Readers and
informatics educators utilizing this site: please let me know how you arrived
here and your
comments. Additional case examples are
also appreciated.
Introduction:
Medical Informatics, Information Technology Leadership, and Healthcare
IT Success
Throughout the past several
decades, significant advancements have been made in the field of information
technology (IT) and in the discipline of biomedical information science or biomedical
informatics. As a result of these advancements and the
synergies they have enabled, in recent years many of the early and often highly
overstated predictions about the use of computers in medicine have become at
least technically feasible.
Yet diffusion of healthcare
information technology after 30-plus years of effort and billions of dollars
spent remains limited. As per the 2008
statistics in the NEJM article "Electronic Health
Records in Ambulatory Care - A National Survey of Physicians", NEJM
359:50-60, just four percent of physicians in the
This website is concerned
with the reasons for this apparent paradox.
While clinically oriented IT is now potentially capable of achieving
many of the claimed benefits (improved medical quality and efficiency, reduced
costs, better medical research and drugs, earlier disease detection, and so
forth), there is a major caveat and essential precondition: the benefits will be
realized only if clinical IT is done well. For if clinical IT is not done well, as often occurs in today’s environment of medical
quick fixes and unquestioning exuberance about IT, the technology can be injurious
to medical practice and biomedical R&D, and highly wasteful of scarce
healthcare capital and resources.
Those two short words “done well” mask an underlying, profound,
and, as yet, largely unrecognized (or ignored) complexity. This website is about the meaning of "done well" in the context of
clinical computing, a computing subspecialty with issues and required expertise
quite distinct from traditional MIS (management information systems, or
business-related) computing.
This site is dedicated to the
improvement of healthcare via enlightened approaches to clinical information
technology, and to the field of Medical Informatics that helps inform the innovative thinking
required to achieve the full benefits of clinical IT. Medical informatics is a cross-disciplinary
field that studies information-seeking activities and tools, analytic
processes, and workflows in biomedical research and clinical care delivery. The
field focuses upon the innovative use of computers in clinical medicine,
molecular biology, neuroscience, and other areas of biomedical research.
This website is also
dedicated to my early mentor, Dr. Victor P. Satinsky, who ran NSF-funded
science programs for high school students at
Medical informatics research
informs the best approaches to clinical information technology design,
implementation and ongoing maintenance and adaptation. (Paraphrasing Dr.
William Hersh at OHSU, "It is
unwise spending millions on Electronic Medical Records without investing
thousands in Medical Informatics expertise.") Unfortunately, that
advice is often not followed, even though specialized postdoctoral training
in Medical Informatics is funded by The U.S. National Institutes of Health
(NIH) at a number of major universities, and is provided by other universities
in the U.S. via internal funds and internationally
as well.
This website focuses on the leadership aspects of clinical IT. It promotes viewpoints based on repeated
observations by this author and Medical Informatics colleagues of a concerning
phenomenon. We observe healthcare
support personnel in information technology, and the non-medical management
personnel overseeing them, assigned roles and responsibilities (largely via
custom and tradition) that increasingly fall outside their core competencies,
due in large part to advancements in biomedical information sciences and in the
computer applications used at the point of care.
These advances mandate cross disciplinary expertise
and insights for optimal leadership
of clinical IT initiatives. The existing
leadership model, where technology-focused healthcare support personnel (facilitators of healthcare) control the
information tools and assets essential to clinicians (the enablers of healthcare), can adversely affect clinical medicine and
biomedical research. More on the issue
of cross disciplinary expertise is below.
It should be noted that the
purpose of these writings is not to
assign "blame", but to start a needed dialog on addressing solutions to the
multidimensional problems faced by healthcare organizations with regard to
clinical IT.
------------------------------------------------------------------------------------------
All is unfortunately not well in the world
of clinical information technology. In medicine, a field characterized by
significant risk and unpredictability, a somewhat remarkable and unexpected
atmosphere of "technologic determinism" (a belief that
computer-based automation is almost magically beneficial) seems common. The
appropriate levels of critical thinking and skepticism essential in a demanding
area such as introduction of computer automation in medicine appear largely
absent, on a worldwide basis, to the point that those who've led IT projects to
automate traditional business activities (e.g., accounting, finance,
manufacturing) are deemed the appropriate leaders to automate clinical
medicine.
