HIT Toolkit System Build Health Information Technology Toolkit for Physician Offices

HIT ToolkitSystem Build

Health Information Technology Toolkit for Physician Offices

2

Presenter• Margret Amatayakul

RHIA, CHPS, CPHIT, CPEHR, FHIMSS

President, Margret\A Consulting, LLCSchaumburg, IL

• Independent consultant, who focuses on achieving value from electronic health records, HIPAA/HITECH, and health information exchange. Developer of tools in Toolkit

• Adjunct faculty College of St. Scholastica, Duluth, MN, masters program in health informatics

• Founder and former executive director Computer-based Patient Record Institute, associate executive director AHIMA, associate professor University of Illinois

• Active participant in standards development, former HIMSS BOD, and co-founder of and faculty for Health IT Certification

2

3

Stratis Health● Stratis Health is a nonprofit organization that leads

collaboration and innovation in health care quality and safety, and serves as a trusted expert in facilitating improvement for people and communities

● Stratis Health works toward its mission through initiatives funded by federal and state government contracts, and community and foundation grants, including serving as Minnesota’s Medicare Quality Improvement Organization (QIO)

● Stratis Health operates the Health Information Technology Services Center for health care organizations seeking to use health information technology in support of their clinical transformation

4

• Understanding system build• System build tasks• Data conversion• Chart conversion• Interfaces• Legal health record

Agenda

5

• Install– Setting up hardware– Loading software onto hardware

• Implement– All activities associated with not only installation but hardware

configuration, workflow and process improvement, loading tables with your organization’s specifications and building the system to meet your requirements (“system build”), testing, training, and support for actual use (i.e., go-live)

• Adopt– The state where intended users actually use the system to

achieve specified, measureable goals

Definition of Terms

Copyright © 2005-8, Margret\A Consulting, LLC. Used with permission of author.

6

• Configuration of the software to meet internal policies, workflows, and process requirements– Also called “software configuration”

• “Configure” = arrange parts in a specific way for a specific purpose

Understanding System Build


Floor

4Bed

2W

Mary Smith

7

• Master files and tables/data dictionary build– Relational database– Values of variables– Metadata– Change control– Screen layout– Data entry shortcuts– Alerting strategies

System Build Tasks


8Copyright © 2005-8, Margret\A Consulting, LLC. Used with permission of author.

Relational DatabasePatient File

Med Rec # Name Birth date Street City, State Insurance

123498 Pam Bell 01121989 123 River Small, ST BCBS

125678 Jo Smith 10301972 RR 3 Rural, ST Aetna

Admission File

Name Date Time Mode Attending MD Adm Dx

Pam Bell 02142008 0900 Ambulance Dan James, MD AMI

Insurance Table

Aetna

BCBS

Paywell

Provider

TableNPI

Pat Carson, PA 8876

Dan James, MD 7543

Ted Smith, DO 1264

Dx Table

AMI

COPD

Fracture

Dan James, MD

Attending MDVariable

Value

9

• All computers that use a relational database to process data collect values of variables from tables to form files

• Some of these variable values will be known to you and relatively stable, such as the names and credentials of providers, nurses, etc.

• Vendors ask that you capture these values so they can be pre-loaded into the system master files and tables – either by the vendor or through you using a wizard

Values of Variables


10

• Data values for system build may come from a variety of sources:– A database or directory you maintain (data dictionary/metadata)– Various forms you use (Form and Reports Inventories)

• Data values relative to payer rules, formularies, drug knowledge are acquired by the vendor from their source (e.g., PBM) or a third party (e.g., ICD, CPT), with any subscription fees passed to users

• Standard vocabularies recommended for use by the federal government include:

– SNOMED – LOINC– RxNorm– UMDNS

Sources of Data Values


11

• SNOMED (SNOMED International) – Originally developed by College of American Pathologists and now distributed by

International Health Terminology Standards Development Organization, Denmark, SNOMED is licensed for use in U.S. by National Library of Medicine. It is a systematically organized computer processable collection of over 0.5 M medical concepts and 1.5 M semantic relationships covering most areas of clinical information such as diseases, findings, procedures, microorganisms, and pharmaceuticals for consistent indexing, storage, retrieval, and aggregation of clinical data across specialties and sites of care.

