A Common Language Architecture for Health Care
In the United States and a few other countries, notably the United Kingdom, the past decade has witnessed the development of components necessary to a health information system infrastructure. Development efforts have included such critical infrastructure elements as standards for data transfer and common clinical terminology sets, which make possible electronic health records (EHRs), computerized physician order entry, various imaging systems, decision support systems, data warehouses, and other increasingly indispensable health information management tools.
The EHR, however, remains one of the most fundamental aspects of an electronic health information system. Standardized clinical terminology is the content backbone of the EHR, reducing the variability in the way data are captured, encoded, and used for the clinical care of patients and medical research.
In addition to supporting a range of outcomes research and improvements in clinical care and safety, a fully functioning EHR can assist in the detection of emerging public health threats and facilitate necessary research for population health studies. Health care professionals and officials in countries such as the United Kingdom, the Netherlands, Denmark, and Australia, as well as the United States, have made important headway toward this end.
In order to increase international recognition that to deliver quality patient care and population health outcomes, a whole health system approach is needed. The specifics of the approach may vary by country. Health systems struggle with managing diverse medical records that must be coordinated not only across a patients lifetime, but also among groups of patients and entire populations to ensure proper treatment, track disease, and provide the best possible health outcomes.
A whole health system approach of almost any stature will depend on the availability of accurate, integrated information that:
- Embodies clinical integrity of shared meaning within and across stakeholders in the public and private sectors;
- Incorporates resources, such as knowledge bases and decision support tools that link to the clinical detail; and
- Enhances systems that can exchange interoperable information electronically and confidentially at the local, national, and global levels.
In the growing digital health care environment, data needs to be captured once, then reused over and over again. Interpreting health information within electronic health information systems requires clinical data that is recorded at the appropriate level of detail, is consistent over time and across boundaries, can be transmitted without loss of meaning, and can be aggregated at general levels from multiple perspectives.
The design of SNOMED Clinical Terms (SNOMED CT), the health care terminology used for indexing, storing, retrieving, and aggregating patient data in a computer readable format, meets the requirements of good terminology practice. This includes unique, meaningful identifiers; concept-based design; and the use of description logic. Developed by SNOMED International, a division of the College of American Pathologists (CAP) in the United States, in collaboration with the United Kingdoms National Health Service, SNOMED CT offers multilingual, controlled health care terminology with comprehensive coverage of diseases, clinical findings, etiologies, therapies, procedures, and outcomes, as well as flexibility in expressing clinical concepts. It provides the core general terminology for an EHR, containing more than 357,000 health care concepts with unique meanings and formal logic-based definitions organized into hierarchies. Concepts are further linked to a total of 957,000 synonyms that may be used for a particular term. For example, arthritis in SNOMED CT is linked to synonyms that may be used to describe the same condition, such as joint inflammation and inflammatory arthritis. SNOMED users can refer to any of these terms with the assurance that they are linked to a common definition.
The digital repository of coded text and images within the EHR creates a searchable, virtual filing system. With greater than 1.37 million relationships between clinical concepts, the core terminology makes retrieval easier and more reliable than plodding through paper records.
Currently, SNOMED CT has been adopted in more than 30 countries worldwide. Such widespread adoption has earned SNOMED International a global reputation as a leading-edge provider of a sustainable, scientifically validated terminology and infrastructure. For a whole health approach to succeed both nationally and internationally, however, remaining barriers to the adoption of SNOMED CT must be removed.
Barriers to Adoption
Increasingly, coordination and continuity of care require that relevant information about a patient be integrated from several different clinicians, settings, and sites of care. The divergent health information technology employed within and across settings, however, presents an added hurdle that must be leaped before accurate and reliable electronic communication of medical information can occur.
For health data of any kind to be exchanged between systems, clinical terminology must be standardized. Such standardization enables system interoperability, that is, the ability for data to be exchanged between systems, regardless of the technology used. System interoperability; remains among the most critically necessary functions for helping to ensure that health information is available and easily accessible where and when it is needed. Without standardization, custom interfaces and other workarounds become necessary, which make processes difficult at best, error prone at worst.