Why is this an important issue? As just one example, as a Director of Medical
Informatics I encountered apparent endangerment of ICU patients from
inappropriate, dust-laden, air-circulating, business class computers subject to
bacterial colonization that were mounted on the ceilings above each ICU bed.
They were malfunctioning as well due to a system architecture inappropriate for
mission critical settings. My advice on changes to more appropriate hardware
and other measures to ensure patient safety were simply overruled by IT
personnel. My counsel was overruled and ignored by the IT staff and CIO on
grounds that the IT staff were unfamiliar with existing, ICU-appropriate
computer hardware and wouldn't support or even evaluate "nonstandard"
(to them) computers in any case. Patients
remained at risk. Having spent much
time in medical and cardiac ICUs during my residency and having done what it
takes to provide the very best of care to extremely ill patients with little
consideration for my own convenience, I found this experience rather
remarkable. This experience informs the first
case example of health IT dysfunction at this website.
Unfortunately, even well-designed healthcare IT applications often lack the
sustained agility in critical functions such as charting, information
retrieval, and decision support to keep up with the pace of the hospital and
clinic, and with the pace of change in medical science and practice. These
applications can distract clinicians and make their work harder and more
stressful. Worse, much of the technology now available comes nowhere near the
optimal design possible.
Further, the use of
electronic health records, without a major change in health care delivery,
would not significantly reduce overall health care costs, the director of the
Congressional Budget Office said at the release of the agency's 2007 report on
long-term health care spending. Peter
Orszag, CBO's director, said that according to data from the report, the return
on investment for EHRs "is not going to be as substantial as people
think."
A 2003 press release about
research at OHSU (one of numerous NIH-funded centers for Medical Informatics
research and training) entitled "Most hospitals don't use latest ordering
technology" (link)
is particularly illuminating:
Computers
programmed to screen out errors and standardize physicians' orders for
prescriptions, tests and other care have been a source of hope in reducing
medical errors and improving patient safety. The problem is that most hospitals
aren't using this technology, known as computerized physician order entry
(CPOE) ... Reducing medical errors gained a sense of urgency in 2000 when the
Despite
a wide array of quality, policy and financial incentives to use CPOE, fewer than
10 percent of American hospitals make it completely available to their
physicians. This was among the findings of a study conducted by researchers in
the Oregon Health & Science University School of Medicine and recently
published in online edition of the Journal of the American Medical Information
Association (JAMIA). [Use of CPOE] is
not yet widespread because it has a reputation for being difficult to implement
successfully. Patient care information systems like CPOE ... can create
unintended or "silent" errors, according to a separate study
conducted by the same author in the
"Many
information systems simply don't reflect the health care professional's hectic
work environment with its all too frequent interruptions from phone calls,
pages, colleagues and patients. Instead
these are designed for people who work in calm and solitary environments.
This design disconnect is the source of both types of silent errors …Some
patient care information systems require data entry that is so elaborate that
time spent recording patient data is significantly greater than it was with its
paper predecessors," the authors wrote. "What is worse, on several
occasions during our studies, overly structured data entry led to a loss of
cognitive focus by the clinician."
Software designed for calm
and solitary environments, intended for physicians, nurses and other clinical
personnel? This is nothing short of
remarkable.
In 2005
In effect, systems that
ignore the healthcare workplace's well-known realities are finding their way
into real products costing millions of dollars that are acquired and
implemented by hospitals, and these systems might actually be causing adverse
effects. How is this possible? This phenomenon should raise a number of
questions to the critical thinker, such as:
who are the CPOE designers, exactly, and what are their backgrounds? How
could investor dollars have been spent in such a fashion as to ignore the
fundamental realities of clinical settings? How could IT companies have
designed and implemented systems that "led to a loss of cognitive focus by
the clinician" and created error?
