Electronic Health Records

A Practical Guidefor Professionals and Organizations

Margret K. Amatayakul, MBA, RHIA, CHPS, FHIMSS

Copyright ©2004 by the American Health Information Management Association. Allrights reserved. No part of this publication may be reproduced, stored in a retrieval system,or transmitted, in any form or by any means, electronic, photocopying, recording, orotherwise, without the prior written permission of the publisher.

ISBN 1-58426-133-1AHIMA Product No. AB102604Production No. IPC-1500-0504

AHIMA Staff:

Marcia Bottoms, Acquisitions Editor and Director of PublicationsCarol A. Quinsey, RHIA, CHPS, HIM Practice ManagerKatherine Kerpan, Senior Editor

Formerly published as The Role of Health Information Managers in CPR Projects:A Practical Guide, ©1999 AHIMA.

AHIMA strives to recognize the value of people from every racial and ethnicbackground as well as all genders, age groups, and sexual orientations by building

its membership and leadership resources to reflect the rich diversity of the Americanpopulation. AHIMA encourages the celebration and promotion of human diversity

through education, mentoring, recognition, leadership, and other programs.

American Health Information Management Association233 North Michigan Avenue, Suite 2150

Chicago, Illinois 60601-5800

http://www.ahima.org

About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ivForeword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .viiAcknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xi

Chapter 1 Introduction to Electronic Health Records . . . . . . . . . . . . . . . . . . . . . . . .1

Chapter 2 Challenges to EHR Adoption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Chapter 3 Roles in EHR Design and Implementation . . . . . . . . . . . . . . . . . . . . . .43

Chapter 4 EHR Migration Path Impact on Quality . . . . . . . . . . . . . . . . . . . . . . . .71

Chapter 5 Strategic EHR Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93

Chapter 6 Healthcare Process Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111

Chapter 7 Functional Needs Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .129

Chapter 8 Data Infrastructure Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .145

Chapter 9 Technology Infrastructure Assessment . . . . . . . . . . . . . . . . . . . . . . . . .171

Chapter 10 EHR Functional Requirements Specification . . . . . . . . . . . . . . . . . . . .191

Chapter 11 Return on Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .229

Chapter 12 EHR Selection and Contract Negotiation . . . . . . . . . . . . . . . . . . . . . . .253

Chapter 13 EHR System Implementation and Ongoing Maintenance . . . . . . . . . .273

Appendix A A Vision from the e-HIM™ Future:A Report from the AHIMA e-HIM™ Task Force . . . . . . . . . . . . . . . .285

Appendix B HL7 Draft Standards for Trial Use . . . . . . . . . . . . . . . . . . . . . . . . . . .307

Appendix C HIM Checklist for EHR Project Involvement . . . . . . . . . . . . . . . . . . .361

Appendix D AHIMA’s Data Quality Management Model . . . . . . . . . . . . . . . . . . . .367

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .375

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .401

Contents

iii

Margret K. Amatayakul, MBA, RHIA, CHPS, FHIMSS, has more than thirty years of expe-rience in national and international healthcare information management. She is a leadingauthority on electronic health record (EHR) strategies for healthcare organizations and hasextensive experience in EHR selection and project management.

After having served as director of medical record services at the Illinois Eye and EarInfirmary, associate professor at the University of Illinois at the Medical Center, associateexecutive director of the American Health Information Management Association, andexecutive director of the Computer-based Patient Record Institute, Margret formed herown consulting firm, Margret\A Consulting, LLC, in 1999. She provides information sys-tems consulting services, freelance writing, and educational programming to the health-care industry.

Margret earned her bachelor’s degree from the University of Illinois at the MedicalCenter and her master’s degree in business administration, with concentrations in market-ing and finance, from the University of Illinois at Chicago. She is a much-sought-afterspeaker and has published extensively. She also has an impressive list of professional serviceawards to her credit. Finally, Margret continues to serve the health informatics communityas adjunct faculty of the College of St. Scholastica and University of Illinois at Chicago.

Margret can be contacted at margretcpr@aol.com.

iv

About the Author

The vision of the electronic health record (EHR) has been in existence for almost twodecades. Initially, it was called the computer-based patient record (CPR); in the mid-1990s,it became known as the electronic medical record (EMR). Although now known as theEHR, the vision never really changed throughout the years and the various name changes.From its earliest conception, the promise of the EHR was great. From improving the qual-ity of care to supporting research and education to facilitating disease monitoring and thedevelopment of health policy, the EHR was—and still is—the concept that would foreverchange the way healthcare is practiced.

But with great promise comes great complexity. As attempts were made to implementthe EHR, it became clear that it was not a single system but, rather, a collection of inter-locking systems that were tied to a series of complex clinical and administrative workflows. The changes required by EHR implementation are massive. Work flows need to bereengineered. At times, technology infrastructure needs to be upgraded. Legacy systemsoften need to be replaced or upgraded. All of this requires strong project management.Moreover, such technological and organizational changes can be costly to an organization,causing pressure to achieve the maximum return on investment.

Clearly, a book that can sort out all of these issues and provide solid recommendationson how to implement an EHR is as invaluable a reference to the practicing professional asit is an educational tool for the student learning about EHRs in the classroom. Few in thehealthcare industry can write as authoritatively on this topic as Margret Amatayakul can.With her background and experience as a health information management professional,past executive director of the Computer-based Patient Record Institute (CPRI), consultant,and government advisor, Margret has dealt with the issues from vantage points that fewcan claim in one professional career. She has taken that experience and marvelously inte-grated the history of the EHR—its functions, high-level technical infrastructure, anddetail-level data architecture—and spiced it with practical tips on how to make an effec-tive business case and manage what could be an overwhelming project—all in one book.

Foreword

v

vi Foreword

As the EHR matures and truly becomes the national standard to be developed andimplemented in software products over the remainder of this decade, this book, rather thansitting on a shelf, is going to be well worn by practicing professionals constantly referringto it as they plan for, select, and implement an EHR. Students are hereby advised to keepthis book and take it with them into their jobs.

Donald T. Mon, Ph.D.

Vice President, Practice LeadershipAmerican Health Information Management Association

Co-chair, EHR Functional Outline Direct Care Work GroupCo-chair, Care Setting Definitions and Profiles Work Group

Health Level 7 (HL7) EHR Functional Model and Draft Standard

Co-chair, EHR Outreach Work GroupEHR Collaborative

It is very apropos that this second edition of a book on planning and managing electronichealth record (EHR) projects is written at the time of a tremendous surge of interest in not onlyautomating healthcare documentation, but also in gaining the benefits that can accrue fromcomprehensive clinical information decision support and knowledge management systems.

From a mention in President Bush’s State of the Union Address in 2004 to the myr-iad individual projects focused on improving healthcare and patient safety, the EHR’stime has come.

Still, many healthcare professionals and organizations seek guidance on EHR defini-tion, migration path, planning, selection, and implementation. The changes that accom-pany the computerization of health information are huge. Many healthcare professionalsstill find it a challenge to fully adopt the use of computers that store information securelyout of sight and require structured data entry and controlled vocabularies to dictate whatinformation is collected, who collects it, and how it is recorded. Clinical guidelines andprotocols embedded in EHR systems direct the development of care plans. Standardizedreports, reminders, and alerts for clinicians are generated automatically by the system.

