Footnote 6 Publication

P u r su i t

O f

Q uali t y

Physicians’ Use Of Electronic
Medical Records: Barriers And
Solutions
A survey of physician practices shows slow but steady progress in
adopting this new technology.
by Robert H. Miller and Ida Sim
ABSTRACT: The electronic medical record (EMR) is an enabling technology that allows physician practices to pursue more powerful quality improvement programs than is possible
with paper-based records. However, achieving quality improvement through EMR use is neither low-cost nor easy. Based on a qualitative study of physician practices that had implemented an EMR, we found that quality improvement depends heavily on physicians’ use of
the EMR—and not paper—for most of their daily tasks. We identified key barriers to physicians’ use of EMRs. We then suggest policy interventions to overcome these barriers, including providing work/practice support systems, improving electronic clinical data exchange, and providing financial rewards for quality improvement.

I

n f o r m at i o n t e c h n o l o gy i s i n c r e a s i n g ly r e c o g n i z e d as an important tool for improving patient safety and quality of care, especially by promoting the practice of evidence-based medicine.1 Of all the health information
technology (IT) in current use, the electronic medical record (EMR) has the most
wide-ranging capabilities and thus the greatest potential for improving quality.
Research has demonstrated the quality benefits of electronic documentation and
viewing, prescription and test ordering, care management reminders, and
messaging, among other EMR functions.2
Despite this potential for quality improvement, however, few physician practices use EMRs. Nevertheless, interest in EMRs is substantial.3 In a survey of 1,200
mostly solo/small physician groups, less than 13 percent of respondents said that
their practice had EMRs; 32 percent expressed interest in EMRs, and half of these
were “very interested.”4 Clearly, the EMR is of growing importance for many physician practices.
In this paper we present key results of our qualitative study on the opportuniRobert Miller ([email protected]) is associate professor of health economics in residence at the Institute for
Health and Aging, University of California, San Francisco (UCSF). Ida Sim is an assistant professor of medicine
in the Division of General Internal Medicine, Department of Medicine, and associate director for medical
informatics, Program in Biological and Medical Informatics, at UCSF.

116
DOI 10.1377/hlthaff.23.2.116 ©2004 Project HOPE–The People-to-People Health Foundation, Inc.

M a r c h /A p r i l 2 0 0 4

Use

Of

E MRs

ties and barriers that ambulatory physician practices encounter when using
EMRs for quality improvement. Based on insights from this study, we then suggest policy interventions that can promote opportunities for and lower barriers to
using EMRs for quality improvement.

Methods
We conducted nearly ninety interviews between mid-2000 and the end of 2002
with EMR managers and physician champions in thirty physician organizations
that had implemented an EMR. The organizations were purposefully selected for
their diversity in sizes, EMR used, duration of EMR use, affiliations, and extent of
capitated payment. Studied organizations included nine large medical groups of
more than seventy physicians each, eighteen solo/small-group practices of ten or
fewer physicians, and three medium-size groups. Most of the small groups were
primary care only, while eight of the large groups were multispecialty; we focused
primarily on EMR use by primary care physicians. Some data were obtained from
the same organization at several points in time.
The physicians and managers that we interviewed were likely more positive
than their colleagues about the promise of EMRs, and practices participating in
the study were likely more successful than nonparticipating practices. Nevertheless, as the findings suggest, interviewees tended to be open about the challenges
of using EMRs. We also interviewed representatives from EMR vendors, professional medical associations, and IT consulting firms.
The interview questionnaire was semistructured, allowing interviewees to respond in their own words. We taped the interviews and coded the transcripts using qualitative research software (QSR Nvivo), to organize and summarize data by
concept (for example, EMR costs, financial benefits, and market environment).
Our analysis methods included pattern-matching and explanation-building techniques often used in qualitative research.5 From the analyzed interviews, we identified persistent and important themes and patterns. We then constructed a
typology of barriers to adopting EMRs for quality improvement.