How could hospitals have acquired and put such systems into
operation? What is behind this
irrational exuberance and blind faith, or, as I have described it, an
“inappropriate overconfidence in computers?” Why has commerically available clinical
IT, developed under the management information systems (MIS) paradigm, not
lived up to expectations that in turn are based on the experiences of medical
centers with advanced information systems developed internally under the aegis
of clinicians and experts in Medical Informatics?
In a 2007 research study involving the University at Buffalo and other
institutions, "Design
of Patient Tracking Tools May Have Unintended Consequences", researchers
found that “Virtual Patient Status Whiteboard” systems to replace traditional,
dry-erase patient status boards in hospital Emergency Departments (ED’s), but
designed by technologists without adequate domain expertise, often do not work
as expected. The computer-based system
interferes with staff communications.
From the report:
In some cases, providers noted that computer
systems hid some of the information;
if only three comments could be viewed per screen, they had to click to get to
another screen, requiring them to search for information that might demand
immediate attention. [How such an ill-conceived design for an ED
status board could have been put into production is remarkable – ed.] The study also found that there were fewer visual cues with the computational
system. Some providers noted that they used to be able to get a sense of
the status of the emergency department just by walking through the room and
visually checking the manual whiteboard.
“Without that public display, providers have to sit down at the computer
and check it, which can add time or
reduce awareness,” said a principal investigator.
These are potentially disastrous consequences in an ED environment where
patients can be highly unstable and serious events transpire rapidly and
irreversibly. An investigator in the
study observed that "the results provide an important case study of what
can happen when new technologies are developed without sufficient understanding
by designers of the nature of the work in which they will be used."
This last observation raises a more fundamental question: why does such an axiomatic, common-sense
statement, especially in a domain as complex as medicine, need to even appear in print?
With regard to electronic
health records (EHRs), a research article in the Archives of Internal Medicine
entitled “Electronic Health Record Use and the Quality of Ambulatory Care in
the United States” (Arch Intern Med. 2007;167:1400-1405, link to abstract here)
reached what to many was a counterintuitive and paradoxical conclusion. The authors examined electronic health
records (EHR) use throughout the
An informal poll taken by
Dr. David Brailer (the first director of the U.S. Office of the National
Coordinator for Health IT) while speaking to a group of clinicians, was reported
as indicating that as many as 20% of physicians have tried to implement EHR. Of
those, 50% had experienced a failure. Over 70% of the clinicians knew of a
colleague who had experienced an EHR implementation failure.
These findings and others
like it are indeed troubling. An EHR for
small-group and solo-practice physicians costs $44,000 per physician, and
generates an average ongoing $8,500 per year in annual costs, ACP president
Lynne Kirk, MD told the house Subcommittee on Regulations, Healthcare and Trade
of the House Committee on Small Business in October 2007. "The business
case does not exist to make this kind of capital investment," Kirk told
the Subcommittee.
It appears the clinical case
may not exist, either, for clinical IT as
it is designed and implemented today.
Promises made about how this technology, as it is today, will
revolutionize healthcare might better reflect a “magic bullet theory” of
IT-enabled transformation than clinical and social reality.
How can it be that “as
implemented, EHR’s were not associated with better quality ambulatory care?” It
has been said that there are no paradoxes, only false assumptions. To medical informaticists, the key phrase
that explains these findings about EHR’s is “as implemented”, to which I would
also add “as designed.” In other words,
EHR design and implementation across a wide variety of settings were so
suboptimal that in some cases and for some clinical quality indicators, clinicians using paper performed better
than clinicians using EHRs.
Considering the expense of these information systems, the lack of
discretionary funds within healthcare for IT experimentation, and the scarcity
of resources to be wasted upon “IT misadventures” that might be better spent on
providing clinical care, these results are troubling.
These issues are international
in scope. Richard Granger, former head
of the UK's “Connecting for Health” national clinical IT program, had this to
say about a program described by some UK members of Parliament as “the largest
government IT debacle ever” (see here
for more):
"Sometimes we put in stuff that I'm just ashamed of ... Some
of the stuff that [our large American clinical IT vendor] has put in recently
is appalling ... [vendor] and [prime contractor] had not listened to end users
... Failed marriages and co-dependency with subcontractors ... A string of
problems ranging from missing appointment records, to inability to report on
wait times ... Almost a dozen cancelled go-live dates ... Stupid or evil people
... Stockholm syndrome -identifying with suppliers' interests rather than your
own ... A little coterie of people out there who are "alleged
experts" who were dismissed for reasons of non-performance."