More than ever before, EHR system creation and use encompasses the need for allstakeholders to work together. Although its basis is clinical information, the EHR is usednot only by clinicians, but also by virtually every other health professional who manageshealthcare payment, risk, quality, research, education, and operations. Patients, too, areincreasingly gaining access to their health information and making important contributionsto their personal healthcare by doing so. EHRs serve as the foundation for populationhealth and, ultimately, the potential for a national health information infrastructure.

The preface to the first edition of this book spoke extensively about the power of infor-mation and the fear that giving up paper records somehow would result in a power shift inthe healthcare community. It was noted that:

What physicians, nurses, allied health professionals, and health information professionals need torecognize is that the CPR [EHR]—and the information system technologists who operate the com-puter systems supporting it—are not taking control of health information. The CPR [EHR] is merely

Preface

vii

one tool for change in the evolving healthcare system. The true forces behind the sweeping changesin health information systems as in healthcare services and reimbursement are worldwide economictrends and ongoing developments in sophisticated communications and computer technology thatare beyond everyone’s direct control.

The ultimate goal of everyone who creates, uses, and manages health information should be toprovide the best healthcare possible (Amatayakul 1999, vii).

This remains as true today as it did five years ago.

Objectives of This Practical GuideElectronic Health Records: A Practical Guide for Professionals and Organizations isintended to be a guide for individuals who are interested or involved in the EHR planningand implementation process, whether in a healthcare organization, working for a vendor, orin a policy-making position. The book covers the elements of the EHR process in chrono-logical fashion, from developing a migration path to implementation and ongoing manage-ment of systems. Along the way, it provides a solid background about EHR history, trends,and common pitfalls. Where the first edition of the book was targeted specifically to thehealth information management professional, this edition broadens the target audience to allstakeholders. Whether it is the board or executive management that needs an overview ofEHR, clinicians who want to understand the level of change to expect from an EHR, opera-tional managers who must make change happen, information technologists new to the health-care environment, or even politicians and government officials who must understand thescope and complexity of the project, this book is intended to serve as a professional resource.

Although not intended for use exclusively as a textbook, Electronic Health Records: APractical Guide for Professionals and Organizations can help translate complex discus-sions about EHRs into everyday, understandable language. The book is suitable for appliedhealth informatics students and others who are training to play an active role in their orga-nization’s journey toward a fully functional EHR.

The first step in any journey is to decide whether the trip is worth taking. Chapter 1 ofthis book defines EHR, identifies early attempts at development, and relates current status.Chapter 2 discusses challenges to adoption, with the intent of overcoming myths and cap-italizing on lessons learned.

Chapter 3 describes the importance of an EHR project plan and identifies the roles ofthe various stakeholders in EHR planning. Chapter 4 is new to the second edition becauseof the intense need to address return on investment (ROI) in an age of diminishing capitaland tight resources. It not only describes the concept of the migration path that was dis-cussed in the first edition but also relates each step in the migration to specific benefits.Chapter 5, then, emphasizes the importance of coordinating EHR planning with the orga-nization’s strategic initiatives.

Chapters 6 through 9 cover the various forms of assessments that serve as the baselinefor EHR planning. Chapter 6 focuses on healthcare processes and work flow analysis.Chapter 7 provides a functional needs assessment. Chapter 8 describes the data infrastruc-ture necessary for an EHR. Chapter 9 provides a primer on EHR technology.

viii Preface

Chapter 10 pulls the various assessment components together to describe a require-ments specification. Chapter 11 includes specific methodologies for determining ROI andimpact on quality. Chapter 12 provides detailed advice on how to select an EHR vendor,and chapter 13 provides a summary of the implementation requirements and ongoingactivities associated with continuing to build a comprehensive EHR.

Although the book covers the steps in EHR planning and implementation in a logicalsequence from the beginning of the project to the end, each reader may want to focus oncertain chapters. Each reader brings a different background to the EHR project. Some willbe knowledgeable about computers and simply want to know more about the healthcarecontext and unique features of EHR systems; others may need a refresher on the latestinformation technology.

Any Web sites listed in this book were current and valid as of the date of publication.However, Web page addresses and the information on them may change or disappear atany time and for any number of reasons. The user is encouraged to perform general Websearches to locate any site addresses listed here that are no longer valid.

Note to educators: Instructor materials for this book include lesson plans, chapter slides,and other useful resources. The instructor materials are available in online format throughthe Assembly on Education (AOE) Community of Practice (CoP). Instructors who areAHIMA members can sign up for this private community by clicking on the help iconwithin the CoP home page and requesting additional information on becoming an AOECoP member. An instructor who is not an AHIMA member or an AHIMA member who isnot an instructor may contact the publisher at publications@ahima.org. The instructormaterials are not available to students enrolled in college or university programs.

ReferenceAmatayakul, Margret. 1999. The Role of Health Information Managers in CPR Projects: A Practical Guide.Chicago: American Health Information Management Association.

Preface ix

The first edition of this book acknowledged the contribution of Rita Finnegan, past presi-dent and former executive director of AHIMA, who brought me into the field of healthinformation management and proposed to the board of directors that the AHIMA supportthe Institute of Medicine’s patient record study. In assigning me to participate in that pro-ject, Rita gave me the opportunity to help form the Computer-based Patient Record Insti-tute (CPRI) and to extend my network of associates and influence well beyond traditionalboundaries.

That network has only continued to grow and be enriched. My colleagues and my stu-dents are my extended family in learning, and this book would not have been possible with-out their support, shared expertise, and assistance. The current members of the AHIMA staffand the external reviewers have contributed enormously in their trust, comments, and ideas.These individuals include:

Elizabeth D. Bowman, MPA, RHIA Nadinia A. Davis, MBA, CIA, CPA, RHIA Susan Helbig, MA, RHIA Frances Wickham Lee, DBA, RHIA Shirley Eichenwald Maki, MBA, RHIA, CPHIMS Donald T. Mon, Ph.D.Lisa E. Morton, MLIS, RHIA Carol A. Quinsey, RHIA, CHPS Nanette B. Sayles, Ed.D., MPA, RHIA, CCS, CHP Scott D. Stratton, MPH Catherine N. Turner, BSN, MBA, RNC Kathy C. Trawick, EdD, RHIA

Of course, my husband Paul remains my reason for perseverance.

Acknowledgments

xi

In his 2004 State of the Union Address, President Bush noted: “By computerizing healthrecords, we can avoid dangerous medical mistakes, reduce costs, and improve care.” Thisstatement has been applauded by many who recognize that electronic health records (EHRs)represent a huge opportunity to improve patient care and health system operations. However,efforts to achieve an EHR have represented a long journey from early visions to today’s real-ity of possibility. The EHR is not a simple computer application; rather, it represents a care-fully constructed set of systems that are highly integrated and require a significant investmentof time, money, process change, and human factor reengineering. This chapter:

• Describes the framework of the EHR as defined by the landmark work of theInstitute of Medicine

• Compares the functionality of various efforts to achieve the envisioned benefits

• Summarizes the migration path of information systems technology needed toachieve the EHR goal

• Discusses the current status of EHR development and implementation

• Provides an overview of the stages of EHR implementation

Electronic Health Record Definition

Defining the electronic health record is not simple. Indeed, there is no standard definition. TheEHR is not an information system that is purchased and installed as a word-processing pack-age is, or even billing systems or laboratory information systems that may have to be con-nected to other information systems and devices and customized to the specific environment.The EHR is more an information system framework that accomplishes a set of functions.

