The Voluntary Universal Healthcare Identifier System
It was a simple error in identification, but a female patient's routine electrocardiogram appointment ended in tragedy because of it.
She was healthy when she went with her husband of 55 years to a clinic, and as is usual practice, earlier test results were evaluated by the attending physician's assistant. Unfortunately, the irregular cardiac EKG report presented during her visit belonged to another patient. The report had the patient's name handwritten across the top, not printed by the EKG device, which would normally be done automatically. The clinician did not take time to investigate this irregularity since the readout showed that the patient was on the verge of a heart attack.
The patient was rushed to the hospital where two additional EKGs showed normal results but were ignored. She was treated with unnecessary and very potent medications that caused a stroke. She underwent surgery, lapsed into a coma and died five months later. All of this was the result of one error: a test result associated with the wrong patient.
These incidents occur far too frequently.
An average of 195,000 people in the United States died due to potentially preventable, in-hospital medical errors in each of the years 2000, 2001 and 2002, according to a 2004 study of 37 million patient records that was released by HealthGrades, an independent source of physician information and hospital quality outcomes. In fact, The Commonwealth Fund reported that more than one of five Americans stated that they or a family member had experienced a medical or prescription drug error.1 Patient identification errors are one of the most common root causes of medical errors.2 Improved patient identity management could have an important impact on reducing those errors.
Automation in the healthcare environment lags behind most other industries, but there is extensive use of technology and information management systems in virtually every hospital and physician's office across the country. From medical devices such as the patient's EKG machine to electronic medical records, virtually every one of these systems has at least a minimal capability, and need, to discretely identify patients. Imagine a large hospital organization with affiliated clinics and doctors' offices. There are literally hundreds of computer systems that must communicate to share patient information in order for it to be used for care and administrative services. Recall our patient and the domino effect that occurred because someone pulled the wrong EKG.
To illustrate the problem in another way, consider Harris County Hospital District in Houston, Texas. Based on the analysis of 12 years of data from their enterprise master person index consisting of 3.4 million patients, 249,213 patients have the same first and last name as at least one other person. Some 76,354 patients share both names with four other individuals\; 69,807 pairs share both names and birth date. There are 2,488 patients named Maria Garcia, and 231 Maria Garcias have the same birth date.3 It is clear that the risk of patient identity confusion due to similar and overlapping demographic information is neither insignificant nor isolated.
David C. Kibbe, M.D., M.B.A., senior adviser, American Academy of Family Physicians\; chairman of ASTM International Committee E31 on Healthcare Informatics\; and principal, the Kibbe Group LLC, understands the challenge. "Separating identity from specific healthcare IT systems and applications is a critical need for modern high quality healthcare. As long as we rely on isolated software and ‘data islands' at each healthcare provider organization to specify a person's identity, we are literally designing errors and poor quality of care into the system. Mistakes and mismatches are inevitable. Use of a unique healthcare identifier is one way to achieve accuracy and interoperability for personal identity, and if used properly, it will save lives," Kibbe says.
Incorrectly identifying a patient has many unwanted consequences, including adverse drug events, unnecessary and duplicate testing, clinical complications, prolonged hospital stays, hospital readmissions, increased administrative costs and patient dissatisfaction. Important technologies have been developed to help mitigate this problem: enterprise master person indices are designed to manage the many patient-identifying applications that may exist in a healthcare delivery organization. However, even the best of these EMPI systems have been shown to have an 8-10 percent error rate in identifying a patient from their demographic information.
In response to this situation, ASTM Committee E31 developed two patient identification standards that were approved in 2007:
Based on these standards, the voluntary universal healthcare identifier (VUHID) system was designed and developed. The underlying principle of the VUHID system has been to provide a solution that is simple, inexpensive and effective in both improving the accuracy of patient identification and providing additional privacy protection for healthcare information.
