Reconceiving Health Care To Improve Quality

WILLIAM C. RICHARDSON

The Tanner Lectures on Human Values

Delivered at

University of California, Santa Barbara

January 29, 2001
William C. Richardson is president and chief executive ofŠcer of
the W. K. Kellogg Foundation. He is also professor and president
emeritus of the Johns Hopkins University. He was educated at Trinity
College, Connecticut, and at the University of Chicago, where he re-
ceived his Ph.D. He was executive vice president and provost of the
Pennsylvania State University, and was dean of the Graduate School and
vice provost for research of the University of Washington. He is a mem-
ber of the Institute of Medicine of the National Academy of Sciences
and a fellow of the American Academy of Arts and Sciences and the
American Public Health Association. He has served in many capacities
in the U. S. Department of Health, Education and Welfare, and the U. S.
Department of Health and Human Services; on advisory committees
for the Robert Wood Johnson Foundation, the Rockefeller Foundation,
and the Pew Charitable Trusts; and on the board of trustees of the Henry
J. Kaiser Family Foundation and the Glenmede Trust Company. He is
currently serving on the board of the Council of Michigan Foundations
and as chairman of the board of trustees of the Council on Foundations.
He is the author of numerous articles, monographs, and reports on issues
related to public health.
Good evening and thank you for the warm welcome.
Our society and economy have been undergoing some marked
changes over the past twenty years. My topic this evening is envisioning
a new set of relationships in health care and a degree of professional and
organizational rearrangement that rešect these changes and that will, in
the process, materially improve health care quality (Institute of Medi-
cine 2000).
Let me begin with the basic premise of any encounter or relationship
with a health care provider. We expect—at the very least—to be the
better for it, and certainly no worse off. Since the days of ancient Greece,
“First, do no harm” has been the Šrst principle of medicine.
A little over a year ago, however, the Institute of Medicine of the Na-
tional Academy of Sciences (which I will refer to as the IOM) released a
groundbreaking report on medical errors that suggested we are doing
harm (Institute of Medicine 2000). Although we have, in many ways,
the world’s Šnest health care system, the report found a surprisingly
high level of medical errors. And, although Americans are healthier and
are living longer than ever, our over-stressed health care establishment
too often fails to insure the basic safety of its patients.
The report received extraordinary media attention—two days of
near-saturation coverage on the national networks and a great deal of lo-
cal media coverage as well. Further, the story turned out to have what
some refer to as “legs.” I don’t think there has been a week in the last
year that there wasn’t some national coverage of the medical errors issue,
except maybe the second week of November.
Naturally, much of the media’s attention focused on the statistics
and personal stories related to medical errors. And these are certainly a
vital aspect of the quality problem. But this evening I’d like to offer
something of a sequel to the medical errors report. A range of pertinent
demographic trends; technical, scientiŠc, and societal developments;
and changes in health care have been discussed by members of the IOM
Committee that was responsible for the errors report. Based on these

This lecture draws heavily on the deliberations and contributions of members of the
Committee on the Quality of Health Care in America, which I chair. The Šrst report of this
Institute of Medicine Committee is a key source, as are sections of our committee’s second
report, available in the spring of 2001.