The health care industry has significantly lagged behind other major industries, such as banking and aviation, in capitalizing on the best of what information technology and the evolving field of informatics offer. Historically, the development of health information technology has occurred in a highly segmented fashion, with various electronic solutions being developed on a case-by-case basis, outside of a larger vision for an enterprisewide electronic information infrastructure that could communicate with other external electronic information systems. The lack of modernization in the health care information infrastructure has made achieving the vision of a patient-centered electronic health record accessible anytime, anywhere by authorized individuals a much more complex endeavor. Standardized clinical terminology is the cornerstone of the infrastructure necessary to achieve this vision.
Benefits of SNOMED
Using SNOMED CT improves efficiency, productivity, and transaction management. It allows for data to be collected once, then shared and reused for many different purposes. This kind of data sharing increases the accuracy of clinical documentation and communication. For instance, to trigger decision support rules, different types of concepts must be linked to each other, such as allergies to drugs, procedures to devices, and diseases to contraindications. Many of the common rules draw on data that may be recorded at different times and places along the continuum of care, by different clinicians, and at varying levels of specificity. For example, clinical alerts can be created based on disease and allergy coding in relation to specific medication codes. Decision support rules built on SNOMED CT enable the development of a more sensitive and sophisticated decision support system, thereby helping to further reduce the risk of medical errors and adverse outcomes.
From a cost management and reduction standpoint, a SNOMED CT-driven EHR allows for more efficient and efficacious care delivery between a team of primary care providers, specialists, nurses, pharmacists, and others involved in a patients care. An easily accessible digitized record helps avoid duplicate tests and procedures, not only saving time and health care expense, but also delays in treatment that could result in undesired outcomes. The system interoperability facilitated by SNOMED CT can also simplify data integration, thereby reducing associated costs and enhancing quality of care, patient safety, statistical reporting, outcomes measurement, evidencebased medicine, and cost analysis.
As population health issues and threats to public health have taken on more intense global importance, health information technology solutions are needed to collect, aggregate, analyze, and formulate meaningful and timely responses or strategies for prevention. SNOMED CT allows for real-time collection and aggregation of health data and information needed to detect health-related pattern in population health, such as outbreaks of insect-born viruses, or threats to public health, such as bioterrorism
Medicines knowledge base of diseases, treatments, and outcomes continues to grow at exponential rate, showing no signs of slowing down anytime soon. A SNOMED CT-driven EHR helps create and support much-needed research databases that contain more clinically reliable information. The richness of such databases allows health care researchers to draw upon de-identified patient data for a wide range of efforts, including outcomes studies, public health initiatives, and patient safety efforts. This capability will only enhance global efforts to meet the challenges of 21st century health care delivery in the face of rising life expectancies and the associated social, ecologic, and economic conditions that will affect the health of the world population.
While SNOMED CT is comprehensive on its own, it maps to other medical classifications, localized code sets, and terminologies already in use, such as ICD-9-CM. This avoids duplicate data capture, while facilitating enhanced health reporting, billing, and statistical analysis. SNOMED CT also provides a framework to manage language dialects, clinically relevant subsets, and extensions comprised of concepts and terms unique to particular organizations or localities. These unique capabilities help create an invaluable information resource for supporting health care priorities at local (both individual and communitywide) and national levels, whether government- supported or private.
As health care organizations and governments worldwide assess infrastructure requirements, the choice of a terminology standard will have a profound effect on system interoperability, comparability of data, workflow, and decision support. SNOMED CTs global reach continues to expand, increasing its content, reputation, and value not only to users, but also software suppliers. One terminology architecture worldwide for global software suppliers will result in an expanded market and increased benefit and value for SNOMED CT users.
SNOMED CT is recognized by a number of prominent standards development organizations, including American National Standards Institute (ANSI), HL7, Digital Imaging and Communications in Medicine (DICOM), Accredited Standards Committee (ASC) X12 version 4010, and others.
