From 3M Health Information Systems
Reflections from a laboratory scientist: Terminology’s role in U.S. healthcare paradigm shift
The American Association of Clinical Chemistry (AACC) held its 69th Annual Scientific Meeting and Clinical Lab Expo in San Diego in August, in collaboration with the Canadian Society of Clinical Chemists and American Society of Clinical Lab Scientists. With their vision of “better health and health care through laboratory medicine,” the attendance of 21,300 scientists was a record. In addition to meeting clients and colleagues, I enjoyed roaming the exhibit hall of medical laboratory testing equipment, reagents, and vendors who supply our clients with their testing materials and processes. It assisted greatly in our LOINC mapping work to identify what is truly being measured.
The portions of LOINC and SNOMED CT created for laboratory medicine have a bidirectional sequence: The clinician orders the tests or profiles and the laboratory returns results. Interpretive data points may have been canned text with limited ability to personalize comments. Laboratories were only allowed to reflex on testing with incurred charges when the clinicians had ordered the reflexive testing cascade. If clinicians didn’t opt for the reflexive testing, the pace to reach diagnosis might be limited by the clinician’s knowledge or willingness to consult. There is now a paradigm shift in health care towards precision medicine, integrated with known patient genomic data, past diagnostic information, imaging, medication list and other information.
The changes in laboratory medicine reimbursement is escalating through Centers for Medicare/Medicaid Services (CMS). Thus, the regulatory affairs session at the annual meeting was a packed room. Laboratories recognize their prior payment model is dissolving before their eyes, and are working towards a new sustainable payment model. My 3M colleague Mary Zeigle covered in her recent blog. Individual and group physicians are being challenged through CMS reimbursement to work toward better outcomes for their patients. What better way than by growing their knowledge team to include laboratory testing, data mining and pathology consultative services?
Order entry errors and diagnostic errors hurt patients. Physician studies with Johns Hopkins in 2013 showed that diagnostic errors are two times more likely fatal than other errors, and that 30 percent of diagnostic errors led to death or permanent disability. The AACC regulatory affairs panel described laboratorian impacts, including developing expertise to develop decision support tools, guidelines and algorithms. Bayesian Network and Markov Decision Support provided examples of advanced computational machine learning methods that could interpret the type of clinician, the patient and patient conditions, leading to intelligent identification of the next most likely tests. Information could be provided on conditions that include or exclude the ordering of a particular test, as well as alternative logic. The interface would support clinicians’ logical analysis, not dictate the path. Having an information system armed with the advanced computational algorithms could interface with the health care team and reduce such errors.
Does an information system exist that can carry this weight? Is this an LIS2.0, an EHR 2.0 or middleware the likes of which have never been seen before? Perhaps the next generation will not even follow past naming conventions. Whatever the system’s name, terminology’s part in health care will jump from being the standardized record keeper of what has been done, (encoding procedures, assays, treatments, diagnosis) to the thinking engine of what should be the next step, holding a myriad of terminology coded data elements from the patient’s integrated records. A system running logically through all the possibilities, providing support and clarity to the patient’s health care team.
Pam Banning, healthcare data analyst with 3M Health Information Systems.