The Criteria for an Electronic Health Record

Possibly the simplest way to describe the EHR is to refer to the early work of the Computer-based Patient Record Institute (CPRI). The CPRI, now part of the Healthcare Information

Chapter 1Introduction to Electronic Health Records

1

and Management Systems Society (HIMSS), identified three key criteria for an EHR. Theelectronic health record must:

• Integrate data from multiple sources

• Capture data at the point of care

• Support caregiver decision making

The intersection of these criteria may be considered the EHR, as illustrated in figure 1.1.

Components of the Electronic Health Record

Many contemplating what an EHR is want to understand what information system (IS)components comprise an EHR. Figure 1.2 displays a conceptual model that depicts thetechnical system components that capture and integrate data and support caregiver deci-sion making. However, because the EHR is as much a set of functions that provide valueas it is technical system components, the conceptual model also shows the relationship ofthe technical components to the value of EHR. The value of integrating clinical, financial,and administrative data contributes significantly to improvements in quality, cost, andaccess to healthcare. Furthermore, the EHR is not limited to a single location but, rather,should include remote access for providers and consumers. Potentially, it should even becapable of integrating data across providers and from personal health records to form alongitudinal view of an individual’s health status and healthcare.

Therefore, the technical system components of the EHR include:

• Source systems that capture data to support the EHR infrastructure. Thesesource systems include all administrative, financial, and clinical departmentalsystems that relate in any way to the health record.

2 Chapter 1

Figure 1.1. EHR criteria

Collect datafrom multiple

sources

Used asprimarysource

of informationat point of

care

Providedecisionsupport

EHR

• Supporting infrastructure that integrates data, which may include a data reposi-tory that centralizes data from other components (or other means to integratedata); a rules engine that supplies programming logic for decision support, suchas alerts and reminders, order sets, and clinical protocols; knowledge sourcesthat make information available from various external sources; and data ware-houses from which specific data can be mined (that is, aggregated and analyzed)to provide useful information.

• Human–computer interface that helps capture data at the point of care and, byvirtue of access to data, rules, knowledge, and mined data, supports caregiverdecision making. It may be any means by which users enter and retrieve data,such as a computer workstation, personal computer (PC), notebook computer,personal digital assistant (PDA), voice recognition system, handwriting recogni-tion system on a tablet PC, and so on.

The end result of implementing the technical components of the EHR is that quality,cost, and access to healthcare are enhanced through clinical, financial, and administrativedata support. The American Health Information Management Association’s (AHIMA’s)Health Information Vision 2010 states that “health information will be used concurrentlyfor multiple and diverse purposes, including healthcare delivery and treatment, outcomesmeasurement, finance, and support of health services and policy research, clinical trials,

Introduction to Electronic Health Records 3

Figure 1.2. Conceptual model of EHR

Copyright © 2004, Margret\A Consulting, LLC. Reprinted with permission.

Source

Lab

Rx

Radiology

Devices

SystemsImages

OtherWarehouse

Data MiningTools

KnowledgeSources

RulesEngine

Repository

QualityCost

Access

Human–ComputerInterface

ConsumerAccess

RemoteAccess

Clinical

Adminis

trat

iveF

inancial

and disease prevention and surveillance at the individual, community, national, and inter-national levels” (AHIMA 2003). (See appendix A.)

Origins of the Electronic Health Record

The concept of the EHR has existed since early use of computers in healthcare and hasundergone a significant transformation over time. Figure 1.3 illustrates the major mile-stones in the history of EHR implementation.

Pioneers of the Electronic Health Record

The first major efforts to automate clinical information occurred in the late 1960s and early1970s. Several forward-thinking universities and companies recognized the value ofemerging information technology (IT) for healthcare.

Some of the early attempts at automating the health record were highly successful andare precursors of today’s products. For example, efforts by Wiederhold at Stanford Uni-versity and El Camino Hospital conducted with scientists at the Lockheed Company areoften cited among the forerunners of the first commercial products. This early clinicalinformation system (CIS) effort was subsequently taken over and further developed bythe Technicon Corporation (subsequently TDS Healthcare Systems Corporation, which isnow part of the suite of products available from Eclipsys, Inc.).

Another pioneering provider–vendor partnership took place at Latter Day Saints (LDS)Hospital, now Intermountain Healthcare, with the HELP system on which 3M has basedproducts. Other organizations have contributed significantly to research in the design ofCISs. Massachusetts General Hospital in Boston; Kaiser Permanente in Oakland, Califor-nia; and Regenstrief Institute in Indianapolis have contributed significantly to the EHRbody of knowledge (Shortliffe and Perreault 2001).

The National Library of Medicine (NLM) also has contributed greatly to early effortsand continues an extremely active role in data and vocabulary development. Its IntegratedAdvanced Information Management Systems (IAIMS) grants have helped—and continueto help—fund many EHR-related projects. (For a comprehensive history of medical infor-matics, see M. F. Collen’s A History of Medical Informatics in the U.S. 1950–1990.)

Early Limiting Factors

For the most part, applicability of the early EHR projects was limited to the environmentsin which they were created. Products often could not be “commercialized” or made readily

4 Chapter 1

Figure 1.3. History of EHR implementation

AcademicExperiments

Feeder Systems

2000s

EHR Hypeand Components

Integrationand Connectivity

1970s 1980s 1990s

able to be implemented in other settings because they were so closely linked to processes atone organization.

Furthermore, in the early days of computer use, most healthcare organizations lackedthe source systems—the laboratory, radiology, pharmacy, and other ancillary services—tosupply an EHR with the data needed to provide users with much value. Thus, the automa-tion that ensued in the 1980s focused on the relatively simpler, but critical, source systemsand those that produced more immediate payback. Initially, these were administrative andfinancial systems such as registration–admission/discharge/transfer (R-ADT), master per-son index (MPI), and accounts receivable (A/R). Later, departmental systems for labora-tory, radiology, pharmacy, dietary, materials management, and others were developed andimplemented. In general, new interest in automating the health record itself waned as othersystems were being implemented.

Landmark Effort

In the mid-1980s, frustrated by the inadequacies of the paper health record and slowprogress toward automating clinical data, the Institute of Medicine (IOM) initiated a studyon “improving the patient record in light of new technology.” In 1991, the IOM released areport of its study entitled The Computer-based Patient Record: An Essential Technologyfor Health Care. The IOM coined the term “computer-based patient record” (CPR), morecommonly referred to today as the electronic health record (EHR). At the time, the CPRrepresented a huge leap from the concept of documentation that primarily supported theprovider to documentation that focused on the person receiving care. The IOM’s landmarkwork laid the conceptual foundation for a vision of a system that would:

provide a longitudinal (i.e., lifelong) record of events that may have influenced a person’s health(IOM 1991, 137) and reside[s] in a system specifically designed to support users by providing acces-sibility to complete and accurate data, alerts, reminders, clinical decision support systems, links tomedical knowledge, and other aids (IOM 1991, 11)

EHR Vision

EHRs may not yet be as widely implemented as the IOM would have desired, but the IOMvision remains firm. In 1997, a second edition of the original 1991 study report was under-taken. The result was retention of all original content as being valid and current for thetime, with updates relative to implementations in the United States and Europe. The IOMvision for EHR remained committed to the use of IT for a number of purposes, including:

• To document all episodes of healthcare wherever they take place

• To provide immediate access to data

• To process data in a variety of ways to support better decision making for patientcare and clinical and health services research

