AS THE ARTICLES in this issue demonstrate, the electronic health record (EHR) has reached a sufficient level of adoption and sophisticated use in ambulatory care that our focus needs to shift beyond initial implementation, including the transition from paper records to electronic records, to a dependence on EHR during a lifetime of practice in both the inpatient and ambulatory settings. Now that the computer-based patient record has become the essential technology for health care that the Institute of Medicine declared it to be in 1990 (Dick, 1990/1997), we need to shift our thinking toward supporting connected practice, secondary uses of data, portability of data, and the time value of information to avoid information overload. It is inevitable that organizations may need or desire to change their EHR vendor or move from an in-house to a vendor-based system, that patients will change practices or move to another part of the country, and that physicians will change their employer and location over time. Our systems must consider future migration needs from the time they are installed.
The Decade of Health Information Technology began in 2004 with the creation of the Office of the National Coordinator for Health IT and the publication by Tommy G. Thompson, Secretary of Health & Human Services, and David J. Brailer, the first National Coordinator for Health IT, of a Framework for Strategic Action called The Decade of Health Information Technology (Thompson & Brailer, 2004) that set a goal of providing electronic medical records for most Americans by 2014. The goal was achieved through the ARRA HITECH Act that created the Meaningful Use EHR Incentive Program and not only drove adoption but also created disappointment regarding ease of use and achieved value of the technology.
We are now in the second Decade of Health Information Technology that began in 2015 with the publication of Connecting Health and Care for the Nation: A Shared Nationwide Interoperability Roadmap (Office of the National Coordinator for Health IT, 2015), with a goal to achieve nationwide interoperability to enable a learning health system by 2024. The gap that now exists between having an EHR and using an EHR in an optimal fashion depends on portability of data and tools. Critical to successful interoperability is the retention of codes and metadata within the EHR when data are received from other systems. Data that are received from another practice, hospital, or ancillary service should be retained so that the data that are reexported match the data that were imported subject to data reconciliation and error correction. Without retention and preservation of these critical data, future migration of patient records to other systems is jeopardized and an appearance of errors in the records is created.
Current interoperability standards focus on the external representation of data in messages and documents rather than constrain how that information will be stored in the EHR. Without bringing these standards into the EHR, we will never be assured that a patient summary or report that we receive can always be imported as discrete data that can be used for computation and decision support. Coding and terminology systems will continue to evolve, and our EHR systems must function in this changing world. In some cases, this may require dual coding of data for support of different uses such as ICD-10-CM and SNOMED CT for problem lists, National Drug Code (NDC) and RxNorm for medication lists, or NDC and CDC CVX codes for immunizations (Office of the National Coordinator for Health IT, 2016). While one can map from one code system to another to create external interoperable messages or documents, it is difficult to import data interoperable into the EHR if the internal structure and terminology do not match the external representation.
The HL7 Fast Health Interoperability Resource (FHIR, n.d.; Mandel et al., 2016) offers an exciting new approach to both interoperability and extending functionality in an EHR through use of an Application Programing Interface that can support third party applications to enhance an EHR by adding new views, clinical decision support (CDS), and quality assessments (QA) tools. This approach will fail to be universally available if we do not focus on compatible coding and structure of data inside the EHR in addition to external representation of data in a patient summary document.
The content of medical records (paper or electronic) has always been tightly linked to the method of payment for health care. With capitated systems or in national health services such as the British NHS, documentation is sometimes minimal and focused on key medical decisions and data. In our current system based on E&M (Evaluation and Management) coding, documentation of normal data will be detailed. In the new value-based payment systems including Affordable Care Organizations and MACRA, documentation will focus on assuring quality and supporting cost-effective decisions. Documentation for clinical trials requires uniform structured and coded data often forcing duplicate data entry and parallel systems. As our payment policies evolve, as in the case of Accountable Care Organizations, we need to consider their needs for structured and coded data in the EHR that are readily interoperable and can be merged across practices. Sharable tools for CDS and QA will enhance our ability move toward value-based care, with limited implementation overhead providing that fully supported interoperability makes these tools widely available to many EHRs.
We have reached an exciting point in the evolution of the ambulatory EHR-adoption is high and best practices are showing the way to truly meaningful use of this technology. Realizing the full potential of this technology in ambulatory care will require portable data across vendor systems and institutions because of patients' will to change providers and receive care in multiple locations.
REFERENCES