Finally, At a website of the
Working Group for Assessment of Health Information Systems of the European
Federation for Medical Informatics (EFMI)
entitled “Bad Health Informatics Can
Kill” (link) are
summaries of a number of reported incidents in healthcare where IT was the
cause or a significant factor in patient harm.
Is there a unifying issue that
can explain these findings about clinical information technology systems?
------------------------------------------------------------------------------------------
As clinical and biomedical
IT becomes increasingly more complex, and as it supports increasingly complex
medical science, research and practices, the number of ways that failures and
mishaps can occur from errors in judgment, inadequate knowledge, mismanagement,
and related factors increases markedly. Competence, excellent management, logical
decision making, and the wide-angle view of true cross-disciplinary
expertise have therefore become imperatives for leadership and success in
this field. Unfortunately, the reality
in today's hospital, EHR vendor and biomedical R&D IT departments falls
short of this.
Studies of IT failure have
been limited, but statistics on failure rates from some of the studies
performed in the business IT sector may be found here. Clinical IT failure rates may be even higher
than those of business IT, due to the greater social and technological
challenges of the healthcare IT subspecialty.
A 2007 Medical Records
Institute survey of about 800 health IT stakeholders is summarized in the
Modern Healthcare article “Failure,
de-installation of EHRs abound” at this
link. Nineteen percent of
respondents indicated they either have experienced the actual abandonment of an
EMR system or are now going through a de-installation, and some had actually
gone back to paper. I believe these
results markedly under-represent true problem rates, as the sample size was
small and health IT stakeholders are quite reluctant to share such negative
information due to internal healthcare organization and health IT industry
backlash.
As far back as 1969, EMR and Medical Informatics pioneer Donald A. B. Lindberg,
M.D., now Director of the U.S. National Library of Medicine at NIH, made the
following observation. He wrote that "computer engineering experts per se
have virtually no idea of the real problems of medical or even hospital
practice, and furthermore have consistently underestimated the complexity of
the problems…in no cases can [building appropriate clinical information
systems] be done, simply because they have not been defined with the physician
as the continuing major contributor and user of the information" (Lindberg
DAB: Computer Failures and Successes, Southern Medical Bulletin
1969;57:18-21).
Surprisingly, there has been little change in this issue in thirty-five years.
Today the IT personnel and non-medical managers (e.g., non-degreed IT staff, BS
or MS in computer science, MBA's, even PhD's) who by custom and tradition are
assigned leadership roles in EMR and clinical data research initiatives via
control of critical decisions, budgets and resources, often lack clinical
experience and insight. Specifically, personnel of an information technology
background, with little or no background in the biomedical sciences, often are
positioned by senior management as enablers, rather than facilitators,
of such initiatives. They retain a major say in what is -- and is not -- done,
and in the tools provided to perform clinical care and biomedical R&D.
From a dual perspective as
both a clinician and computer professional, it is evident that this arrangement
is faulty, and that critical clinical computing projects benefit greatly from
an alternate approach to project preparation, development, implementation,
customization and evaluation as compared to management information systems
(MIS) projects. Clinical computing
and business computing are different, highly distinct subspecialties of
computing, much as neurosurgery and psychiatry are highly distinct medical
subspecialties. (Both of the latter
fields focus on the same body part, but the approaches and methodologies of
neurosurgery do not work well in psychiatry, and vice versa.)
The true nature of clinical
settings and the major fault in the management model for healthcare IT is
captured quite pithily in the article “Hiding in Plain Sight: What Koppel et al. tell us about Healthcare
IT” (Nemeth & Cook, Journal of Biomedical Informatics 2005;38:262-263, link
to pdf):
On
the surface, healthcare work seems to flow smoothly. That is because the
clinicians who provide healthcare service make it so. Just beneath the
apparently smooth-running operations is a complex,
poorly bounded, conflicted, highly variable, uncertain, and high-tempo work
domain. The technical work that clinicians perform resolves these complex
and conflicting elements into a productive work domain. Occasional visitors to
this setting see the smooth surface that clinicians have created and remain
unaware of the conflicts that lie beneath it. The technical work that
clinicians perform is hiding in plain sight. Those who know how to do research
in this domain can see through the smooth surface and understand its complex
and challenging reality. Occasional
visitors cannot fathom this demanding work, much less create IT systems to
support it.
