2.1+Standards

=Standards=

//Before you begin//: Lessons are based on community-based, continually updated online sources such as [|Wikipedia]. Relevant terms for this lesson are listed under Topics and presented in a narrative format in the Read about sections. Click on each of the linked items and visit the Wikipedia article to get the most out of the lesson, and then hit the Back button on your browser to return to the lesson.

toc

=Goals=
 * Basic**
 * To be familiar with the major standards organization important to pathology practice
 * To understand the current and potential uses of structured medical languages

=Topics= [|diagnosis codes], structured medical language, [|natural language processing], [|Health Level 7](HL7), [|American Standard Code for Information Interchange] (ASCII), [|American Society for Testing and Materials] (ASTM), [|Logical Observation Identifier Names and Codes] (LOINC), [|Digital Imaging and Communications in Medicine] (DICOM), [|Systematized Nomenclature of Medicine] (SNOMED), [|International Classification of Disease]: ICD-9, ICD-10, [|Current Procedural Terminology] (CPT)

=Read about=

The importance of a standard
At times, it is in everyone’s best interests to follow a standard. For example, imagine a world without traffic rules, where everyone drove in a way that seemed most efficient and best to them. Most likely, there would be many more road accidents and a lot more traffic jams. In order to achieve gains in safety and speed for all involved, it is in everyone’s best interest to stop at a red traffic light, even if the light poses a temporary inconvenience to the people in front of it. (In game theory and economics, these are known as [|coordination problems] .) Now, let's say we visited another world where the traffic lights alternated between orange, purple, white, and pink. You can see what a problem this would present if we were inside a car in front of such a traffic light. Standards may be arbitrary in some ways, such as the colors used for traffic light signals, but it doesn't matter as long as everyone agrees on how to interpret them. Using standards, we can cross streets safely, drive on the "correct" side of the road, send email, [|land spacecraft safely on Mars] ... you get the idea.

Standards organizations
[|Standards organizations] develop, promote, revise, and maintain standards for the various industries and sectors in society, including healthcare. For instruments used in pathology laboratories, standards would presumably allow for inter-operability and the ability to share data between devices, even if they come from different vendors. In an ideal world, laboratory instruments from different vendors built on the same standards would be able to be connected with one another without much fuss, and communicate easily with the laboratory information system. In the real world, standard adherence is strictly voluntary, and not always closely followed. In addition, as technology moves forward, standards must also change to keep up, and therefore, it's a constant challenge to make sure everyone is on the same page.

There are a number of standards organizations important to pathology, among them [|Health Level Seven] (HL7) and [|Digital Imaging and Communications in Medicine] (DICOM). HL7 is a standards organization for hospital information systems accredited by the [|American National Standards Institute] (ANSI), and publishes messaging protocols that simplify the interface between healthcare software applications and various vendors. Many current LIS's use the messaging protocol HL7 version 2.x for sending data to and from from laboratory instruments. Laboratory instrument data representing results, quality control, normal range, patient medical record number, etc., is highly structured, and each set of numbers needs to be interpreted differently. In HL7 2.x, the data are interpreted by their position in a string of characters. Start and stop markers for positions are defined by special characters, such as a vertical bar ("|") or a caret ("^"). Even if a data field were to be blank, delimiters would still be included so that subsequent fields are interpreted appropriately. [|Extensible Markup Language] (XML) is the basis for HL7 version 3.0, and promises to provide even greater flexibility in data sharing.

DICOM is a standards organization for medical imaging. As the importance of digital imaging of gross specimen photos and histology slides grows in pathology, so too does the importance of imaging standards. The committee within DICOM that specifically works on pathology imaging standards is Working Group #26.

Structured medical language
A shared vocabulary for describing illness, diagnostic testing, and therapeutic interventions is incredibly useful for public health statistics, reimbursement schema, and health information systems.

Nomenclature for describing disease, like the [|International Classification of Disease] (ICD), has been used for the statistical study of morbidity and mortality in populations. For example, ICD codes are used on death certificates in the United States, and used to generate mortality figures. Incidence and prevalence statistics are also commonly created from ICD codes. Such data guides the allocation of national resources for medical research and education. Hospital reimbursement schemes by Medicare and Medicaid also use ICD (currently ICD-9 and ICD-10) codes to pay hospitals and clinicians a similar rate for caring for patients within one “[|diagnosis-related group] ” (DRG), with the assumption that patients in the same diagnostic category will utilize a similar level of hospital resources.

Reimbursement schemes for pathologists commonly use [|Current Procedural Terminology] (CPT) codes, rather than ICD codes. CPT tends to be clustered around procedural terminologies, rather than a disease process, and perhaps better capture what pathologists do.

Building inter-operability in pathology information systems also requires that devices share a language for testing purposes. For example, “chem 7,” “basic metabolic panel,” and “sodium, potassium, bicarbonate, chloride, BUN, creatinine, glucose” may be understood as the same panel of tests to a pathologist, but they would all mean different things to a computer. Having a standard vocabulary to differentiate laboratory tests can be useful in these situations. [|Logical Observation Identifier Names and Codes] (LOINC) is one such vocabulary set for laboratory test orders and results. LOINC and the [|Systematized Nomenclature of Medicine - Clinical Terms] (SNOMED CT) are part of on-going efforts to standardize words in all aspects of medicine, i.e. signs and symptoms, social history, drugs, disease processes, laboratory results, procedures, etc.

The trouble with narrative reports
Anatomic pathologists and radiologists have traditionally generated narrative reports for every specimen or film, with non-standard vocabulary and descriptors. As such, the information in the narrative reports has so far eluded capture for systematic study or quality control purposes. On its own, a computer would not be able to group similar clinical terms, nor would it recognize the very significant difference between terms like “malignant” and “benign.” Thus, one of the ongoing developments in standards for pathology is the creation of structured reporting for anatomic pathology reports. Use of a structured report across pathology institutions can form the basis for computer search functions for research or clinical purposes, communication of report contents to outside institutions, and identification and tracking of specimens and their derivatives.

=Activities= Visit the HL7 and DICOM websites and read the latest minutes from their last meeting.

=Online Resources=
 * Shaver, D. [|HL7 101: A Beginner’s Guide]. For The Record. 2007;Vol.19(1):22
 * Brodsky, VB and Gilbertson, JR. [|Data, Messaging Standards and Interfaces] (powerpoint). APIII Conference: October 2008.
 * Tuthill JM and Crowley. [|Next Generation Structured Data Reporting in Pathology: Closing the Communication Gap from Practice to Research] (powerpoint). APIII Conference: October 2008.
 * Goldsmith JD, Siegal GP, Suster S, Wheeler TM, Brown RW. [|Reporting guidelines for clinical laboratory reports in surgical pathology] . Arch Pathol Lab Med. 2008 Oct;132(10):1608-16.
 * AHRQ (Agency for Healthcare Research and Quality) . [|Health IT bibliography: Standards and Interoperability].

=Questions=

=Advanced courses=
 * Suzanne Flynn. Course materials for 24.900 [|Introduction to Linguistics, Spring 2005] . MIT OpenCourseWare, Massachusetts Institute of Technology.
 * Robert Berwick. Course materials for 6.863J / 9.611J [|Natural Language and the Computer Representation of Knowledge], Spring 2003. MIT OpenCourseWare, Massachusetts Institute of Technology.

=Expert corner= Help with Wikipedia article(s): structured medical language (no article), [|natural language processing]

flat This page was last modified by user:{$revisioneditor} on {$revisiondate}.