• LOINC (Logical Observations Identifiers Names and Codes) – Developed and maintained by Regenstrief Institute, includes over 41,000 names of

laboratory terms, as well as nursing diagnosis, nursing interventions, outcomes classification, and patient care data set. Each database record includes six fields for the unique specification of each identified single test, observation, or measurement:

Standard Vocabularies


12

• RxNorm – Provides standard names for clinical drugs (active ingredient + strength + dose form)

and for dose forms as administered to a patient. It provides links from clinical drugs, both branded and generic, to their active ingredients, drug components (active ingredient + strength), and related brand names. NDCs (National Drug Codes) for specific drug products (where there are often many NDC codes for a single product) are linked to that product in RxNorm. RxNorm links its names to many of the drug vocabularies commonly used in pharmacy management and drug interaction software, including those of First Databank, Micromedex, MediSpan, Gold Standard Alchemy, and Multum. By providing links between these vocabularies, RxNorm can mediate messages between systems not using the same software and vocabulary.

• Universal Medical Device Naming System (UMDNS)– A standard international nomenclature and computer coding system for medical

devices used in applications ranging from hospital inventory and work-order controls to national agency medical device regulatory systems and from e-commerce and procurement to medical device databases. UMDNS is maintained by ECRI and contains nearly 7,500 unique medical device concepts and definitions (preferred terms), along with an additional 8,000 entry terms to facilitate classifying of biomedical information

Standard Vocabularies



Metadata• Data about data

– Describes structured, or discrete, data element properties

– Must be kept up-to-date – Changes must be

documented (“change control” or configuration management)

Attributes Original

Name Attending MD

Table Physician

DB Name ATTMD

Synonyms Admitting Physician

Definition Member of medical staff who may admit patients

Reference Medical Staff Bylaws

Source Admission Screen

Derivations None

Valid Values Alphabetic

Conditionality Required

Default None

Lexicon (Standard

Vocabulary)

None

Relationship None

Access Any staff

(CDS) Process Rule

Convert to NPI for billing

Dan James, MD

Attending MD

14

• External reporting:– Common meaning?– Common representation?

• All data required to “fire” a rule must be present for a clinical decision support rule to work correctly– Changing the definitions of the data or requirements for their entry puts use of

the rule at risk– Changes to metadata may be admissible in a court of law if there is a question

as to spoliation of evidence

• A data administrator is often responsible for maintaining the integrity of a data dictionary; a database administrator makes the physical changes in the metadata registry

• Clinicians should approve all changes to the data dictionary, especially as they impact clinical decision support

• Degree of flexibility an organization has in managing changes to metadata varies with product

Importance of Metadata


15

• A data element may now be required to be entered (e.g., dose for a drug prescribed), but someone asks for it to be changed to optional (e.g., so dose would not always have to be entered).

• The decision to make such a change should be a thoughtful one, with an appreciation for its impact – If dose is not recorded by the user, will the system automatically

record a default, or can the user expect a call from the pharmacy?– If something goes wrong in the future (e.g., a default dose was not

changed when necessary), will you be able to track why and when the system change was made if necessary?

– If a future version of the software depends on this data element to be required and will not work properly as a result, will you be able to track that it was this change that is the cause of the problem now?