[3]
4 The Tanner Lectures on Human Values

discussions, I will provide an overview of what a safer and more patient-

centered health care system might look like. I think the ideas that are
emerging can reduce medical errors but also lead to a much more effec-
tive and productive health care enterprise. Our report on these broader
considerations should be out in early spring 2001 (Institute of Medicine
2001).
But Šrst let me provide a brief recap.
In December 1999, along with several health care leaders, I joined
President Clinton at the White House to discuss the implications of the
medical errors report. As I said, the report had made headline news
across the country, but as important to me was the groundswell of pub-
lic interest that was prompted by its release. For weeks afterward the In-
stitute of Medicine was šooded with calls from hundreds of people who
had questions or personal stories to share. (The number I received per-
sonally was startling.) The report clearly touched a nerve with the
American people. While health policy can be highly technical and ob-
scure, everyone can relate to the tragedy of a serious medical mistake.
Local and national and private and public sector leaders have all taken
notice. Many professional groups (such as the American Medical Associ-
ation, American Nurses Association, American Hospital Association,
American College of Physicians/American Society of Internal Medicine,
American College of Surgeons, American Board of Medical Specialties,
and the Accreditation Council for Graduate Medical Education, to name
but a few) are taking action, including institutional practices, a careful
examination of the roles of board certiŠcation and recertiŠcation, and
undergraduate, graduate, and continuing health professions education
in improving safety. The business community, including the Business
Roundtable and the National Business Coalition on Health, has an-
nounced purchasing initiatives to promote patient safety. Last Novem-
ber the so-called Leapfrog Group unveiled a market-based strategy to
improve safety and quality, including encouraging the use of computer-
ized physician-order entry, evidence-based hospital referrals, and the use
of intensive care units staffed by physicians credentialed in critical care
medicine.
The Clinton administration and the Congress have also been active.
The Senate held its Šrst hearings on the issue within a few weeks of the
report’s release, and additional hearings were conducted by committees
of both the House and the Senate. The administration launched an am-
bitious plan of federal action to reduce medical errors. In general, then,
[Richardson] Reconceiving Health Care to Improve Quality 5

there is broad bipartisan support for taking action, and I would be sur-
prised if quality of health care were not a signiŠcant issue for the Bush
administration.
What the report found, after an exhaustive review of the literature,
was a staggering number of medical errors. Based on the Šndings of one
major study, about 44,000 patients die each year from medical errors in
hospitals alone (Thomas et al. 2000). Another study puts that Šgure
even higher, at 98,000 patients a year (Leape et al. 1991). Even if we use
the 44,000 patient Šgure, medical mistakes in hospitals rank eighth in
the United States among the leading causes of death. More people die
each year in the United States from medical errors, for example, than
from trafŠc accidents, breast cancer, or AIDS.
Within a short time of the report’s release, some began to question
the numbers. “Could the problem really be this serious?” they asked.
We will never know the exact number of medical errors, but there is lit-
tle doubt that the evidence is strong and consistent. Let me take a mo-
ment to point out several things about the nature of the evidence base.
First, the conclusions of the IOM Committee were not based on just
these two studies. There are two large studies—the Harvard Medical
Practice Study using about 30,000 medical records and a study in Col-
orado and Utah (Thomas et al. 2000; Leape et al. 1991) based on about
15,000 records—and more than thirty others (chapter 2 in Institute of
Medicine 2000) in leading peer-reviewed journals in the last ten years.
Second, in some ways, we have really only looked “under the lamp
post” for errors. Nearly all studies focus on hospital settings, yet most
care is delivered in doctors’ ofŠces, ambulatory surgery centers, clinics,
patients’ homes, and nursing homes. Granted, most would concur that
the hospital is a higher-risk environment for errors, because patients in
hospitals are generally receiving more medical interventions and more
drugs, sometimes under hurried or critical circumstances. But we have
seen a steady migration of many procedures such as some laparoscopic
surgery to ambulatory settings. One study has found that some 7,000
patients die each year just from medication errors that occur both in and
out of hospitals (Phillips et al. 1998).
Third, most studies identify errors from information that is docu-
mented in handwritten medical records. It is likely that many errors
never get documented in medical records. Providers are acutely aware of
liability concerns. In addition, some errors are not recognized or noticed
by clinicians. Studies that rely on “automated signals” to detect errors
6 The Tanner Lectures on Human Values