Study Results
n

EMR use and potential benefits. An EMR can provide the electronic infrastructure for eight types of clinical and administrative activities normally conducted in physician practices. Commercial EMR systems vary in their capabilities
for each type of activity, while practices and physicians vary greatly in how extensively they use available EMR capabilities.
Here we summarize our findings on how physician practices used and generated clinical benefits from each type of EMR capability and on the major barriers
these practices encountered. Our key insight with policy implications is that the
path to quality improvement and financial benefits lies in getting the greatest
number of physicians to use the EMR (and not paper) for as many of their daily

H E A L T H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 2

117

P u r su i t

O f

Q uali t y

tasks as possible. The key obstacle in this path to quality is the extra time it takes
physicians to learn to use the EMR effectively for their daily tasks.
Viewing. All practices used EMR viewing capabilities, which improve chart
availability, data organization, and legibility. Quality benefits depended on the
amount of viewable clinical data. The amount of initially viewable data depended
on efforts to type in existing paper-based medical record data and to electronically import data from lab, billing, and other systems. As patient data accumulated
over time, financial savings accrued from less staff time spent finding, pulling, and
filing charts and less physician time spent locating information.
Documentation and care management. We identified a consistent relationship between greater electronic documentation by physicians and greater quality improvement and financial benefits. Just as computerized physician order entry
(CPOE) appears central to generating benefits in hospital settings, physician electronic documentation appears central to generating benefits in ambulatory care
settings.
Although most clinicians maintained electronic problem and allergy lists, physicians varied greatly in how they documented progress notes. Basic EMR users
had their dictated notes transcribed and imported into the EMR, or they typed
their own progress notes into unstructured text boxes. More advanced users
typed data into templates (electronic forms) that included physical exam and documentation prompts.
Basic use of the EMR improved the legibility and accessibility of progress notes
and increased the availability of electronic problem and allergy lists. More advanced use of documentation templates led to greater opportunities for improving
quality of care. For example, problem-specific templates (such as a sore throat
template) with embedded prompts reminded clinicians to ask about particular
symptoms, order particular tests and prescriptions, or perform preventive or disease management activities. Also, templates that help clinicians enter data in
coded rather than free-text form facilitated more advanced computer-based decision support for such tasks as care coordination and chronic disease management.
In most practices the bulk of EMR-related financial benefits came from reductions in medical records and transcription staff as physicians moved from dictation to typing their own notes. Some practices increased billing revenue through
more complete capture of services and decreased undercoding of services provided.
Ordering. Basic use of electronic ordering typically consisted of physicians’ typing in prescription orders, responding to drug interactions and drug allergy alerts,
and printing out prescriptions. All but three practices we studied used electronic
prescribing. In large practices, basic ordering often also included electronic ordering of referrals and laboratory and radiology tests. More advanced ordering capabilities included additional decision support, electronic transmission of orders to
pharmacies and laboratories, and better tracking of test-order status and test re-

118

M a r c h /A p r i l 2 0 0 4

Use

Of

E MRs

“Designing easy-to-use software for knowledge workers is a
challenge that spans the software industry beyond health care.”
sults, all of which can improve quality and decrease errors.
Messaging. Basic use of electronic messaging among providers improved the
availability, timeliness, and accuracy of messages and increased completeness of
documentation, thus potentially reducing “dropped balls” and safety problems.
Much less common was advanced messaging, which included messaging with
outside providers (to improve care coordination) and with patients (to improve
patient satisfaction and, potentially, patient self-care and compliance).
Analysis and reporting. Few practices initially used physician performance monitoring and feedback capabilities to improve quality and efficiency. Over time,
some practices—especially larger ones—used reporting capabilities more widely:
For example, some practices generated reports to physicians on diabetic patients
with hemoglobin A1C levels greater than 8 percent and on the percentage of a physician’s patients having such levels.
Patient-directed functionality. Most practices had limited or nonexistent practice
Web sites for patients. A few large-practice Web sites enabled patients to schedule visits, send secure e-mail messages to providers, receive e-mail reminders, order medications, access their charts, and obtain more individualized educational
patient care information—all of which have the potential to improve quality.6
Billing. Increased integration between billing and EMR software, combined
with electronic documentation, can yield financial benefits through more complete capture of services provided, more defensible Medicare coding at higher
coding levels, and reductions in data-entry staff.
n Barriers to EMR use. Key surface barriers to EMR use that emerged as persistent themes from our interview data included high initial financial costs, slow and
uncertain financial payoffs, and high initial physician time costs. Underlying barriers included difficulties with technology, complementary changes and support,
electronic data exchange, financial incentives, and physicians’ attitudes. These barriers were most acute for physicians in solo/small-group practice, a mode in which a
substantial majority of U.S. physicians practice.7
High initial cost and uncertain financial benefits. The high up-front financial costs of
implementing EMRs is a primary barrier to their adoption. This barrier is compounded by uncertainty over the size of any financial benefits that may accrue
over time. In most practices we studied, up-front costs ranged from $16,000 to
$36,000 per physician. Some practices incurred additional costs (in the form of decreased revenue) from seeing fewer patients during the EMR transition period.8
Financial benefits varied greatly, from none in practices that made few work practice changes and retained paper processes to more than $20,000 per physician per
year in the few practices that eliminated most paper processes.

