HIMSS: Obama Advisors On Wrong Track
Inaccurate patient identification methods could sink the vision of the President’s Council of Advisors on Science and Technology for establishment of a "universal exchange language" to accelerate health information exchange, according to the Healthcare Information and Management Systems Society.
Inaccurate patient identification methods could sink the vision of the President's Council of Advisors on Science and Technology for establishment of a "universal exchange language" to accelerate health information exchange, according to the Healthcare Information and Management Systems Society.
In a comment letter, Chicago-based HIMSS expresses concern that the council's report issued in December "continually states that there is no role for universal patient identifiers, but does not offer a detailed approach as to how the healthcare community will achieve error-free patient identification across healthcare organizations."
The methods to ensure enough accuracy in patient identification to meet the goals of the council are currently not available, the organization contends. "HIMSS has concerns that the PCAST project will indeed exacerbate patient identification problems because it will involve searching larger patient populations and communications across environments that have no ability to coordinate their automation systems nor common techniques for patient identification unless such capabilities are made available," the organization asserts. "Therefore, if the effort outlined in the PCAST report is going to succeed, it must find ways to improve the accuracy of patient identification over current methods or risk failure just on the basis of inadequate patient identification. HIMSS encourages a study of available technologies and techniques in this regard."
In particular, HIMSS disputes as technically incorrect an assertion from PCAST that "unique patient identification is not necessary because multi-factor or algorithm solutions can be leveraged to identify a patient." The council appears to confuse the intended purpose of such solutions, according to HIMSS. "Current algorithms for medical record/identity resolution leverage matching across multiple fields (which may contain errors) and normally never achieve 100% accuracy, which means that false positives and false negatives are possible when joining multiple records for the same person."
When tackling patient matching issues, federal officials should study integrated delivery systems--notably Military Health, the Veterans Health Administration and Kaiser Permanente--that have complex electronic health records systems and have established identifiers for millions of patients, HIMSS advises. The 16-page comment letter is available here.
In a new blog entry, Beth Israel Deaconess CIO and HIT Standards Committee co-chair John Halamka, M.D., lays out general principles and 10 characteristics that a universal exchange language should have, at http://geekdoctor.blogspot.com.
--Joseph Goedert
In a comment letter, Chicago-based HIMSS expresses concern that the council's report issued in December "continually states that there is no role for universal patient identifiers, but does not offer a detailed approach as to how the healthcare community will achieve error-free patient identification across healthcare organizations."
The methods to ensure enough accuracy in patient identification to meet the goals of the council are currently not available, the organization contends. "HIMSS has concerns that the PCAST project will indeed exacerbate patient identification problems because it will involve searching larger patient populations and communications across environments that have no ability to coordinate their automation systems nor common techniques for patient identification unless such capabilities are made available," the organization asserts. "Therefore, if the effort outlined in the PCAST report is going to succeed, it must find ways to improve the accuracy of patient identification over current methods or risk failure just on the basis of inadequate patient identification. HIMSS encourages a study of available technologies and techniques in this regard."
In particular, HIMSS disputes as technically incorrect an assertion from PCAST that "unique patient identification is not necessary because multi-factor or algorithm solutions can be leveraged to identify a patient." The council appears to confuse the intended purpose of such solutions, according to HIMSS. "Current algorithms for medical record/identity resolution leverage matching across multiple fields (which may contain errors) and normally never achieve 100% accuracy, which means that false positives and false negatives are possible when joining multiple records for the same person."
When tackling patient matching issues, federal officials should study integrated delivery systems--notably Military Health, the Veterans Health Administration and Kaiser Permanente--that have complex electronic health records systems and have established identifiers for millions of patients, HIMSS advises. The 16-page comment letter is available here.
In a new blog entry, Beth Israel Deaconess CIO and HIT Standards Committee co-chair John Halamka, M.D., lays out general principles and 10 characteristics that a universal exchange language should have, at http://geekdoctor.blogspot.com.
--Joseph Goedert
More for you
Loading data for hdm_tax_topic #better-outcomes...