Vital Signs

The Clinical Decision Support Knowledge Artifact Specification, or KAS, is an HL7 standard that gives guidance on writing shareable decision support content in XML, including event-condition-action rules, order sets, and documentation templates. This post explains what a knowledge artifact is, how the HL7 specification evolved to use the Clinical Quality Language, and what five studies found on sharing CDS: a RAND prototype that built 22 artifacts, a national CDS Connect evaluation, a survey of 13 knowledge repositories, and two standards-based CDS projects.

Normalization has two meanings in health IT: creating a uniform, agreed set of standards, policies, definitions, and technical procedures so systems can interoperate, and organizing the fields and tables of a relational database to reduce redundancy. This post explains both and reviews three studies: a survey in which 19.6 percent of reported LOINC codes were incorrect, a 32-author consortium that normalized structured and unstructured EHR data for quality measures, and an FHIR-based pipeline with 62 normalization rules.

The Health Quality Measure Format, or HQMF, is an HL7 standard that represents a quality measure as an electronic document containing the information needed to compute the measure and its result. This post explains how HQMF works and how it relates to eCQMs and Clinical Quality Language, and reviews studies that translated HQMF queries into i2b2, executed measures at scale with MapReduce, and implemented 39 national quality indicators in CQL on FHIR.

SOAP, the Simple Object Access Protocol, is a messaging protocol that lets programs on different kinds of operating systems exchange information using HTTP and XML. This post explains what a SOAP message is, how SOAP differs from REST and FHIR, what a Minnesota public health project found when it used HL7 and SOAP web services to connect an immunization system to a disease surveillance system, and why message size has been a concern in hospital networks.

PIX, or Patient Identifier Cross-referencing, is an Integrating the Healthcare Enterprise profile that links the identifiers a patient has in different systems so that information about one person can be correlated. This post explains its actors and identifier domains, what a software architecture study said about combining master patient indexes with deduplication engines, what an Austrian project learned from IHE-based shared records, and what a machine learning study found about tuning a master patient index.

A national patient identifier is a system for uniquely identifying every recipient of health care. This post explains the idea and its link to HIPAA, why standards exist for provider and employer identifiers but not for individuals, and what studies found about duplicate records in master patient indexes, the demographic fields available for matching across nine health systems, and biased estimates when discharge data lack a unique identifier.

The Continuity of Care Record, or CCR, is a standard for a summary of a patient’s health information that a clinician can send at referral or transfer and a patient can carry. This post explains how the CCR was developed, how it relates to the Continuity of Care Document, and what studies found on its use in medication reconciliation, a Korean transfer note proposal, and personal health record data models.

The virtual medical record, or vMR, is a data model for representing the data that clinical decision support engines analyze or produce. This post explains why decision support needs a simplified record, how HL7 standardized one, what a multinational study found when a vMR-based data integrator supported guideline-based decision support, and how the vMR compares with openEHR archetypes in two comparison studies.

The semantic web is an effort to give the content of web documents meaning that machines can understand, using standards, markup languages, and processing tools. This post explains RDF and OWL in plain language, what a 37-paper systematic review found about semantic technologies and EHR data quality, what an 81-publication review found about ontologies in decision support rules, and what a review of 14 studies found about semantic interoperability in EHRs.

The Public Health Information Network, or PHIN, is described in the HIMSS Dictionary as the CDC’s vision for fully capable, interoperable information systems across the many organizations in public health, and as a national initiative for a multi-organizational business and technical architecture. This post explains the idea, what studies found about electronic laboratory reporting, electronic case reporting, and data quality, and what clinicians can do to make reporting work.

A detailed clinical model is an information model that expresses one or more clinical concepts and their context in a standardized, reusable way, as a set of discrete logical data elements. This post explains how DCMs relate to archetypes and FHIR resources, what a nursing study found when building 240 DCMs for perinatal nursing actions, and what a hospital study found when DCMs were used to reuse EHR data for COVID-19 research.

A lab results interface, or LRI, supports the electronic transmission of clinical results from a laboratory information system into an electronic health record. This post explains the HL7 implementation guide behind the term, why code mapping matters, what a laboratory review of EHR test builds found, what the CLIA test report rule requires of results systems, and what a large health system’s surveillance program found about missed abnormal results.

In the HIMSS Dictionary, key management services generate and manage the keys that identify records when data arrive from several source systems into an EHR repository. This post explains why source keys can clash, how identity keys differ from the cryptographic keys that protect data, what a multihospital study found when checking keys across 1,573 tables, and what a machine learning study found for patient record linkage.

QRDA is the HL7 standard that gives quality measure results a structured, consistent format so they can be sent between systems and reported to programs such as those run by CMS. This post explains the difference between patient-level Category I files and aggregate Category III files, how federal certification rules and CMS implementation guides use them, and what research shows about eCQM design, testing, and the burden of reporting.

A HISP, or health information service provider, is the Direct Project’s term for both a function, managing security and transport for directed exchange, and an organization that performs that function for senders and receivers. This post explains what a HISP does behind a Direct address, why a HISP can be a HIPAA business associate, what the regulation says about Direct in certified EHRs, and how directory management and growth statistics from the literature show the importance of this layer.

Direct messaging is a secure, encrypted, email-like way to send sensitive clinical and administrative information from one identified sender to one identified recipient. This post explains how it fits into ONC certification, what a radiology study found when Direct addresses grew from 8,724 to 752,496 in two years, why address directories decide whether a message can be sent, and what happens when recipients receive more outside data than they can manage.

An archetype is a named, machine-readable model that specifies how a clinical concept, such as a blood pressure observation or a medication order, should be captured and stored. This post explains how openEHR separates clinical models from software, how archetypes combine into templates, and what case studies of a coronary CT registry, a pediatric sepsis alert, a blood pressure monitoring module, and a corneal transplant record found.

NEDSS is an initiative that promotes data and information system standards to build efficient, integrated, and interoperable disease surveillance systems at federal, state, and local levels. This post explains how NEDSS fits with public health reporting, what a 2011 assessment found about state system capacity, what an Illinois pilot found when electronic case reports were compared with laboratory reports, and what the US Virgin Islands learned from migrating its NEDSS Base System to the cloud.

An IHE profile is a documented, tested specification from the Integrating the Healthcare Enterprise initiative that shows how existing standards should be combined to meet a specific integration need. This post explains how profiles are developed and tested, how Taiwan used the cross-enterprise document sharing profile for its national electronic medical record exchange across 321 hospitals, and what pilots of the XDS-I imaging profile found.

An interface is hardware or software designed to pass information between devices, between programs, or between a computer and a person. This post separates system interfaces, such as HL7 v2 feeds, from user interfaces, and explains what a scoping review of 28 studies found about how poorly designed EHR interfaces cause task switching, long navigation, fragmented information, and workarounds.