• To increase the efficiency of healthcare organizations and decrease the cost ofservices

• To ensure the confidentiality of data

• To improve the quality of care and promote the wellness of the population

Introduction to Electronic Health Records 5

The IOM’s contribution was significant not only because it addressed a need, but alsobecause it defined the scope of the transition to the EHR. This transition is characterizedby the following key quotes from the IOM report:

• “Merely automating the form, content, and procedures of current patient recordswill perpetuate their deficiencies and will be insufficient to meet emerging userneeds” (IOM 1991, 2)

• “The CPR encompasses a broader view of the patient record than is currenttoday, moving from the notion of a location or device for keeping track of patientcare events to a resource with much enhanced utility in patient care (includingthe ability to provide an accurate longitudinal account of care), in management ofthe health care system, and in extension of knowledge” (IOM 1991, 3)

• The CPR is “the core of health care information systems. Such systems must beable to transmit data to other types of clinical and administrative informationsystems within healthcare institutions; they must also be able to transmit data toand accept data from other healthcare institutions or secondary databases” (IOM1991, 51)

EHR Terminology

A number of terms have come to be associated with the means to accomplish the IOM’sgoals. Some of these are generic; others are vendor specific.

As noted above, the IOM coined the term “computer-based patient record” (CPR),which it defined as “an electronic patient record that resides in a system specificallydesigned to support users by providing accessibility to complete and accurate data, alerts,reminders, clinical decision support systems, links to medical knowledge, and other aids.”

Another popular term is “electronic medical record” (EMR). This term continues tobe widely used, especially to describe systems based on document imaging or systemsimplemented in physician offices.

“Electronic health record” is the term used in this book and the term chosen in the IOMletter report, Key Capabilities of an Electronic Health Record System, which it submittedto the U.S. Department of Health and Human Services (HHS) in July 2003. Moreover,EHR is the term being incorporated into a standard description of an EHR being developedby Health Level Seven (HL7), a standards development organization, in collaborationwith the IOM. (See appendix B for the Draft Standard for Trial Use [DSTU] developed byHL7 in 2004 and chapter 9 for a comprehensive analysis of this work.) EHR is believed torepresent the most comprehensive vision of an information system that would support alltypes of caregivers, in all settings, including the individual who may be using it to recordpersonal health status information.

EHR Migration Path

Although the variety of terms initially may have come about because of vendors attempt-ing to differentiate their products, some in the industry have tried to use the different termsto describe either different visions or a migration path toward an ultimate vision.

6 Chapter 1

The many functions that an EHR may address, from simple to sophisticated, include:

• Document-scanning/imaging systems

• Order communication/results retrieval (OC/RR) systems

• Clinical messaging systems

• Patient care charting

• Computerized physician/provider order entry (CPOE) systems

• Clinical decision support systems

• Provider–patient portals

• Personal health records

• Population health

These are the most common and significant functions. There are many other, moredetailed functions—or at least subfunctions—that others might highlight.

Document-Scanning/Imaging Systems

Document-scanning/imaging systems represent a wide range of functionality. Some merelycapture images of the forms in the paper record for storage in a computer system for laterretrieval. Most systems today index the data on the forms to help retrieve specific data.Some document-imaging systems combine scanned forms with forms that have been com-puter generated. This is known as computer output to laser disk (COLD). Finally, documentimaging can be integrated with work flow technology.

Work flow technology helps direct work on documents. For example, it can determinewhen a record is ready for coding and put it into the appropriate coder’s work queue.Simultaneously, clinicians can view the documents when access is needed for subsequentpatient care. The patient financial services department also may access the documents togenerate a claims attachment. However sophisticated the ability to select blocks of text toview or how easily work may be distributed for processing, document imaging primarilyaffords access to what was originally paper record content from multiple locations.

Order Communication/Results Retrieval Systems

Order communication/results retrieval (OC/RR) systems provide the capability of trans-mitting orders to various ancillary departments and viewing results of laboratory andother diagnostic studies or the status of orders. OC/RR systems go a long way towardintegrating the various source systems for operational purposes. However, they are essen-tially paper based because they rely on handwritten orders that are key-entered into acomputer by clerical or nursing staff and diagnostic studies results that are typically gen-erated in a paper format. These documents may be from ancillary information systems,such as laboratory systems, or from transcription systems where radiology and otherreports are dictated and transcribed. Some organizations use COLD feed to compile thesedocuments into a central automated archive; others merely file the documents into thepaper chart folder.

Introduction to Electronic Health Records 7

OC/RR systems afford the functionality of automatically transmitting orders insteadof telephoning, using a courier, or faxing, and of providing access to results from multi-ple locations.

Clinical Messaging Systems

Clinical messaging systems add to CR/RR systems the dimension of real-time interactionthrough Web-based technology. The Web-based technology may be applied within theorganization’s own internal network, or intranet, or may entail exchange of informationthrough a secure Web portal from the Internet.

Clinical messaging systems enhance paper-based systems because they make availableelectronically the content of what would otherwise be on paper.

Patient Care Charting

Patient care charting is a system in which caregivers enter data (notes, vital signs, med-ications administered, and so on). Early patient care charting systems were focused pri-marily on nursing staff documentation but now have come to be used by any caregiver. Dataentered into patient care charting systems may be structured or unstructured. Figure 1.4illustrates both of these data-entry concepts, as further discussed below.

• Structured data refers to data that have been predefined in a table or checklist,such as shown in figure 1.4 for recording severity of symptoms. The user selects thedata through the keyboard, touch screen, light pen, or any other human–computerinterface via a checklist. In other cases, the user may be asked to enter a numberor numeric score. For example, to record the severity of symptoms, a nurse mayselect a numeric score from 1 to 10 from a standard pain scale. It is important tonote that the purpose of structured data entry is to help ensure standardization. Itis interesting that in healthcare, there are actually two pain scales. One consistsof scores from 1 to 10 and is used by nurses; the other consists of scores from1 to 4 and is used by physical therapists. When setting up structured data entry, itis important to ensure that the appropriate scale, scoring system, or code set isspecified in some way. Structured data capture also may be accommodatedthrough choices in a drop-down menu, such as the selection of a problem in theproblem list in figure 1.4. Additionally, medical devices (such as cardiac moni-tors) or laboratory instruments generate structured data.

• Unstructured data essentially refers to narrative data, such as in the section for“presentation” in figure 1.4. Here, areas are reserved for caregivers to key in,dictate through voice recognition, handwrite through handwriting recognition, oreven dictate and have later transcribed anything they wish to record. Suchunstructured data is more difficult to use in searches and generally is not con-verted into tabular or graphical form.

Early in their introduction, patient care systems required significant changes in workflow. They were often time-consuming and difficult to use. Attempts were made to use

8 Chapter 1

so-called bedside terminals because returning to a small number of terminals at a nurs-ing station was inadequate. Today, patient care charting systems have become muchmore sophisticated. Data entry may be performed using handheld devices, many ofwhich are wireless. Data also may be captured by using smart text (or what would becalled macros in word processing), where only a few key letters have to be entered torepresent an entire word or phrase. The most sophisticated systems would apply naturallanguage processing, which permits narrative text to be converted to discrete data forprocessing by the computer. This last form of data entry is still very much in the devel-opmental stage.