The Office of the National
Coordinator for Health Information Technology (ONC) has recognized numerous barriers to clinical IT
adoption. One barrier is a high
failure rate for electronic health record implementation:
Current Market Barriers and Challenges to Widespread
Adoption of Health Information Technology (ONC)
·
Payers don't reward efficiency or quality; they pay based on
volume
·
Adoption issues
o
There is a negative business case for typical health
information technology adopter
o
There is a significant electronic health record adoption gap
based on organization size
o
There is a first mover disadvantage for health information
technology buyers
·
High failure rate for electronic health record
implementation
o
There is variable availability of IT expertise in physician
offices
o
There is a high failure risk for business re-engineering
o
There is limited implementation support for 75,000 small
practices
·
Limited capacity for interoperability
o
Few health information technology products include standards
o
Standards are not rigorous and lag behind commercialization
o
There is no viable health information exchange
infrastructure
IT personnel in healthcare
often believe that success in implementing management information systems
applications ("business computing") supersedes or actually renders
unnecessary the mastery of medicine in leading and controlling implementation
of clinical computing tools. Yet, mastery of applied IT towards implementing
management information systems is in large part mastery of process (e.g., in
acquiring and supporting vendor-written software) and repetition, as
opposed to the practice of medicine, which requires mastery of complexity.
In other words, applied IT is a field of a relatively small number of principles,
a large number of arbitrary conventions and rules, and a narrow body of
knowledge applied repetitively and programmatically, often without scientific
rigor. This may be illustrated by the fact that most areas of applied IT can be
done well, and often are, by those with little or no formal training. This is
not to imply that applied IT is itself easy, which it is not. There is no
substitute for talent and real-world experience.
In clinical IT settings, however, there must be the right experience.
Experts in clinical computing must provide effective solutions via seasoned
application of the concepts, techniques, knowledge, and processes of medicine,
and display an expert level of critical thinking in applying principles,
theories, and concepts on a wide range of issues that are unique to clinical
settings. Business IT experience alone does not provide a sufficient background
for such responsibilities to be carried out effectively. Further, medicine is a
domain of many difficult, nonintuitive principles, experimentally-derived
natural laws, and a large body of knowledge applied in a broad, interconnected
manner, ideally with critical scientific rigor. It cannot be practiced
successfully without significant mastery of an enormous body of biomedical knowledge
and significant hands-on patient care. The IT model of "If it's
information, we do it" starts to fall apart and impede progress in such
organizationally and sociologically-complex environments.
Leaders in clinical IT must be experienced in medical sciences and in the
complex social and organizational issues of healthcare, such as the need for
multiple, contextual levels of confidentiality, the politics and psychology of
medical practice and referral, the complex medical workflow and the need to rapidly
improvise due to the unexpected ("there are no committees in cardiac
arrest situations"), and societal and personal sensitivities towards the
physician-patient interaction.
In effect, management information systems and clinical systems are highly distinct.
The belief that mastery of IT process and repetition for management information
systems implementation entitles IT personnel to lead and control implementation
and operationalization of essential tools in complex domains such as medicine
(e.g., electronic medical records systems) is presumptuous and creates an
environment strongly misaligned with the business of healthcare delivery.
The belief often results in the exclusion or misutilization of Medical
Informatics experts, appropriate clinicians, and other forms of mismanagement
exemplified in the typical cases below.
Unfortunately, healthcare IT journals avoid covering these issues due to an
apparent but understandable “prior restraint” of items that ruffle industry
feathers. Remarkably, those same
publications commonly offer articles acclaiming the value of IT personnel allowing
clinicians to participate in clinical systems implementation. Clinician
involvement is so obviously necessary that such articles might be compared to
the New England Journal of Medicine publishing articles on the value of
employing sterile technique during surgery. A critical reader should question
why articles about IT personnel needing to allow clinicians to participate in
healthcare IT still appear in print.