(such as abnormal or unexpected laboratory test results) reveal higher

rates of errors than do studies that rely on the documentation or report-
ing of errors by clinicians (Classen et al. 1991).
At this point, anyone who is not familiar with the report might
gather that it is an indictment of health professionals. But this is not at
all the case. Indeed, a main conclusion of the report is that no particular
group or entity is to blame. A fundamental conclusion is that this is not
a “bad apple” problem. And indeed that pinpointing and placing blame
is actually a counterproductive exercise. To reduce medical errors and
improve patient safety we need to recognize this as a systems problem.
The comprehensive strategy put forth by the IOM report calls for ac-
tion by government, industry, consumers, and health care providers.
The report notes that it may be human to err, but it’s also human to cre-
ate solutions, Šnd better alternatives, and devise new approaches for the
challenges ahead. In order to meet these challenges, however, we must
Šrst face facts. Our health care system is a decade or more behind other
high-risk industries in its attention to ensuring basic safety. The ordi-
nary risk of dying in a domestic airline šight or at the workplace has de-
clined dramatically in recent decades. That is due in part to increased
attention by industry efforts, and in part because federal agencies were
created to improve safety. Drawing on these models, my colleagues and
I urge several courses of action, which are detailed in the report. Let me
just quickly summarize them here before moving on to the topic of
quality of health care more broadly deŠned.
First, we urged Congress to create a national center to set national
safety goals, track progress in meeting them, and support research to
learn more about preventing errors. It would also act as a clearing-
house—an objective source of the latest information on patient safety
for the nation. I am pleased to say that prompt action has been taken.
We also recommend that a nationwide, mandatory public reporting
system be established by the states. Currently, only about one-third of
the states have reporting requirements. And yet, this information is
critical if we are to learn, in any systematic way, about practices that lead
to serious injury or death. It’s not unlike our being required to report se-
rious trafŠc accidents or deaths from workplace accidents. During the
last year, the National Academy for State Health Policy has convened
leaders from both the legislative and executive branches of the states to
discuss approaches to improving safety and with the support of private
foundations has recently completed an inventory of current state report-
[Richardson] Reconceiving Health Care to Improve Quality 7

ing initiatives (National Academy for State Health Policy, forthcom-

ing).
At the same time, we also recommend federal legislation to protect
the conŠdentiality of certain information. This protection would cover
medical mistakes that do not result in harm—so-called near misses—
where information is collected and analyzed solely to improve safety and
quality. Such legislation would encourage the growth of voluntary,
conŠdential reporting systems so that practitioners and health organiza-
tions can correct problems before serious harm occurs. Without such
legislation, a doctor or hospital may be afraid to gather such informa-
tion, for fear that it would be subpoenaed and used to establish a pattern
of error in a subsequent lawsuit.
Again, the majority of errors in hospitals do not result from impaired
professionals or recklessness. Most mistakes occur because of basic šaws
in how elements of the health system are designed and organized. For ex-
ample, equipment controls that vary from one manufacturer to another,
or from year to year, can contribute to errors. Stocking patient care units
with drugs that are potentially lethal unless they’re diluted has led to
deadly overdoses. And, because of illegible handwriting in medical
records, patients have been given drugs in wrong doses or given drugs to
which they have a known allergy.
There is no single solution sufŠcient to bring about the degree of
needed change. Rather, we should think and act systemically, to create
an environment where safety will become a top priority for health care
organizations and providers. The goal here is not to blame individuals
or to seek retribution. Instead, we want to design new systems that pre-
vent, detect, and minimize hazards and the likelihood of error. In short,
we want to create a new culture of safety in the American health care
system. We want a system in which it is hard to make a mistake.
We are not, however, calling for a nationwide master plan to solve
the problem of patient safety. The American health system doesn’t work
that way. There are too many markets and too many variables. What
works well for the Henry Ford Health System in Detroit may not be ap-
plicable in Santa Barbara—much less in Maine.
While the Šndings on medical errors are sobering, perhaps more
sobering is the realization that medical errors are but one manifestation
of an even larger problem. Medical errors, the failure to execute a care
plan as intended, are not the only type of “system weakness.” There is
ample and growing evidence that much of health care is not strongly
8 The Tanner Lectures on Human Values

