The purpose of these principles is to provide the American College of Cardiology’s (ACC) position on Unique Patient Identifiers. The principles set forth on this page are intended to educate our members and advocate our position to policymakers.
Quality Reasons for a UPI
The ACC is a 36,000 member non-profit professional medical society. Its purpose is to foster optimal cardiovascular care and disease prevention through professional education, promotion of research, and leadership in the development of standards and formulation of health care policy.
The ACC is considered a leader in health care technology within the physician community. The College supports the national agenda to accelerate the integration of health information technology (IT) into physician practices. To meet the health IT needs of its members, the ACC established the Informatics Committee to coordinate its health IT activities and policies. The ACC participates in many activities in the health informatics domain, and is involved with efforts related to interoperability, standards harmonization and health IT adoption.
For the purposes of this document, the ACC defines a national Unique Patient Identifier (UPI) as it is defined in the 1997 National Committee on Vital and Health Statistics report “Analysis of Unique Patient Identifier Options ”:
The identity of an individual consists of a set of personal characters by which that individual can be recognized. Identification is the proof of one's identity. Identifier verifies the sameness of one's identity. Patient Identifier is the value assigned to an individual to facilitate positive identification of that individual for health care purposes. Unique Patient Identifier is the value permanently assigned to an individual for identification purposes and is unique across the entire national health care system. Unique Patient Identifier is not shared with any other individual.
The ACC believes that the benefits resulting from implementing a UPI, thus improving quality of care, are primarily realized by patients. The ACC advocates the following positions regarding a UPI:
1. The UPI shall require mandatory national participation
2. Congress shall encourage the Department of Health and Human Services (HHS) to investigate UPI development
3. The UPI shall only be used for accessing health information and not tied to a social security number
4. Breaches of patient confidentiality via UPI shall be strictly monitored
5. Current health care privacy statutes and regulations regarding breaches of patient confidentiality via UPI shall be enforced
Quality Reasons for a Unique Patient Identifier
The U.S. health care system is composed of numerous independent systems, each with their own method of identifying patients, which makes linking patients across systems very difficult. Currently, these health care systems rely mostly upon statistical matching for patient identification. The false positives this method creates, which occur when there is a link to the wrong patient records, are a serious quality issue and can lead to deleterious consequences. Statistical matching may also create false negatives, in which not all of a patient’s records are found. This results in a physician having incomplete information when treating a patient.
The Institute of Medicine defines quality as the “degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.” Central to the idea of quality of care are related issues of patient safety, care coordination and data portability. With an increasingly mobile population, the ability to coordinate care across clinicians and to ease the difficulty of sharing electronic patient data is critical to maintaining the high standards of care in the U.S.
ACC supports conclusions outlined in 2008 RAND study, “Identity Crisis: An Examination of the Costs and Benefits of a Unique Patient Identifier.” The report states the benefits of a UPI to quality of care and continued improvement of care for patients. Specifically, it concludes:
1. The broad adoption of a UPI should enhance the U.S. health care system
2. A hybrid system utilizing both statistical matching and a UPI will be necessary for the foreseeable future
3. Security and privacy could be strengthened with a UPI
4. The costs of a UPI are significant but probably much less than the value associated with error reduction, efficiency and interconnectivity of the health care system
Furthermore, the ACC believes the following reasons support the need for a UPI.
1. UPI is integral to maintaining an accurate list of a patient’s allergies and current medications. Studies have found potential reductions in adverse drug events rates from the use of health IT of between 50 percent and over 90 percent 4, 5, 6, 7
2. UPI allows for the improved exchange of health information, which contributes to reducing medical errors. Patient safety incidents among 41 million Medicare patient records resulted in 238,337 potentially preventable deaths during 2004 through 2006
3. Published data on duplicate or split records in medical-records databases indicate a high rate of these errors, averaging about 8 percent overall and trending higher for larger systems 9
4. Current patient identifiers, such as Social Security Numbers, are targets for identity theft. The World Privacy Forum estimates that each year 250,000 to 500,000 Americans are victims of medical identity theft10. This may account for 80 to 120 billion dollars of loss per year11. If implemented properly, a UPI could limit susceptibility to medical identity theft. If there is a breach of a patient’s record, then the matter could be rectified by reissuing a number.