H E A L T H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 2

119

P u r su i t

O f

Q uali t y

High initial physician time costs. Interviewees reported that most physicians using
EMRs spent more time per patient for a period of months or even years after EMR
implementation. The increased time costs resulted in longer workdays or fewer
patients seen, or both, during that initial period.
Three underlying barriers—difficulties with technology, complementary
changes and support, and electronic data exchange—increased physicians’ initial
time costs and reduced physicians’ EMR use, financial benefits, and quality improvement.
Technology. Most respondents or their colleagues considered even highly regarded, industry-leading EMRs to be challenging to use because of the multiplicity of screens, options, and navigational aids. Problems with EMR usability—especially for documenting progress notes—caused physicians to spend extra work
time to learn effective ways to use the EMR. These substantial initial time costs
are an important barrier to obtaining benefits, as greater burdens on physicians’
time decrease their use of EMRs, which lowers the potential for achieving quality
improvement.
Although vendors are slowly improving EMR usability, most vendor interviewees doubted that any “silver bullet” technology (for example, voice recognition,
tablet computers, or mobile computing) will dramatically simplify EMR usage.
Designing easy-to-use software for knowledge workers is a challenge that spans
the software industry beyond health care.9
Difficult complementary changes and inadequate support. EMR hardware and software
cannot simply be used “out of the box.” Instead, physician practices must carry out
many complex, costly, and time-consuming activities to “complement” the EMR
product. Across industries, such complementary changes have been found to be
critical for generating benefits from new technology.10
These complementary changes exact a great deal of time from physicians—especially physician EMR champions in solo/small-group practices—for months or
even years after implementation. In our study, EMR champions in small practices
spent much time arranging for EMR installation, receiving and assisting with
EMR training, and encouraging EMR use among their colleagues and staff. These
physician champions also had to patch together and deploy technical support
from the various software, hardware, networking, and service vendors when technical glitches occurred. Both champion and nonchampion physicians had to work
with their staffs to summarize and enter patient data from existing paper charts
into the EMR. All physicians spent substantial time customizing their own visitor disease-specific electronic forms and documentation shortcuts to speed visit
documentation. Moreover, physicians had to redesign their workflow (how they
worked in the exam room) and office workflow (who did what tasks).
As a general rule, larger physician groups could implement complementary
changes more easily than smaller groups could because large groups tended to
have stronger organizational resources such as management expertise, experience