Introduction to Electronic Health Records 9

Figure 1.4. Structured and unstructured data entry in an EHR

Computerized Physician/Provider Order Entry Systems

Computerized physician/provider order entry (CPOE) systems came about fairly recently asa result of the IOM’s quality initiative, which began with the National Roundtable on HealthCare Quality (1998) and subsequent studies on patient safety issues and medication errors inparticular (To Err is Human, 1999; Crossing the Quality Chasm, 2001; and Patient Safety:Achieving a New Standard for Care, 2004). CPOE systems are intended for use by physi-cians and other providers to enter orders directly into the computer system and be givenprompts, reminders, or alerts about the order entered. These systems enhance legibility toavoid errors, and their decision support capability enhances patient safety and healthcare effi-ciencies. Decision support might include calculating an appropriate dose and alerting thephysician as the order is entered that, given the patient’s weight, an alternative dose might beappropriate. Another example might be indicating that a prescribed medication is contra-indicated under certain circumstances, such as when the patient is taking another medicationor being prepped for a certain diagnostic study. The CPOE system might identify that a spe-cific drug is not covered by the patient’s insurance and might offer equivalents.

Some CPOE systems are more sophisticated than others. Some have minimal clinicaldecision support, such as merely checking an order against a formulary. In many of theseless-robust implementations, physicians have complained that the order-entry process takesmuch longer than writing an order on paper and do not believe they are gaining any value.

Clinical Decision Support Systems

The term “clinical decision support system” (CDSS) is generally reserved to describe thehelp provided in association with data entry into an EHR system performed directly by thecaregiver at the point of care. The help may come in the form of alerts or reminders thatare generated by preprogrammed logic, or rules. For example, if a patient has a diagnosisfor which a certain medication has been found to be effective and no order has been givenfor that drug, the system could be programmed to suggest that the physician consideradding the medication to the patient’s treatment regime. This is what supports the mostsophisticated CPOE system. Another example might be support for coding in a physicianoffice EHR system, where the complexity of the patient encounter must be reflected in theEvaluation and Management (E&M) CPT-4 code and ultimately contributes to the levelof reimbursement for the visit.

The help provided through a CDSS also may be in the form of a clinical practiceguideline, care path, or critical pathway. Moreover, the CDSS may tap external knowledgesources to provide more comprehensive information. For example, if a physician is unfa-miliar with a new drug that is offered as an alternative suggestion, he or she may be ableto click on a link to a reference that provides more complete information. A physicianfaced with an unusual set of symptoms and signs may look for reference material on theWeb to develop a differential diagnosis. A pharmacist may need to make suggestions foralternative medications when a patient has an allergy, or research the efficacy of variousdrugs when there is an unusual diagnosis. The CDSS may produce tailored instructions tobe given to the patient. Clinical decision support is considered concurrent (at the time dataare entered) or even prospective (when presenting “best practices” in anticipation of care).

Clinical decision support should be contrasted with executive decision support, whichis typically a stand-alone system that analyzes a large volume of aggregated data and pro-vides trending information. Executive decision support is typically retrospective, providingquality improvement, productivity, staffing, and/or marketing information for executives.

10 Chapter 1

Provider–Patient Portals

Provider–patient portals are the secure means to use the Internet for communicating pro-tected health information (PHI) among providers, between provider and patient, orbetween provider and payer. In some cases, the provider–patient portal is used to supportsecure e-mail. In other cases, it may be used for remote connectivity to results retrievalsystems (and clinical messaging systems). In still other cases, patient monitoring may beperformed through such a connection. Some examples of areas where this is being usedvery effectively is for pacemaker monitoring in heart disease patients, blood sugar moni-toring for diabetic patients, and breath sounds monitoring for babies at risk for suddeninfant death syndrome.

Personal Health RecordsPersonal health records (PHRs) are systems designed to support patient-entered data. Ifthey are associated with a provider, they may serve as a means for patients to access theirown health records (or summaries of their records) or to provide information to their pro-viders about their health status. In some cases, PHRs are integrated with the EHR andaccess is often provided through a Web portal. In other cases, typically when they arestand-alone systems not associated with a provider, they may be anything from a fax-backsystem to one that supports structured data entry by a patient on a Web site.

Population HealthPopulation health also is facilitated through accurate, complete, and timely capture andreporting of public health data, including data relating to homeland security. Populationhealth data collection may be initiated in the provider setting and linked automatically toa state data collection system. Population health may be served by decision support that isprovided to caregivers through alerts from public health departments, such as to notifythem of a new strain of virus or to remind them to seek certain information from patientswho present with certain symptoms. A precursor to population health may be disease man-agement, wherein providers and health plans share data about patients/health plan mem-bers who would benefit from certain educational programs or special monitoring.

EHR Implementation Status

Electronic health record implementation has gone through a number of stages. Initial inter-est in its early history later led to disappointment, but eventually a better sense of reality.

Media HypeThe release of the IOM patient record study report in 1991 initially led to much mediahype. Vendors flooded trade shows and publications with promotions for EHR products.Information systems leadership surveys conducted by HIMSS and other organizationsmade EHR projects the top priority for healthcare institutions.

It was assumed that an EHR product could be purchased and implemented in much thesame way as other systems for keeping track of patient accounts, reporting laboratoryresults, and abstracting health records. Although many of these source systems now havebecome quite complex, they typically are more confined to a single department or a singleset of functions.

Introduction to Electronic Health Records 11

Disappointment

As EHR project teams started evaluating institutional readiness for the EHR and assessingvendors’ products, they found that the project was not as simple as buying software off theshelf or installing departmental systems. They came to understand that the EHR is a highlycomplex concept. Virtually every system in the organization would touch, and be touchedby, the EHR. Many organizations simply had not integrated all the necessary source sys-tems to support the EHR. Furthermore, for an EHR to provide benefit, all caregivers hadto use it in place of the paper health record and use information in new and different ways.Caregivers had never before been touched by any system in the way they would be affectedby an EHR.

Finally, vendor offerings were typically found to be less complete than anticipated.Early in the evolution of the EHR, many vendors developed only an EHR concept andwere looking for organizations to be development partners. Some vendors had one com-ponent of the EHR but could not integrate it with other applications into an organization’stotal operations. Unfortunately, many vendors failed to recognize the complexity of theEHR system. Few vendors took the time to study the nature of clinical information or theflow of clinical data through the healthcare delivery system. Still fewer attempted to intro-duce a system that would truly improve clinical information management, not just dupli-cate the existing paper system. One vendor went so far as to advertise that its productwould make significant improvements in productivity and the quality of care for physi-cians’ offices—without changing a thing! The fundamental lesson learned from these earlyprojects is that improvement requires change.

Reality

Since the IOM released its first study on patient records, hundreds of vendors have beenselling what they have promoted as EHR products. Many of these vendors quickly wentout of business or were sold to save their investment. Today, the lure of the “better mouse-trap” still attracts new vendors, but there is a more stable set of vendors with much morerobust products.

The EHR market is beginning to settle down. Some of the early adopters have workedclosely with vendors to develop mutually acceptable products. Although the market wassegmented by acute care and ambulatory care in early systems, vendors now are acquiringcompanies and/or skills to develop systems that address the needs of integrated deliverynetworks (IDNs). EHR vendors are beginning to work on the key issues that have beenstumbling blocks to success. Today’s environment is much more realistic about what canbe accomplished and what still needs to be done. When organizations undertake an EHRproject today, they have a far greater understanding of what an EHR is and how it must beimplemented. The jury may still be out on whether EHR systems pay for themselvesdirectly, but many organizations are recognizing their strategic value and are planning toimplement systems that address longer-term goals.

