The AMDIS Connection | IT: New Stethoscope for Better Care
The sad reality is that after you’re through designing, implementing and refining an EMR, you’re just beginning to get to the point of data integration. Yet the idea of being able to integrate data from different cases and learn something is one of the key premises of meaningful use Stage 1 and Stage 2.
Most major commercial data warehouse vendors have a product suite that isn’t the same as their standard EHR, but includes a clinical data repository. Usually, the oldest parts of that are built for billing and business analysis and support, but the new generation of tools are focusing on clinical info, and that’s where we’re getting into this whole idea of requirements for coded diagnoses that need to be rich enough and varied enough so that they are understandable clinically.
There are companies whose business model is about acting as a Rosetta Stone that cross-links the work of CPT, ICD-9 and -10, and other coding schemas with clinically relevant terminology. There are taxonomies that are relevant to a particular type of medical practice—for example, SNOMED comes out of pathology. There’s also ICPC, which diagnoses categories that are relevant to primary care practices.
Each taxonomy provides a different view of the world, and to succeed with Stage 2 and beyond, your computer system must generate compliance with different aspects of the qualification criteria. You need to have data integration, and your system needs to be able to look at and understand a range of coded information.
As the criteria were discussed for meaningful use, it was by design that clinical information is required to be in machine-readable form, with data elements besides one set—not just diagnoses, not just medications or lab results, but cross-cutting with demographic data and other information to gauge progress. Doing that will allow any organization to answer questions like what have we done? How many patients with a particular condition are we seeing? What are our medical outcomes? This new level of data integration can only be done with tools that are built to do that job.
Data requirements going forward with meaningful use are intended to get hospitals, health systems and practices competing on the basis of outcomes of care results from their patients. Reporting also demands data integration and information synthesis. You need to cross-walk between what may be the usual diagnostic category groups, the usual therapeutic interventions, usual results of testing, and then you need to say something about the population that you’re looking at. Your decision support tools and rules are critical in answering these questions.
Even in today’s data integration/reporting infancy, there’s a growing body of evidence that this era of automation in healthcare is not the haves versus have-nots, and there’s no way to compete with the results of the big academic centers. If you’re talking about demonstrating improvement in safety and quality of care, that’s not true. For example, the first Medicaid health IT incentive payments went to practices in Tennessee and Oklahoma, not in larger, more urban states.
That’s the way it’s supposed to be. You don’t have to be a huge organization to bring together information and mine the data within it for the benefit of the betterment of a whole system of care process.
Using an information environment to improve the practice of medicine is like a new stethoscope—it’s the new instrument in the evolution of American healthcare.
Most major commercial data warehouse vendors have a product suite that isn’t the same as their standard EHR, but includes a clinical data repository. Usually, the oldest parts of that are built for billing and business analysis and support, but the new generation of tools are focusing on clinical info, and that’s where we’re getting into this whole idea of requirements for coded diagnoses that need to be rich enough and varied enough so that they are understandable clinically.
There are companies whose business model is about acting as a Rosetta Stone that cross-links the work of CPT, ICD-9 and -10, and other coding schemas with clinically relevant terminology. There are taxonomies that are relevant to a particular type of medical practice—for example, SNOMED comes out of pathology. There’s also ICPC, which diagnoses categories that are relevant to primary care practices.
Each taxonomy provides a different view of the world, and to succeed with Stage 2 and beyond, your computer system must generate compliance with different aspects of the qualification criteria. You need to have data integration, and your system needs to be able to look at and understand a range of coded information.
As the criteria were discussed for meaningful use, it was by design that clinical information is required to be in machine-readable form, with data elements besides one set—not just diagnoses, not just medications or lab results, but cross-cutting with demographic data and other information to gauge progress. Doing that will allow any organization to answer questions like what have we done? How many patients with a particular condition are we seeing? What are our medical outcomes? This new level of data integration can only be done with tools that are built to do that job.
Data requirements going forward with meaningful use are intended to get hospitals, health systems and practices competing on the basis of outcomes of care results from their patients. Reporting also demands data integration and information synthesis. You need to cross-walk between what may be the usual diagnostic category groups, the usual therapeutic interventions, usual results of testing, and then you need to say something about the population that you’re looking at. Your decision support tools and rules are critical in answering these questions.
Even in today’s data integration/reporting infancy, there’s a growing body of evidence that this era of automation in healthcare is not the haves versus have-nots, and there’s no way to compete with the results of the big academic centers. If you’re talking about demonstrating improvement in safety and quality of care, that’s not true. For example, the first Medicaid health IT incentive payments went to practices in Tennessee and Oklahoma, not in larger, more urban states.
That’s the way it’s supposed to be. You don’t have to be a huge organization to bring together information and mine the data within it for the benefit of the betterment of a whole system of care process.
Using an information environment to improve the practice of medicine is like a new stethoscope—it’s the new instrument in the evolution of American healthcare.