5. Several European countries, including Belgium, Germany, Luxembourg and Portugal, have plans to establish UPI, and confirm that good progress on strategies and roadmaps has been made12. A review of the European experience with UPIs has revealed no significant breaches of the security of individual health information.2
A UPI would improve patient care by promoting quality improvement derived from patient data registries. Registries rely on the accuracy and completeness of their data to provide valid analyses to participants. A UPI provides the capability to link patient information across registries and patient encounters. This capability would not only increase the accuracy and completeness of registry data, it is critical for registry data analysis and would enable a wide array of new scientific analyses. Without a UPI, registries rely on statistical matching logic based on patient identifiers (i.e. name, date of birth, etc.), which suffer from the same errors in statistical matching as cited earlier.
In the Health Insurance Portability and Accountability Act of 1996, Congress mandated that the HHS Secretary adopt standards providing for a “unique health identifier for each individual, employer, health plan and health care provider for use in the health care system. However, in 1998, as a result of privacy groups’ efforts, Congress subsequently prohibited HHS from expending funds to develop a UPI without explicit congressional approval.
Not having a UPI negatively affects current patient care. Below are cardiologists’ accounts of situations where having a UPI to integrate patient information would have led to higher quality of care.
A Case in Waste and Frustration
Three weeks ago I saw an elderly woman for a second opinion. She came to my clinic seeking an opinion on whether she needed an implantable cardioverter-defibrillator. Although the patient and daughter provided pertinent medical history, my decision-making depended on outside records. Within the prior three months she had received an echocardiogram, a nuclear stress test and a cardiac catheterization. She had received care from two different cardiology groups at two different hospitals as well as her primary care provider. When she arrived in my clinic she had a new incision from a pacemaker and was wearing an external defibrillation vest.
In total, we needed the patient to sign five separate release of information forms. It took over a week to receive all of the necessary information before I could see the patient again and her daughter needed to take another day off from work. After reviewing all of her records I was able to reassure her that no further testing or therapies were necessary. The case was frustrating – and mostly nonproductive – for all involved: the patient, her daughter, my nurse and assistant and me. All the information I needed to make some very complex medical decisions was completed before I saw the patient but it was separated in five different locations under five different ID numbers. The information I needed was contained within ten pages of information; however, over ten hours of time was spent communicating with the various offices and hospitals and over one hundred pages of information was faxed in hopes that the right information was sent. Is there any question why it would have been easier for many cardiologists to simply repeat a battery of tests?
With a unique patient identifier, the patient’s permission and digital access to those records, I could have seen, evaluated and counseled that patient in one visit.
A Case in Name Changed
This account involves a 45-year-old female with acute presentation to a regional emergency room with ST-Elevation Myocardial Infarction (STEMI). She had ongoing chest pain, and accordingly was given morphine and transferred. We fielded her in our usual rapid fashion but she was basically non-communicative because of the effects of morphine. She continued to have significant ST elevation. She had had previous coronary artery bypass graft, and previous stents at our institution, but no medical records came up under her name. We had to wing it when I would have liked to know the previous location of grafts, size and types of stents used previously and known allergies. However, none of this information was available.
We subsequently learned that she had had a name change, a history of reasonable compliance and that she could have been a reasonable candidate for a drug eluting stent sandwich rather than a bare mental stent sandwich. Fortunately, we got all of the grafts, but it probably took more contrast than necessary. It might have been unsafe to give abciximab, a platelet aggregation inhibitor, if she previously had had a thrombocytopenic response but luckily she hadn’t. We dodge these bullets fairly often, I'm sure. They could say that this is an uncommon problem, but as we know, there are a lot of STEMIs per year in the nation.
A Case in Duplicate Names
I have three patients who all have the same name and are related. In this family, the grandfather, the father, and the son share the same name without suffixes. Each is a separate entry in our electronic medical record, which should be identifiable by date of birth. Both the grandfather and the father have had their mitral valves repaired, and the son is being followed for mitral regurgitation. When they see me they undergo subsequent testing, which is then scanned into their medical records. Without fail, despite the separate birthdates, there is confusion as to the correct patient and frequently the wrong test is scanned into the wrong patient. Moreover, when each patient is evaluated both the patient and I are always squeamish that we have the correct patient data. A UPI would help resolve this confusion and given more confidence to patients and providers in the accuracy of the data records.
ACC believes that the need for a national UPI is principally an issue of improving the quality of patient care. While there are many obstacles that must be addressed in order to implement a national UPI, the ACC believes that the benefits of increased patient safety, care coordination and data portability outweigh these obstacles.