Because of the variability of terms used to describe EHR, the multiplicity of visions, andthe migration path of functionality, it is difficult to determine the true status of EHR imple-mentation. Indeed, some believe the success of achieving the IOM vision of an EHR mayhave been undermined by the multiplicity of visions associated with it and the wide varietyof functionality available (Fox and Thierry 2003).

12 Chapter 1

Currently, it is estimated that less than three percent of hospitals have a true EHR. Asurvey of 250 hospitals conducted by the Medical Records Institute in winter 2002 indi-cated that only four facilities had their “entire medical record computerized.” Although thisfigure is probably about accurate, many would attempt to paint a rosier picture. The arti-cle describing this survey was accompanied by graphics describing the current status ofcomputerized records in small, medium, and large hospitals. For example, it indicated thatnearly half (45.2 percent) of the small hospitals surveyed have a computerized record butgoes on to indicate that only a quarter of the computerization in these records was online.Clearly, many hospitals are starting down the path toward an EHR but are finding thatautomated components in themselves are not EHRs.

EHR Limitations

Designing, marketing, and implementing information systems that provide access to clin-ical data and process data into information that contributes to knowledge for improvedquality of care has been challenging. A number of major stumbling blocks are beingaddressed in clinical and technological areas.

Clinical Data Limitations

Clinical data are textual and contextual, but computers have been designed primarily tomanipulate discrete, factual data. Computers are very good at storing large volumes of dataand performing mathematical formulas or clearly defined retrieval functions. However,they do not have the human capability of “thinking” or making associations or assumptionson their own. A lot of work is being done to program computers to perform more “rea-soning” functions and to “learn” to offer decision-making support. But these functions arevery sophisticated and still very much under development.

A good example of clinical data limitations may be to consider how a computer canprocess a simple statement such as “the skin is red.” Interpretation of “red skin” dependson the context to define what is meant by “red.” Does red describe a burned area, a rash,or an increase in temperature? What is the cause—fever, embarrassment, allergy, burn,high blood pressure, or something else? Unless we are satisfied with simply recording thisinformation and making it available as documentary evidence of something caregiversobserved, the field of computer science must learn how to structure data to associate themproperly with other data for future processing.

One significant initiative that is helping the adoption of a standard, clinical vocabularyhas been the licensure of the comprehensive SNOMED vocabulary by the NationalLibrary of Medicine (NLM). Adoption of a standard vocabulary, such as SNOMED, andits use in structured data will open the door for much broader use of clinical decision sup-port systems. (See chapter 8 for a more detailed discussion of SNOMED.)

In addition to data comparability that could be achieved through adoption of a stan-dard, comprehensive vocabulary, clinical practice requires more information today than itdid in the past. First, there is much more to know—thousands of new drugs, new strains ofviruses, and so forth. The field of medicine is continually changing and expanding. Physi-cians are having to track numerous diagnoses, procedures, diagnostic tests, clinicalprocesses, devices, and drugs, and in many cases the payment rules associated with some

Introduction to Electronic Health Records 13

of this. Just keeping up with the literature is daunting. For example, MEDLINE, an onlinebibliographic database of medical information compiled by the NLM, indexes nearly halfa million new articles each year from biomedical literature alone. Connectivity to the Weband its resources has opened a huge opportunity to access information. Now the task is toconvert that information into usable knowledge, which includes evaluating its reliabilityand validity.

Another factor that has presented limitations to clinical use of information systemsincludes the volume of patients and thus productivity concerns surrounding what typicallyhas been a more time-consuming method to record information. It is not uncommon for aprimary care physician to treat eighty patients per day during flu season. To quickly recorddata for this volume of patients, the way in which the caregiver enters or retrieves datafrom the computer (the human–computer interface) needs to be perfected. We are begin-ning to see many small, wireless input devices, including personal digital assistants(PDAs) and tablet computers.

Data reuse also is becoming popular. This refers to the ability to “cut and paste” datafrom templates, previous visit documentation, or even documentation developed for use inother patients’ records. Obviously, this must be applied very carefully and cautiously toensure that the same data directly apply to the current patient and episode of care. How-ever, the function represents an opportunity that is important for adoption of EHRs(Amatayakul, Brandt, and Dougherty 2003).

A final clinical data limitation is that many caregivers find it foreign to rely on infor-mation systems for clinical decision support. But the IOM patient safety reports high-lighting medication errors, and the subsequent efforts of employer groups and majorcorporations to sweeten contracts for those providers using IT to improve patient safety,has led to greater acceptance of clinical decision support systems.

Technological Limitations

Technological limitations have made clinical information systems difficult to use. Even asa new generation of caregivers emerges that is accustomed to using computers and engag-ing patients in their use, physical limitations still abound because healthcare is such amobile profession. The care of patients requires direct interaction between patients andcaregivers. Pen and paper that slip into a pocket are much easier to manage when a care-giver is making rounds and administering to patients. New, smaller, wireless devices, suchas PDAs, notebook computers on carts, and even cellular phones with data capture capa-bility and improvements in voice and handwriting recognition are beginning to addresstechnological limitations. So, too, are efforts to redesign care processes that better incor-porate the use of computers.

Another technological limitation is the extent to which disparate computer systemscan be made to work together and exchange data. Standard protocols have been devel-oped to help, but vendors must adopt the standards and conform to their requirementsexplicitly. In some cases, vendors have developed highly proprietary systems to encour-age providers to buy all components from one vendor. When the vendor does not offer aspecific component, the provider is faced with doing without until the vendor creates thecomponent or buying the component from another vendor and hoping an interface (a spe-cial program to enable data exchange) can be written that will permit the data to flowacross the two different vendor platforms. The lack of interoperability between systems

14 Chapter 1

has sometimes meant that providers have been unable to adopt EHR systems as rapidlyas they would like.

Cost and Value Limitations

A major consideration for any provider adopting EHRs is cost. Today, all healthcareproviders are seeing reduced revenue and increased costs. The Healthcare Financial Man-agement Association (HFMA 2003) reports that providers are often strapped for cash andmany have very limited access to capital. The EHR is considered an investment that mustpay for itself. The systems undoubtedly cost a significant amount of money in addition tothe time required to tailor them to the environment and to manage the degree of changethey create. For example, it may be necessary to create interfaces enabling independentsystems to communicate with one another and with the EHR. All of the application tem-plates must be populated with the provider’s specific requirements, clinical practice guide-lines, any unique terminology, their own formulary and charge data, and many otherrequirements. In addition, although it generally does not take very much time to learn howto document in an EHR if it is properly designed, it does take time for the adoption processto occur and to manage work flow change.

Many have questioned whether the EHR can truly pay for itself. Part of the issue isthat both cost and benefits are somewhat elusive. Because the EHR depends on the inte-gration of all other clinical and administrative systems, sometimes some of this cost isattributed to the EHR when it should be attributed to processes that needed attention any-way. Although there are many direct, monetary benefits, many benefits either cannot bequantified or are very difficult to quantify—even when the effort is made to do so, whichis often not the case.

Some initiatives have attempted to address the cost–benefit issue. For example, theOpen Source EHR project spearheaded by the American Academy of Family PracticePhysicians has attempted to significantly lower the cost of EHRs for physician offices bysharing the source code for development purposes. A number of industry awards that tar-get the “most wired” or “exemplary implementations” have attempted to demonstrate thevalue of EHR systems (for example, “Most Wired” Hospitals and Health Networks, HIMSS).

