieve all the data for an identifier that no one else shares. For example, Estonia’s health system uses a patient ID code to link every piece of health information together.
The implications are immediate:
In an emergency situation, a doctor can use a patient’s ID code to read time-critical information, such as blood type, allergies, recent treatments, on-going medication or pregnancy. The system also compiles data for national statistics, so the ministry can measure health trends, track epidemics, and make sure that its health resources are being spent wisely.
This is not an aspiration or dream. It is what happens in Estonia today and can happen in the United States.
700 people a day die in the United States because of hospital medical errors. This is the third leading cause of death. Another 10,000 people a day suffer serious complications because of these mistakes. Communication breakdowns, diagnostic errors, poor judgment, and inadequate skill are all to blame. How can it be fixed? A part of the solution is more reliable data. And we can’t have reliable data without a reliable way to tie all of our data together.
In the United States, policymakers prevented the use of a unique identifier in the name of protecting a patient’s right to privacy. And privacy is very important.
There are well-intentioned efforts underway to try and work around having a unique national identifier. Most start with the promise of using algorithms and different data elements to match and link patients. Sadly, these efforts all fail, because they are either not 100% accurate or not easily implementable. Furthermore, these approaches erode privacy, because they often require more personal and health information about a patient to be shared to guarantee a match.
When Estonian policymakers were asked if their unique identifiers weaken privacy, they remarked it is quite the opposite. By having their patient id code, they are better able to protect a patient’s private health information because any access of their health data is logged and because no personally identifiable information is shared when the data is shared.
The Path Forward
It is time for the United States to allow patients to have and use a unique national identifier. What needs to happen is:
- Lift the Funding Restriction: Congress needs to remove the restriction prohibiting the funding of unique IDs at the Federal level.
- Designate and Fund a Responsible Organization: Either HHS or a non-governmental organization needs to be responsible for issuing unique identifiers to the public.
- Always Protect Patient’s Right to Privacy: A majority of the public will be fine with a single unique identifier for all of their health records. Some will prefer to have multiple identifiers or to use no identifier at all. It is the patient’s decision to de-couple, opt-out, and even request the deletion of their records.
- Pick Proven Technology: When deciding how to generate the identifier, it’ll be important to keep things simple and pick proven technologies.
This is a self-inflicted wound with a straightforward fix.