The familiar stories of healthcare IT failure and organizational discord in
hospitals and academia below reflect, as their root cause, basic mismanagement
due to significant inadequacies in organizational thinking, structures and
support of healthcare information technology. Such technology is vital to
healthcare quality improvement and prevention of errors. As these stories
illustrate, however, this technology is not always treated as such by
healthcare leadership, including officers at the "C" level (CEO, COO,
CIO etc.) and Boards of Directors.
It should be remembered that failed healthcare IT projects are not caused by
immutable organizational or political issues. Importantly, failures are caused
by the mismanagement of the organizational and political issues and of
the people who create the problems associated with these issues.
The direct economic costs of such IT failures (often caused by a minority of
personnel in an organization) is in the millions of dollars per year per
healthcare organization. The resultant less tangible costs of lost opportunity
are more difficult to quantify, but are probably much greater than the direct
losses in the long term.
Medical professionals are being held to increasingly stringent standards of
quality and accountability at the same time they are becoming highly dependent
on healthcare IT in taking care of patients. Those who are responsible for
healthcare IT, including senior healthcare management, have not been held to
the same standards of quality and accountability as the medical professionals
dependent on this critical IT. This needs to change.
Healthcare, pharmaceutical, and other biomedical R&D organizations that
conflate "information provision" with "information
technology" and depend on business-IT personnel to do the work of medical
informaticists and clinical information scientists do so at their peril.
We believe there are a
number of key issues that need to be addressed.
These might be categorized as follows:
1. Resistance of some clinicians to rigorous
information practices that support quality and safety (which is inclusive of,
but not exclusive to, IT resistance);
2. Resistance of the healthcare IT vendors to
high-quality user-centered design practices;
3. Resistance of IT personnel within delivery
organizations to user-centered design practices i.e., in customizations of
vendor-acquired products, or internal development of specialized systems;
4. A belief in IT solutions by many stakeholders
as a "magic bullet" or panacea - i.e., build an IT system and
miracles in clinical quality, operational, compliance and documentation
improvements will occur.
5. Financial disincentives for many providers,
expecially community based clinicians, to adopt clinical IT;
6. Knowledge that existing systemic
organizational faults do contribute to errors in health care delivery.
As leaders in the
transformation of modern medicine, addressing these issues requires consensus
to build strength and voice for which the informatics community must
lobby. The policy initiatives that
informaticists pursue should result in:
·
user-centric
design practices in industry through studying health IT problems that
legitimately cause physician resistance,
·
empowerment of
Chief Medical Information Officers (CMIO's) with executive presence and
managerial authority,
·
reformation of
the hospital MIS department "designer-centric" culture,
·
improved
financial incentives for IT adoption,
·
studies of
systemic organizational and societal barriers with an aim to improve them.
"Blame" as an
issue does not move us forward as a medical subspecialty with mission, vision
and values. If we are to issue
"blame", it should perhaps be directed towards those minority of
individuals among the stakeholder groups who make suboptimal decisions out of
self-interest. Empowered CMIO's and
trained informatics personnel with proven expertise are one remedy to
"filtering" those who have lost sight of the broader perspective of
improving healthcare with appropriate HIT tools.
An article worth reviewing
is "Human error: models and management ", James Reason (a fitting
name!), BMJ 2000;320:768-770 ( 18 March ), http://www.bmj.com/cgi/content/full/320/7237/768
:
Summary points:
·
Two approaches
to the problem of human fallibility exist: the person and the system approaches
·
The person
approach focuses on the errors of individuals, blaming them for forgetfulness,
inattention, or moral weakness
·
The system
approach concentrates on the conditions under which individuals work and tries
to build defenses to avert errors or mitigate their effects
·
High reliability
organizations---which have less than their fair share of accidents---recognize
that human variability is a force to harness in averting errors, but they work
hard to focus that variability and are constantly preoccupied with the
possibility of failure.