120

M a r c h /A p r i l 2 0 0 4

Use

Of

E MRs

“Practices without physician EMR champions may flounder in
their efforts to generate quality or financial benefits from EMRs.”
with past process changes, financial resources, leadership, and information systems support staff.11 As a result, large groups provided more internal technical and
personnel support for complementary changes. Despite these advantages, interviewees from large groups reported that many of their physicians still had to invest
substantial additional time to make needed changes. For solo/small-group practices that lacked much internal support, physicians bore a much greater time burden after initial EMR implementation.
Inadequate electronic data exchange. Another barrier to EMR use was the lack of adequate electronic data exchange between the EMR and other clinical data systems
(such as lab, radiology, and referral systems). Having parallel electronic and paperbased systems forced physicians to switch between systems, thereby slowing
workflow, requiring more time to manually enter data from external systems, and
increasing physicians’ resistance to EMR use. Furthermore, with fewer data in the
EMR, there was less opportunity for intervening electronically to improve quality,
and reduced ability to perform internal analyses or to report performance externally for quality report cards or performance incentive programs.12
Lack of electronic data exchange was most problematic for solo/small-group
physicians. For example, physicians in nine of the eighteen solo/small-group practices we studied could not view any electronic lab results within their EMR, seventeen could not view hospital data, and nine had EMRs that could not exchange
any data with their practice management system. Some labs or hospitals refused to
set up data exchange; less often, the practice failed to make necessary programming changes in its own EMR because vendor or internal IT support was lacking.
In contrast, larger groups tended to have in-house lab and practice management
systems that exchanged data with their EMRs, and had the leverage to obtain the
cooperation of hospitals and other external data producers for electronic data exchange. Large groups also had the IT staff to program any necessary data exchange
interfaces.
Lack of incentives. EMR use could be increased through financial rewards for
quality improvement and for public reporting of multiple measures of quality performance. Yet few interviewees reported any financial incentives for quality, and
none reported public reporting of their quality performance compared with that
of other physician practices. The lone practice that operated under substantial financial incentives for quality improvement intensified its use of the EMR and
reaped sizable financial rewards.
Physicians’ attitudes. Most interviewees were EMR champions who had positive,
“can-do” attitudes toward solving EMR-related problems and who were vital to
getting other physicians to use EMRs. These physicians—innovation “early adopt-

H E A L T H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 2

121

P u r su i t

O f

Q uali t y

ers”—were willing to bear initial financial and time costs to generate benefits.13 In
contrast, nonchampion physicians tended to be less positive toward EMRs and
more easily discouraged by usability problems. Without exhortation and support
from physician champions, these physicians tended to remain as lower-level EMR
users. As a result, practices without physician EMR champions may flounder in
their efforts to generate quality or financial benefits from EMRs.
In summary, the greatest financial and quality benefits are to be had when most
physicians use EMR capabilities for most of their daily tasks. However, especially
in smaller practices, the initial physician time burden can be onerous given current obstacles, resulting in piecemeal and less effective use of EMR capabilities
and fewer financial and quality benefits. Even in large groups, despite the availability of better organizational support and more complete data exchange, many
physicians used only a fraction of available EMR capabilities and generated only a
fraction of potential quality and financial benefits.
In the face of these barriers to using EMRs for quality improvement, one hopeful theme that emerged is that improvements in electronic data exchange, quality
performance incentives, and support can over time help transform many physicians from inefficient, low-level EMR users to efficient, advanced users, thereby
increasing the likelihood of generating quality and financial benefits. Our policy
suggestions below center on facilitating such improvements.

Potential Solutions
We believe that public and private policy interventions can effectively counter
several of the barriers we identified to ambulatory EMR adoption and use for
quality improvement. Indeed, such interventions are essential for solo/smallgroup practices that lack the organizational resources of large groups. Each policy
intervention suggested below either decreases EMR time costs to physicians or
increases financial benefits, thereby increasing the attractiveness of EMR use to
physicians.
n Communitywide data exchange. Ubiquitous, secure electronic exchange of
clinical data among providers would help lessen the disruption from parallel electronic and paper-based medical record systems, thereby decreasing physician time
costs and increasing financial benefits.
Market developments are increasing the amount of data exchange to some extent. Some hospitals and laboratories are proffering electronic data exchange as a
marketing tool to gain the allegiance of small EMR-using practices. RxHub and
SureScripts—consortia of pharmacy benefit managers (PBMs) and pharmacies,
respectively—are competing to provide electronic prescribing connectivity
among physicians, pharmacies, and PBMs.14 Moreover, several market and regulatory developments are setting the stage for even greater future data exchange:
Data-producing organizations are increasingly using data exchange standards
such as HL7 and LOINC; implementation of the Health Insurance Portability and

