Introduction_to_Healthcare_Informatics_Second_EditionChapter

Introduction to Healthcare Informatics, Second Edition

Chapter 11:

Using Healthcare Data and Information

Objectives

Describe the different secondary uses of healthcare data and information

Define unstructured data and structured data

Delineate how natural language processing is used to support secondary uses of healthcare data and information

Explain the opportunities and challenges encountered when using unstructured data for secondary uses

List and briefly define the major datasets, classification systems, clinical terminologies, and other standards utilized for secondary data use

Utilization of Healthcare Data

Primary

Direct patient care

Secondary

Nondirect care use

Collect once, use many

Unstructured Data

Free text or string data

Why use unstructured data?

If used, what is most effective way?

Natural Language Processing

Converts human language into data that can be translated then manipulated by computer systems

Ontology

A branch of artificial intelligence

Uses of NLP in Healthcare

Support high-quality patient care

Discovery of point-of-care data

Biosurveillance

Administrative support

Coded and Structured Data

Also known as discrete data

Binary, machine-readable data in defined fields

Vocabulary

Terminology

Classification system

Healthcare Code Sets

Any set of codes used for encoding data elements, such as tables of terms, concepts, medical diagnostic codes or medical procedure codes

Importance of standardization

HIT Standards Committee

Vocabulary Task Force

Meaningful use EHR certification criteria

International Classification of Diseases

ICD-9-CM

12,000 diagnostic codes

4,000 inpatient procedure codes

Format: XXX.XX

Transition to ICD-10-CM/PCS

October 1, 2014

Development of ICD-10-CM in the United States

Structural changes

21 chapters

Format: XXX.XXXX

Inclusion of alpha characters

Use of a dummy placeholder

Organizational changes

Incorporation of factors influencing health status and external causes

Injuries now grouped by body part not injury type

Excludes notes moved to beginning of each chapter

Postoperative complications in procedure specific body system chapters

Expanding the detail

Laterality

Complications

Seventh character extension to indicate initial encounter, subsequent encounter or sequelae

Procedure Codes

No international standards

US Inpatient Procedure Codes

CMS and 3M Health Information Systems

ICD-10-PCS

Seven character alphanumeric structure

Meaning of ICD-10-PCS Characters

Moving to ICD-11

World Health Organization

Tripartite structure:

An ontological core in partnership with SNOMED

A Foundation Layer of richly interlinked concepts with explicit preferred terms, fully specified terms, synonyms, definitions, and detailed attributes

Several “linearizations” deriving from the Foundation Layer that will resemble the traditional mutually exclusive and exhaustive coding structure of historical ICD tabular versions

Linearization

A subset of the ICD-11 foundation component that is

Fit for a particular purpose: reporting mortality, morbidity, or other uses

Jointly Exhaustive of the ICD Universe (Foundation Component)

Composed of entities that are mutually exclusive of each other

Each entity is given a single parent

Postcoordination of Many Desired Attributes

To avoid excessive repetition

Terms with modifiers created as needed

“Sanctioning rules” to be computationally executed not manually practiced

Pathway for graceful evolution from ICD-10 to ICD-11 is possible

Clinical Terminologies

Healthcare Common Procedure Coding System (HCPCS)

Level I: CPT

Level II: HCPCS

Systemized Nomenclature of Medicine—Clinical Terms

Concept identifier

Highly detailed and granular

Relationship to meaningful use

Multiple Levels of Granularity in SNOMED-CT

Figure reprinted by kind permission of the International Health Terminology Standards Development Organisation (www.ihtsdo.org).

Clinical Terminologies (continued)

Logical Object Identifiers Names and Codes (LOINC)

Laboratory codes and clinical test descriptors

Relationship to meaningful use

RxNORM

Drug names

Corresponding meaningful use specification

LOINC Examples

Other Data Set Standards

Information model and messaging standards

HL7 Version 2 bar-delimited ASCII strings

Consolidated Clinical Data Architecture

Meaningful use health information exchange

Clinical Information Modeling Initiative

Metadata

Descriptive data that characterize other data to create a clearer understanding of their meaning and to achieve greater reliability and quality of information

Essential for understanding of the underlying data

Can be important for investigative or legal reasons

Help to explain the data

Secondary Data Uses in the Future

Data quality metrics

Population-based guideline application

Best evidence discovery

Adverse event detection

Comparative effectiveness analyses

Clinical and outcomes research

Clinical decision support

Big Data

High-volume, high-velocity, and high-variety information assets

Differentiators:

Extremely large databases

Speed the data needs to and does move around the system

Variety of the data

Comparable and Consistent Representations

Normalized to conform to designated standards

High-throughput clinical phenotyping

Summary

Availability of secondary data is a driver for the implementation and use of HIT

Data standards will need to be combined with analytical processes and tools

A Better Understanding of the Chapter 11 (and how it all relates!)

Using Healthcare Data and Information

The scale of data is mind-boggling

The world’s “digital universe” is in the process of generating 1.8 Zettabytes of information

With continuing exponential growth this projects to 40 Zettabytes in 2020

Two put this into perspective, 1ZB is 1000 EB or a billion TB.

Five exabytes (10^18 gigabytes) of data would contain all words ever spoken by human beings on earth

The source?

Over 5 billion mobile devices in use worldwide. Each call, text and instant message is logged as data. Mobile devices, particularly smart phones and tablets, also make it easier to use social media and use other data-generating applications. Mobile devices also collect and transmit location data

Billions of internet transactions – purchases, funds transfers, share trades – every day, including countless automated transactions. Each creates a number of data points collected by retailers, banks, credit card issuers, credit agencies and others

Millions of networked electronic devices – including servers and other IT hardware, smart energy meters and temperature and other sensors – all create semi-structured log data that record every action

Hundreds of millions of social networking users and resources. Each Facebook update, Tweet, blog post and comment creates multiple new data points, both structured and unstructured

Big Data

Data sets of such size, complexity and volatility that organizations value cannot be fully realized with existing data capture, storage, processing, analysis and management capabilities

The systematic use of unstructured data is a Big Data challenge!

blog3

Get help from top-rated tutors in any subject.