SNOMED CT is also continuously evolving and expanding its content to incorporate clinical terminology from an increasing number of specialties. For example, in anesthesia, SNOMED works collaboratively with the Royal College of Anesthetists, the Anesthesia Patient Safety Foundation (APSF), the National Health Service Information Authority in the United Kingdom, and the Association of periOperative Registered Nurses (AORN), to broaden and improve the anesthesia content. A collaborative agreement with AORN has allowed SNOMED to integrate and map the perioperative nursing data set to its core content. Another collaborative agreement with the APSF will enable SNOMED to enhance perioperative content with expanded patient-specific intraoperative content. The APSF collaboration will help support documentation in the operating room, improving data collection and analysis to reduce anesthetic errors and increase patient safety.
SNOMED CT is gaining broader recognition by those working to establish core standards and terminologies for use in a U.S. national health information infrastructure. In a historic five-year contract signed between The National Library of Medicine, part of the National Institutes of Health within the U.S. Department of Health and Human Services (HHS), and the CAP, English and Spanish language editions of SNOMED CT Core content and all version updates are now available free of charge to all U.S. health care institutions.
As part of its efforts toward recommending patient medical record information (PMRI) standards, in November 2003, the National Committee on Vital and Health Statistics within HHS recommended SNOMED CT as the general terminology for the U.S. core set for PMRI because of its breadth of content, sound terminology model, and widely recognized value.
Outside of the United States, the United Kingdoms Clinical Information Standards Board, responsible for approving clinical standards in areas such as clinical data sets terminology and clinical messages for direct patient care, supports SNOMED CT as the clinical terminology of choice for all electronic patient record development efforts throughout the United Kingdom. The United Kingdoms NHS has targeted 2005 for implementation of a first generation electronic record system and 2008 for a full electronic records service to be implemented nationally.
Working for Global Adoption
Reducing or eliminating barriers to adoption and use remains one of several basic principles that provide the foundation for SNOMED CT and guide its development activities. To achieve global adoption as the clinical terminology standard, SNOMED has established a global community of users, including the U.S. government; other national government entities, including the United Kingdom; and the private sector, which spans software and knowledge base suppliers, providers, payers, purchasers, and ultimately patients who actively participate in the future direction of SNOMED.
SNOMEDs development and maintenance efforts are focused on ensuring that its multilingual clinical terminology is recognized as best of breed; identified as a key building block of effective, modern health care information infrastructures; and adopted globally. Its ongoing work with individual experts, specialty professional organizations, and user groups help ensure that this world-class terminology is maintained and updated to keep pace with the evolving structure and content of clinical language. As part of its commitment to an open terminology development process, SNOMED sponsors working groups that vary in scope, from providing input about the direction of the terminology to a detailed review of specific nomenclature domains. Active or planned working groups include nursing, mapping, pharmaceuticals, genomics, pathology, public health, compositional grammar, guidelines, context of care, primary care, and anesthesia.
Moreover, teams of professional medical translators, linguists, and editors work on development of international editions of SNOMED CT. Commercial-grade quality assurance processes and translation tools are used to achieve maximum accuracy of concept representation in target languages. Teams of clinical validators from a variety of medical specialties, representing each country in which the target language edition will be used, review the work of the translation team. Localizations specific to dialectal variations of each language edition are prepared as needed.
SNOMED has also created a portfolio of products and services to support implementation. These include a developers toolkit, technical implementation guide, subset editor kit, annual users group and tutorials, and implementation training sessions. SNOMED also offers an electronic request submission process that provides opportunities for users to recommend improvements and ideas that advance the value of the SNOMED CT Core.
And finally, the design of SNOMED CT has been driven by the expressed needs of software developers for features that improve their ability to develop useful applications. In response to these needs, the SNOMED CT design adds unique numeric identifiers, includes links to legacy codes, supports a sustainable migration and maintenance strategy, permits adaptability for national purposes, and fosters alignment with other terminologies and standards such as HL7, DICOM, LOINC, XML, ISO, and ASC X12.
As an ANSI-approved standards developer, the CAP has committed to a review process that incorporates ANSIs minimum due process requirements. The CAP received ANSI approval for the Healthcare Terminology Structure Standard, specifying a standard file structure for use in distributing health care terminology.