122

M a r c h /A p r i l 2 0 0 4

Use

Of

E MRs

Accountability Act (HIPAA) provisions has cleared the way for vendors to provide
private and secure data exchange in regulatory compliance; and substantial efforts
to implement common safeguards for privacy and security have already been
made.15
To take advantage of these developments and to stimulate additional improvements in clinical data exchange, policies should hasten the creation of communitywide data exchange systems that allow clinicians to view all of their patients’
data, regardless of provider and care site. These systems can build upon data exchange models being developed in Santa Barbara, Seattle, Indianapolis, and elsewhere, which are addressing technology, governance, business model, and privacy
issues.16 A 2002 Institute of Medicine (IOM) report called for government to fund
large-scale demonstration projects for such communitywide data exchange efforts, preceded by the creation of local data exchange governance structures and
followed by EMR-focused demonstrations.17
Government agencies, foundations, and purchasers—especially employers and
the Centers for Medicare and Medicaid Services (CMS)—could play an important role in convening the numerous relevant health care entities to begin creating
plans and governance structures for communitywide data exchange. A prime beneficiary of these initiatives will be solo/small-group practices that otherwise lack
the leverage and resources to establish ubiquitous data exchange.
n Performance incentives and mandates. Financial payback to practices for
achieving quality improvement or mandates for IT use would also increase the adoption and use of EMRs for quality improvement. In a highly positive development, a
small but growing number of purchasers, health plans, and employers are initiating
quality-based reimbursement programs, rewarding practices for publishing performance reports, mandating specific quality improvement actions or use of specific IT
applications, and even rewarding consumers for choosing higher-quality providers
on the basis of these performance reports.18 For example, in the California “pay-forperformance” initiative that started in 2003, health plans are measuring and rewarding some aspects of quality, patient satisfaction, and IT use in ambulatory care.19
Adoption of pay-for-performance programs by both Medicare and private payers
could have a powerful effect on accelerating EMR adoption and use.
The specific design of Medicare or private performance-incentive or -mandate
programs can directly affect the nature and pace of IT adoption. For example, incentive programs that reward performance gains on multiple clinical indices encourage the use of full-fledged EMRs to both achieve and document widespread
gains. Alternatively, programs may selectively promote particular IT capabilities,
such as the Leapfrog Group’s mandate for hospital electronic order entry.20 Many
questions remain about how performance-incentive programs should be designed
to encourage EMR use for quality improvement, including questions about the
type and size of incentives that are needed to induce physicians to improve quality
of care or to entice non-EMR-using practices to adopt EMRs.

H E A L T H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 2

123

P u r su i t

O f

Q uali t y

Even as incentive and mandate programs influence IT uptake, however, these
programs should avoid micromanaging how clinical change must occur. For example, mandating “pop-up” reminders may work in some practices but backfire in
other practices where such reminders are considered intrusive. Also, these programs should avoid setting arbitrarily short deadlines for full EMR use, which
may be counterproductive for some physicians. Rather, full EMR use might be
better achieved via a graduated approach in which only basic EMR use (for example, viewing, problem list management, and prescription ordering) is expected
initially, with use of advanced EMR capabilities such as decision support and care
management expected only after basic use has been solidified. Practices may respond to such graded expectations by focusing their early efforts on demonstrating quality and financial benefits from basic use by EMR-receptive physicians or
sites, which could then encourage greater and more advanced EMR use even
among the practices’ less EMR-receptive physicians.
Government or other funders may not need to directly subsidize the cost of acquiring EMRs, since our study suggests that most practices can secure capital for
purchasing the technology. Policy funds could be better used for rewarding quality improvement, for example, than for replacing available sources of capital. Furthermore, the nascent application service provider (ASP) model promises to offer
EMRs at lower initial cost: ASPs are firms that remotely host, support, and essentially lease EMR software and hardware.21
n Support for complementary changes. Physician practices, especially solo/
small-group practices, require support to carry out the time-consuming workflow
and other complementary changes needed to generate financial and quality benefits
from “out-of-the-box” EMRs. Both established EMR vendors and current ASP
firms, however, tend to provide only technical support for their products.
From a policy perspective, providing support for complementary changes is
crucial but challenging and will require much experimentation. We suggest that
funding agencies support the demonstration and evaluation of various models for
providing comprehensive EMR support services to solo/small-group practices to
catalyze the development of such services in the market. In addition, funders
should support comparative evaluations of existing support services, to intensify
competitive pressure for their development.
n Information for those using and considering EMRs. Across industries, innovations such as IT diffuse more slowly the greater the uncertainty about the costs,
implementation, use, and consequences of the technology.22 For EMRs, uncertainty
is especially high because of the relative dearth of objective data on costs, use, and
comparative features (such as a Consumer Reports–type rating of EMRs). Government
and private funding agencies could fund both comprehensive product comparisons
of EMRs and research characterizing the range of financial, time, and quality outcomes experienced by various types of EMR-using practices.
Demonstration and evaluation projects can help determine how best to encour-