Example of Need for Change Control


16

• Some vendors enable screen design changes. Consider making such changes:– Only when absolutely necessary– For an entire group or organization– Such changes are costly to make and maintain

• Consider size and resolution of display screen– Not only will some screens not display well on smaller devices used for mobile

professionals,– But mobile professionals may use a variety of devices

• Consider user familiarity with computers– Data entry and retrieval must be intuitive

• Instructions must be clear, but not to obstruct power user• Icons must be able to be quickly understood (without mouse-over delays) within context

– Navigating multiple screens may result in even a power user getting lost• Balance reduction of complexity with need for information density

– Regenstrief Institute has found that once a person begins to use an EHR, one denser screen is preferred to multiple screens

Screen Layout



Screen Layout Strategies• Size and resolution of monitor (tablets

vs. notebooks vs. desktops)• User familiarity with computers

– Balance reduction of complexity with need for information density

• Sequencing, nesting, spacing, color, icons, navigation

• Alerting– Active

– Passive

• Variable Selection– Balance flexibility with standardization

• Data entry shortcuts• Templates


Data Entry ShortcutsSTRUCTURED DATA ENTRY• “Click” boxes• Drop down • Type ahead- - - - - - - - - - - - - - - - - - - • “Smart text” or macros• Default values• Cut (copy) and paste• Drag and drop• Drawing tools• Speech commands

ABILITY TO CONVERT VALUES OF VARIABLES TO STANDARD NARRATIVE

• “Click” boxes:– Check box = multiple options

may be selected– Radio button = only one choice

can be selected

UNSTRUCTURED DATA ENTRY• Dictation/speech dictation• Typing• Handwriting recognition

19

Alerting Strategies• Sounds or messages to

pagers, phone• “In-basket” functionality• Color, sound, &/or symbols

and indicators• Pop-up boxes (active alert)• Appearance of message or

icon (passive alert)• Context-sensitive templates


20

• Data conversion = permanently replacing data from one application to another application, such as moving the clinic’s patient schedule from a PMS to an EHR

• Interface = in comparison, an interface sends data from one system to another system, where both systems continue to operate on the data as applicable

• Master file and table build = also in comparison, allows stable data to be pre-loaded into new system

• Chart conversion = Making (selected) content of paper charts accessible/usable in EHR

Data Conversion


OLDNEW

DataConversion

Ch

art

Co

nve

rsio

n

21

• Chart conversion options (largely for ambulatory care)– Scan vs. abstract– Staff/contractor vs. physician– All of record vs. parts of record– All records vs. active records

• Other issues– Policy on chart availability after conversion– Closing charts after conversion– File records after conversion vs. warehousing vs. destruction– Legal aspects

Chart Conversion


22

• Start early– Prior to contracting, identify all interfaces

– Determine if EHR vendor can write all interfaces• Is a third party interface developer (system integrator) needed?

• Interface issues:– Uni-directional or bi-directional

– What data? (all or some)

– Will a portal do as well?• Need is to view information, not data (e.g., results review)• Need is to access and use (hospital) applications (e.g., enter

orders into CPOE)

Interface Build


23

• Interface – data entered into one system is also sent to another system

• Portal – entranceway to access applications and perform work at another site to which you are authorized

Interface vs. Portal


CPOE

HospitalClinic

24

Representative List of Interfaces


25

• Subset of all patient-specific data created or accumulated by a healthcare provider that may be released to third parties in response to legally permissible requests

• Basis is Federal Rules of Evidence, where in the business record is “that information kept in the course of a regularly conducted business activity.”

• Rules of e-Discovery, however, do not preclude metadata (including data dictionary definitions and changes, as well as date/time stamps of user entries and audit trails identifying what user accessed what data) from being further requested via court order

Legal Health Record


26

• Will structured data entry result in:

– Structured data output for subsequent processing?

– Print files intended to represent legal health record but are not customary or interoperable?

– Need to produce a screen shot for achieving the legal health record?

– Reports representing various needs, including for auditors, legal health record, subsequent use?

Converting Data to Output for Legal Health Record


27

Contact:

Stratis Health

2901 Metro Dr., Suite 400

Bloomington, MN 55425

952-854-3306

1-877-787-2847 (toll free)

www.stratishealth.org

For More Support

Documents

HIT Toolkit System Build Health Information Technology Toolkit for Physician Offices