Standardization Limitations

As alluded to in the preceding descriptions, the lack of standardization—to define the EHR,write interfaces, compare data, ensure data quality, and perform many other functions asso-ciated with EHRs—also has made it very difficult to achieve widespread adoption.

It is not that some standards do not exist. For example, there are standards for writinginterfaces, but not every vendor is required to use them and they contain a high degree ofoptionality. Moreover, there are standard vocabularies, although their number, untilrecently converged into SNOMED, has equated to a Tower of Babel.

Even though every vendor should be able to apply its own “bells and whistles” toenhance and distinguish its products, standardization would at least achieve a baselineproduct expectation. The mark of a mature industry is when it can adopt standards thatachieve baseline functionality making the products it uses commodities. For example, theautomotive industry has established a standard construct for what a car is. Yet, everyonealso knows the difference between a Chevy and a Cadillac. The EHR market needs to cre-ate a car to which vendors can apply more or less leather and chrome.

Introduction to Electronic Health Records 15

Change Limitations

Also alluded to in the above descriptions of limitations is the underlying issue of thedegree of change imposed by EHR systems. Although somewhat dependent on the com-puter skills of healthcare professionals, the immensity of change begins with learning howto use a computer. Many healthcare professionals today still do not routinely use a com-puter at home or at work, and need basic computer skills. Work flows and processes inhealthcare also represent enormous obstacles. Despite the fact that new procedures, tests,and drugs are constantly being developed, health professionals have a very ingrained senseof process. In fact, such habits enable them to react quickly to rapidly changing circum-stances. It is extremely difficult to change these processes to accommodate what manyhealthcare professionals still view as “only” documentation rather than information that isa source of knowledge and value.

In fact, the extent to which the health profession rejects process change has resulted inmany EHR product design efforts attempting to replicate the paper environment. This canbe seen in screens that have “tabs,” the volume of printouts generated in a paperless envi-ronment, and, essentially, the rejection of performing work that is typically viewed as cler-ical (such as order entry). When considered from a cost–benefit perspective, is it anywonder that executives question the value of a system that does nothing but automatetoday’s environment?

The issue of change is something of a catch-22 in healthcare. The EHR should intro-duce sufficient change so as to improve quality, cost, and access to healthcare, but still rea-sonably reflect processes essential to healthcare delivery.

Major Initiatives in Overcoming EHR Issues

The EHR projects the industry is now undertaking are based on the understanding that anEHR is not a turnkey product but, rather, is as much a concept as a set of interrelated sys-tems. It is further recognized that technology alone is not the answer. It is possible toimplement a well-designed EHR, but if the persons expected to use it are not engaged inselecting the system, do not aid in its design, are not properly trained, or are not supportedby executive leadership, the system will likely either not be used or not be used effectivelyand thus fail to produce the anticipated value.

Creation of EHR systems depends on a thorough understanding of clinical data andhow health professionals use information in clinical decision making. There must be sup-port for adoption of data standards and data quality measures. The willingness to collabo-rate and share data, both among providers and between providers and patients, may beamong the most difficult of the nontechnical issues to be addressed.

Private Efforts

Many provider organizations have made major strides in achieving the vision of the EHR.In some hospitals, 100 percent of the physicians use CPOE. Many hospitals have seen pro-ductivity increases from document imaging systems. Clinicians who have access toscanned documents or information through clinical messaging “will never return to paper.”Home health agencies have benefited significantly from adoption of handheld devices tocapture and submit data, either via telephone modem, docking to a PC, or wireless at a

16 Chapter 1

base station. Ambulance services are using not only telemetry, but also smart phones andother devices to quickly record their documentation. Each step along the migration pathprovides value. It should not be seen as necessary to wait to implement a comprehensivesystem in order to achieve benefits.

Federal Government Efforts

The federal government also has made EHR an initiative. The Department of Defense, theVeterans Administration, and the Indian Health Service are all focused on adopting someform of EHR. The National Committee on Vital and Health Statistics (NCVHS) has envi-sioned a National Health Information Infrastructure (NHII) in which health informationwould be available across the continuum of care for patient treatment as well as popula-tion health. The NCVHS also is responsible for overseeing implementation of the privacy,security, and transactions provisions of the Health Insurance Portability and Accountabil-ity Act of 1996 (HIPAA) and for developing recommendations for uniform data standardsfor the electronic exchange of patient medical record information (PMRI). NCVHS rec-ommendations relative to data comparability were a significant driver in government licen-sure of SNOMED. Additionally, NCVHS has made recommendations relative to systeminteroperability, hoping to gain voluntary adoption of the latest versions of standards thatwould help interface disparate systems.

In 2003, the Secretary of the Department of Health and Human Services requested theIOM and HL7 to design a functional model and standard for the EHR. (See appendix B.) Inresponse, senior executives and volunteer leaders from eight health-related professional andtrade associations joined forces as the EHR Collaborative in July 2003 to further support thefederal government’s effort to advance toward the EHR. Members of the collaborative includethe AHIMA, the American Medical Association (AMA), the American Medical InformaticsAssociation (AMIA), the College of Healthcare Management Executives (CHIME), theeHealth Initiative (eHI), the Healthcare Information and Management Systems Society(HIMSS), and the National Alliance for Health Information Technology (NAHIT).

Private-Sector Organizational Efforts

In addition to the members of the EHR Collaborative, many other organizations have madeEHR an initiative. As mentioned earlier, the CPRI was created based on the IOM patientrecord study; its work has since been transferred to HIMSS. The AHIMA was a contribu-tor to the original IOM patient record study and has steadfastly taken many initiatives tosupport EHR and the use of health information management (HIM) professionals to sup-port them. The Markle Foundation is another organization that contributed a project onConnecting for Health that carries some of the original CPRI work forward and promotesadoption of clinical data exchange standards, privacy and security, and personal healthrecords. The Leapfrog Group is an organization of more than 150 primarily large compa-nies that provide healthcare benefits to their employees. It promotes contractual incentivesfor hospitals that take specific steps to reduce harm to patients, most notably via CPOE.The eHealth Initiative is another group of healthcare stakeholders whose mission is todrive improvement in the quality, safety, and efficiency of healthcare through informationand information technology.

Introduction to Electronic Health Records 17

Vendor Efforts

EHR vendors also have contributed significantly to research and development of EHR sys-tems. Much of the existing standards development has come about through the dedicationof vendor staff to the effort. Each product success and failure contains lessons learned. Inaddition, vendors have taken different approaches—some highly proprietary and othersvery open. This has provided the industry an opportunity to select applications that meettheir specific needs.

EHR Implementation Stages

Much work remains to be done to achieve the fullest possible vision of the EHR. However,the vision expands as each new system is developed and implemented, and will continueto expand as new technology comes into being. Organizations desiring the quantitative andqualitative benefits of the EHR can choose from a variety of options. Historically, “buildyour own” was the mantra of many, but this mantra is rapidly being replaced with eitherbuy or “borrow.” Most providers today buy a commercial off-the-shelf (COTS) product,which is much more affordable and yet still highly customizable. Others acquire EHRfunctionality through a service, known as an application service provider (ASP), thatmaintains the system and leases usage. Whatever model is chosen to acquire an EHR,strategic planning and a carefully developed migration path need to be constructed. A sum-mary of the implementation stages is provided in figure 1.5; the remaining chapters coverall aspects of the various steps within each stage.