124

M a r c h /A p r i l 2 0 0 4

Use

Of

E MRs

age physicians to adopt and use EMRs to improve quality of care. Such projects
can help provide answers about the ideal design of pay-for-performance incentives, for example. The most instructive demonstrations might be those that evaluate combinations of the policy interventions suggested above: communitywide
electronic data exchange, enhanced incentives for quality, and comprehensive
support services.
We believe that funding agencies and policymakers should generally avoid
funding initiatives that aim to directly lower technology or physician-attitude
barriers to EMR use. Vendors already have strong market incentives for developing easy-to-use EMRs, while physicians’ attitudes toward health IT and the EMR
are already on a favorable trajectory with increasing use of personal digital assistants (PDAs) and the Internet by physicians.23 Both technology development and
physicians’ attitudes are unlikely to be changed by policies separate from those
discussed above.

T

h e e m r i s a n e na b l i n g t e c h n o l o gy for physician practices to pursue
quality improvement in potentially powerful ways. Our research finds,
however, that systematic quality improvement using EMRs is neither lowcost nor easy. There is no simple solution to accelerating EMR adoption and use
for quality improvement. Given the multifaceted nature of the barriers, a range of
policy interventions is needed to spur successful EMR-driven quality improvement. These policy interventions center on improving data exchange among
health care entities, providing financial rewards for quality improvement, and providing work/practice support. This package of policy interventions, in combination with ongoing trends, should hasten adoption of EMRs and their use for quality improvement in ambulatory care.

The authors received funding for this research from the Robert Wood Johnson Foundation’s Changes in Health Care
Financing and Organization (HCFO) initiative and from the California HealthCare Foundation.
NOTES
1.
2.

3.

4.
5.

Institute of Medicine, Crossing the Quality Chasm: A New Health System for the Twenty-first Century (Washington:
National Academies Press, 2001).
Ibid; E.A. Balas, “The Clinical Value of Computerized Information Services: A Review of Ninety-eight
Randomized Clinical Trials,” Archives of Family Medicine 5, no. 5 (1996): 271–278; and D.L. Hunt, “Effects of
Computer-Based Clinical Decision Support Systems on Physician Performance and Patient Outcomes,”
Journal of the American Medical Association 280, no. 15 (1998): 1339–1346.
K.B. Lynam and V.J. Karlan, “Electronic Medical Record Systems: Trends in Large Group Practices,” Group
Practice Journal 51, no. 3 (2002): 18–24; Harris Interactive, “eHealth’s Influence Continues to Grow as Usage
of the Internet by Physicians and Patients Increases,” Health Care News 3, no. 6 (2003), 1–7; and American
Academy of Family Physicians, “Survey Shows High Member Interest in EHR System,” FP Report, March
2003, www.aafp.org/fpr/20030300/8.html (19 January 2004).
R.H. Miller, J.M. Hillman, and R.S. Given, “Physician Use of IT: Results from the Deloitte Research Survey,” Journal of Health Information Management 18, no. 1 (2004): 72–80.
R.K. Yin, Case Study Research: Design and Methods, vol. 5, Applied Social Research Methods Series (Thousand
Oaks, Calif.: Sage Publications, 1994).