Simply stated, VUHID is a web-based identifier generation system that interacts with EMPIs to provide unique identifiers to participating organizations. For patients who present their VUHID identifier, whether printed on a card or represented by a bar code or magnetic stripe, unambiguous identification is possible since VUHID does not create duplicate identifiers. Additionally, patients may elect to have multiple identifiers that can be associated with healthcare providers or specific encounters (e.g., psychiatric services) to which they wish to restrict access.
Hospitals and other provider organizations that use the VUHID system are linked together virtually by VUHID services. If a patient presents a VUHID identifier to any participating organization, even if he or she has never been treated there before, the VUHID system notifies the querying EMPI system that the identifier is valid and that the patient has records at one or more other sites. Contact information for the patient's other healthcare providers is provided, but no other information is provided because the VUHID system does not store anything except the identifier, the date and time it was created, the initial location to which the identifier was issued and other locations where that identifier has been recognized. No confidential or protected health information is ever made known to the VUHID system\; no database of protected health information exists.
In its 2008 report, Identity Crisis: An Examination of the Costs and Benefits of a Unique Patient Identifier for the U.S. Health Care System, the RAND Corp. concluded that "a unique, nondisclosing patient identifier is clearly desirable for reducing errors, simplifying interoperability, promoting NHIN [Nationwide Health Information Network] architectural flexibility and protecting patient privacy."4
An important aspect of the VUHID system is the concept that it is voluntary\; only people who want VUHID identifiers receive them. In 1998, the U.S. Congress put a moratorium on federal funds being spent on a national patient identifier solution, based in part on the concerns of many privacy advocates that the government would misuse a large database of patient information and that breaches of confidentiality might occur. That moratorium is still in effect. Recent efforts - some based on the VUHID approach, which eliminates any requirement for such a database - are attempting to remove this restriction.
Concerns about costs also exist, particularly the early projection of billions of dollars if the government would run a patient identification program. VUHID, soon to become part of a pilot project with a large health information exchange organization, has cost less than $250,000. Estimated operational costs, even for use by millions of patients, is dramatically less than early estimates.
As the nation invests heavily in health information technology - some $40 billion (including direct investment as well as projected savings) was included in the American Recovery and Reinvestment Act of 2009 - the need for a national, unique identifier is becoming more pressing. Projects to define interoperability standards so that healthcare organizations can share patient information as part of a Nationwide Health Information Network as well as health information exchange initiatives, are bringing to light the essential role of unique identifiers.
The VUHID system, under the auspices of Global Patient Identifiers Inc., a nonprofit organization established in 2008, is currently in beta test at a large health information exchange network as part of a grant from the Robert Wood Johnson Foundation's Pioneer Portfolio.** Other healthcare stakeholders, including insurance companies, are watching this project with interest. At the same time, the federal government is expressing interest in a patient identity management solution as it awaits relief from the moratorium.
Until the nation moves ahead to resolve this challenge, it is important for consumers to be informed and alert. Learn more about VUHID.
1. Davis, Karen, et al., Room for Improvement: Patients Report on the Quality of Their Health Care, April 2002. Based on The Commonwealth Fund 2001 Health Care Quality Survey, which consisted of interviews with 6,722 adults conducted by Princeton Survey Research Associates.
2. Agency for Healthcare Research and Quality/Patient Safety Initiative: Building Foundations, Reducing Risk, Interim Report to the Senate Committee on Appropriations, 2003.
3. George, Cindy, "Harris County Hospital District Tries New Kind of Palm Reader," Houston Chronicle, April 5, 2011.
4. Hillestad, Richard, et al., Identity Crisis: An Examination of the Costs and Benefits of a Unique Patient Identifier for the U.S. Health Care System, RAND Corp., October 2008.
**This sentence previously misidentified the name of the foundation.
Elizabeth West is chief marketing officer for Global Patient Identifiers Inc., Tucson, Ariz., which sponsors the Voluntary Universal Healthcare Identifier Project\; she is a certified professional in healthcare information and management systems.