The stages are discussed briefly as follows:

1. Determine readiness for an EHR. Readiness for an EHR may need to be culti-vated if there is resistance to computer use. Determine the readiness state andwhat needs to be done to ensure adoption by all. A key ingredient is user partici-pation. Time and again, systems have failed to be adopted appropriately or yieldtheir true results because the individuals required to use the system were notinvolved from the start.

2. Plan the migration path based on a shared vision of the ultimate goal, and thenplot a migration strategy to achieve that goal. There may not be a quick fix orinstantaneous solution. Although many might like a “big bang” that goes from min-imal automation to the vision of the future, it is difficult to pull off in the best ofenvironments. EHR achievement will more likely be an evolution rather than arevolution, although very recently some provider settings have been successful atachieving a revolutionary approach.

3. Select the EHR system that is right for the organization. Many providers arefirmly entrenched with one major vendor, even though the vendor may not bethe best for EHR. It may be necessary to compromise and extend the migrationpath until the vendor catches up, compromise the functionality (especially thecustomizability) to achieve the best integration with the incumbent vendor, or“bite the bullet” and switch vendors. The organization should conduct a thor-ough cost–benefit analysis to look at the overall picture.

18 Chapter 1

4. Install, design, and test the system. Plan ahead to anticipate the impact on users,work flow, productivity, and patients. There are many issues, from whether andhow implementation will be phased in to how past data will be integrated andold data retained.

5. Train, train, and train. User involvement from the beginning is critical to suc-cess. So, too, is training. It is not unusual to have more trainers than staff on-siteduring the actual rollout of an EHR system. Plan for this, advise staff, and workwith your vendor to ensure a smooth transition.

6. Determine benefits, correct course if needed, and enhance the system. A benefitsrealization study is key to determining the payback. If this is not what wasexpected, it is a signal for course correction. When benefits are achieved, theresults can justify further enhancements.

Conclusion

Planning and implementing an EHR system is a significant undertaking for any healthcareorganization. Although the concept of the EHR is not new, the industry is now just beginning

Introduction to Electronic Health Records 19

Figure 1.5. EHR implementation stages

Det

erm

ine

Ben

efits

—C

orre

ct C

ours

e—E

nhan

ce

DetermineReadiness

PlanMigration

Path

SelectSystem

Train

InstallDesign

Test

to fully appreciate the complexity of the integration required to achieve a comprehensiveEHR. Indeed, many of the means to carry out the objectives of the EHR are still emerg-ing. Major stumbling blocks in the form of content, value of information, and technolog-ical limitations are just beginning to be addressed. But there is no time like the present.Because of the rapid rate of change in information technology, putting off electronic healthrecord implementation until “it becomes more affordable,” “more proven,” or “more accept-able” only puts off a new and different set of issues.

References and ResourcesAmerican Health Information Management Association. 2003. A vision from the e-HIM future. Journal ofthe American Health Information Management Association 74(8): Suppl.

Amatayakul, M., M. Brandt, and M. Dougherty. 2000. Cut, copy, paste: EHR guidelines. Journal of theAmerican Health Information Management Association 74(9): 72, 74.

Aspden, P., J. M. Corrigan, J. Wolcott, and S. M. Erickson, eds. 2004. Patient Safety: Achieving a NewStandard for Care. Washington, D.C.: National Academies Press.

Collen, M. F. 1995. A History of Medical Informatics in the U.S. 1950–1990. Bethesda, Md.: HartmanPublishing.

Computer-based Patient Record Institute. 1995–2004. Annual Nicholas E. Davies Award Proceedings of theCPR Recognition Symposium. Chicago: Healthcare Information Management and Systems Society.

Dickinson, G., L. Fischetti, and S. Heard, eds. 2003. HL7 EHR System Functional Model and Standard,Draft Standard for Trial Use, Release 1.0. and EHR Collaborative Report of Public Response to HL7 Ballot1 EHR, August 29.

eHealth Initiative. 2004. Who We Are. Available at www.ehealthinitiative.org.

EHR Collaborative. 2004. Who We Are. Available at www.ehrcollaborative.org.

Fox, L. A., and P. Thierry. 2003. Multiple visions undermine success of EHRs. ADVANCE for HealthInformation Management 13(14): 8–9.

Healthcare Financial Management Association, in partnership with GE Healthcare Financial Services. 2003.How are Hospitals Financing the Future? Access to Capital in Health Care Today. Westchester, Ill.: HFMA.

Health Level Seven. 2004 (April). HL7 EHR Functional Descriptors, Draft Standard for Trial Use.

Healthcare Information Management and Systems Society. 1995–2004. The Annual Nicholas E. DaviesAward of Excellence. Available at www.himss.org.

Healthcare Information Management and Systems Society. 1990–1999. The Hewlett-Packard/HIMSSLeadership Survey. Chicago: HIMSS.

Hospitals and Health Networks. 2002 Most Wired. Available at www.hospitalconnect.com.

Hughes, G. 2003 (June). Practice Brief: Transfer of patient health information across the continuum (Updated).Journal of the American Health Information Management Association. Available at www.ahima.org.

Institute of Medicine. 1991. The Computer-based Patient Record: An Essential Technology for Health Care.Edited by R. S. Dick and E. B. Steen. Washington, D.C.: National Academies Press.

Institute of Medicine. 1997. The Computer-based Patient Record: An Essential Technology for Health Care.Edited by R. S. Dick, E. B. Steen, and D. E. Detmer. Washington, D.C.: National Academies Press.

20 Chapter 1

Institute of Medicine. 2000. To Err Is Human: Building a Safer Health System. Edited by Kohn, L.T.,J. M. Corrigan, and M. S. Donaldson. Washington, D.C.: National Academies Press.

Institute of Medicine. 2003 (July 31). Letter Report: Key capabilities of an electronic health record system.Washington, D.C.: National Academies Press. Available at http://www.nap.edu/books.

Kibbe, D. 2003 (September 25). Open source electronic health record for office-based medical practice.American Academy of Family Physicians. Available at www.aafp.org.

Medical Record Institute. 2002. Less than 3% of hospitals surveyed have completely computerized records.Medical Records Briefing (winter).

Mon, D. T. 2004. Setting the right expectations for the EHR standard. Journal of the American HealthInformation Management Association 75(3): 52–53.

National Committee on Vital and Health Statistics. 2001. Information for Health: A Strategy for Building theNational Health Information Infrastructure. Washington, D.C.: NCVHS.

National Committee on Vital and Health Statistics. 2002 (February 27). Letter to Secretary Thompson,U.S. Department of Health and Human Services, on Recommendations on Uniform Data Standards forPatient Medical Record Information.

National Library of Medicine. 2003. Fact Sheet: Integrated advanced information management systems(IAIMS) grants. Available at www.nlm.nih.gov/pubs/factsheet/iaims.html.

Rhodes, H., and G. Hughes. 2003. Practice Brief: Redisclosure of patient health information (Updated).Journal of the American Health Information Management Association 74(4): 56A–C.

Shortliffe, E. H., and L. E. Perreault, eds. 2001. Medical Informatics: Computer Applications in Health Careand Biomedicine, 2nd ed. New York: Springer-Verlag.

The Leapfrog Group for Patient Safety. 2004. About Us. Available at www.leapfroggroup.org.

Introduction to Electronic Health Records 21