H E A L T H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 2

125

P u r su i t

O f

Q uali t y

6. D. Masys et al., “Giving Patients Access to Their Medical Records via the Internet: The PCASSO Experience,” Journal of the American Medical Informatics Association 9, no. 2 (2002): 181–191; and J.J. Cimino et al., “An
Evaluation of Patient Access to Their Electronic Medical Records via the World Wide Web,” in Proceedings
of the 2000 American Medical Informatics Association Annual Symposium, ed. J.M. Overhage (Bethesda, Md.: AMIA,
2000), 151–155.
7. American Medical Association, Physician Socioeconomic Statistics, 2000–2002 ed. (Chicago: AMA, 2001).
8. R.H. Miller, I. Sim, and J. Newman, Electronic Medical Records: Lessons from Small Physician Practices (Oakland,
Calif.: California HealthCare Foundation, October 2003).
9. M. Hoffmann et al., A Design Process for Embedding Knowledge Management in Everyday Work (Dortmund, Germany: Informatics and Society, University of Dortmund, 1999); C. Friedman and G. Hripcsak, “Natural
Language Processing and Its Future in Medicine,” Academic Medicine 74, no. 8 (1999): 890–895; and L.K.
McKnight et al., “Barriers to the Clinical Implementation of Compositionality,” in Proceedings of the 1999
American Medical Informatics Association Annual Symposium, ed. N.M. Lorenzi (Bethesda, Md.: AMIA, 1999),
320–324.
10. E. Brynjolfsson and L. Hitt, “Beyond Computation: Information Technology, Organizational Transformation, and Business Performance,” Journal of Economic Perspectives 14, no. 4 (2000): 23–48.
11. R.H. Miller and I. Sim, “The Costs and Benefits of Electronic Medical Records in Large Physician Practices: A Conceptual Model” (Unpublished manuscript, Institute for Health and Aging, University of California, San Francisco, 2003).
12. On inadequate electronic data exchange, see C.J. McDonald, “The Barriers to Electronic Medical Record
Systems and How to Overcome Them,” Journal of the American Medical Informatics Association 4, no. 3 (1997):
213–221.
13. On types of innovation adopters, see E.M. Rogers, Diffusion of Innovations, 4th ed. (New York: Free Press,
1995).
14. M. Buckley, Improving Drug Prescribing Practices in the Outpatient Setting: A Market Analysis (Oakland: California
HealthCare Foundation, 2002). See also these organizations’ home pages, www.surescripts.com and
www.rxhub.net.
15. IOM, Fostering Rapid Advances in Health Care: Learning from System Demonstrations (Washington: National Academies Press, 2002); and Centers for Medicare and Medicaid Services, “Health Insurance Portability and
Accountability Act (HIPAA)—Administrative Simplification,” www.cms.hhs.gov/hipaa/hipaa2 (4 May
2003).
16. Connecting for Health, Data Standards Group, Clinical Data Exchange Efforts in the United States: An Overview,
www.connectingforhealth.org/resources/DSWG_Backgrounder_AppendixA.pdf (2 February 2004).
17. IOM, Fostering Rapid Advances in Health Care.
18. D. Doolan and D. Bates, “Computerized Physician Order Entry Systems in Hospitals: Mandates and Incentives,” Health Affairs (July/Aug 2002): 180–188; and R. Galvin and A. Milstein, “Large Employers’ New Strategies in Health Care,” New England Journal of Medicine 347, no. 12 (2002): 939–942.
19. On the California “pay-for-performance” initiative, see, for example, Integrated Healthcare Association,
“History of IHA’s Pay for Performance Initiative,” www.iha.org/payfprfd.htm (6 May 2003); on the Leapfrog initiatives, see Leapfrog Group, “Factsheet, Computer Physician Order Entry,” 18 April 2003, www
.leapfroggroup.org/FactSheets.htm (6 May 2003). See also Bailit Health Purchasing, Provider Incentive
Models for Improving the Quality of Care (Washington: National Health Care Purchasing Institute, 2002).
20. Doolan and Bates, “Computerized Physician Order Entry Systems in Hospitals.”
21. G. LeGrow et al., ASPs, An Executive Report: Are Application Service Providers Ready for Prime Time? (Oakland:
CHCF, 2000).
22. Rogers, Diffusion of Innovations.
23. Harris Interactive, “eHealth’s Influence Continues to Grow”; J.M. Rothschild et al., “Clinician Use of a
Palmtop Drug Reference Guide,” Journal of the American Medical Informatics Association 9, no. 3 (2002): 223–
229; and S. Fischer et al., “Handheld Computing in Medicine,” Journal of the American Medical Informatics Association 10, no. 2 (2003): 139–149.

126

M a r c h /A p r i l 2 0 0 4