science-based (Schuster et al. 2001). In many instances, patients don’t

receive effective care. That is, many people receive services for which the
potential risks exceed potential beneŠts—some call this “overuse.”
Others do not receive services from which they would likely beneŠt—
generally called “underuse.”
The most extensive reviews of the literature on quality have been
conducted by colleagues at RAND (Schuster et al. 1998). The RAND
review now includes over 100 publications in leading peer-reviewed
journals documenting overuse, underuse, and errors (Schuster et al.
2001). Overuse of health services is common. Examples include perfor-
mance of surgery for hysterectomy, coronary artery bypass graft, and
other procedures without appropriate reasons; provision of antibiotics
for the common cold and other viral upper respiratory tract infections
for which they are ineffective; and insertion of tubes in children’s ears in
the absence of clinically appropriate indications.
Underuse is also a serious concern for all types of clinical conditions.
In a study of approximately 3,700 Medicare patients with a diagnosis of
heart attack and eligible for treatment with beta blockers, only 21 per-
cent received beta blockers within 90 days of discharge. The adjusted
mortality rate for patients with treatment was 43 percent less than that
of patients without treatment (Soumerai et al. 1997). Another study
found that an estimated 18,000 people die each year from heart attacks
because they do not receive effective interventions (Chassin 1997).
Overall, it is not an overstatement to say that the health care delivery
system is unable to provide consistently high-quality care. Or said an-
other way, many people simply do not beneŠt from what medicine has
to offer. There is a large “quality gap.” Several expert panels in recent
years have come to this conclusion (President’s Advisory Commission
1998; Chassin et al. 1998). One example would be the IOM National
Cancer Policy Board. Its report Ensuring Quality Cancer Care (Institute of
Medicine 1999) examined the quality of cancer care in depth and con-
cluded that there is a large gap between what care should be and the care
that many patients actually experience.
The signiŠcance of the report To Err Is Human (Institute of Medicine
2000) was not that it reinforced the messages of these other reports, but
that it focused on one readily understandable aspect of quality—med-
ical errors—and it communicated this problem effectively to a very
broad audience, including the lay public.
These panels, of course, are not the only groups that have been rais-
[Richardson] Reconceiving Health Care to Improve Quality 9

ing issues about quality. The Institute for Healthcare Improvement, the
National Committee for Quality Assurance, the Foundation for Ac-
countability, many local institutions (such as the Institute for Clinical
Systems Improvement in Minnesota), and others have been calling for
action to address quality concerns for some time.
All of these efforts have contributed to what I think is a “sea change”
in the way health care leaders, policymakers, purchasers, physicians,
nurses, and, increasingly, consumers view quality. This range of activity
has achieved widespread recognition that quality in health care is not
what it should be and that this is a problem that must be dealt with. We
have in many ways turned a corner, with much change now possible that
wasn’t possible just a few years ago.
Not only is there increased recognition that we have a serious prob-
lem, but there is also recognition that we cannot address it successfully
by simply tweaking today’s health care system. Telling providers to
work harder on coming up with more quality measurement tools just
isn’t going to be enough. Fundamental change is needed in how we
organize and deliver health care; and for this to happen, we need far-
reaching changes in the culture of medicine and in the environment in
which health care is provided.
As we chart a course for the redesign of health care, it is helpful to un-
derstand some of the major technological and demographic forces that
are currently shaping it and will continue to do so in the future. There
are three that I would like to highlight—the rapidly expanding knowl-
edge base; demographic changes, leading most notably to the prevalence
of chronic disease in the population; and information technology.
The quality gap is attributable in part to the extraordinary increase
in medical knowledge and technology in recent decades. Starting in the
mid-1960s and continuing to this day, investment in biomedical re-
search, public and private, has increased in inšation-adjusted dollars
(National Institutes of Health 2000). These investments have clearly
paid off in terms of new knowledge, procedures, drugs, and medical de-
vices. For example, laparoscopic surgery has dramatically changed the
functional impact of many procedures, such as cholescystectomy; and
thrombolytic therapy has had a major inšuence on the treatment of per-
sons with heart attacks.
These innovations in medicine have signiŠcantly, and positively, af-
fected the health of the population. Yet they also pose a challenge for
health care practitioners who want to keep their skills up to date. Just
10 The Tanner Lectures on Human Values

reading about advances—let alone active training in or experience with

new techniques—is a daunting task. For instance, the number of cita-
tions reporting on randomized controlled trials has increased from an
average of 509 annually in the 1970s to over 10,000 annually today
(Chassin 1998). Although no practitioner needs to follow advances
across all areas of medicine and surgery, rapid expansion of knowledge is
occurring even within speciŠc areas. I have been told that the number of
randomized controlled trials published on diabetes in these same peri-
ods, the 1970s to the 1990s, increased from a few (less than ten) per year
to well over 150 per year.
The process of diffusing knowledge and new tools is also quite slow.
The lag between discovery of more efŠcacious forms of treatment and
their incorporation into routine patient care is in the Šfteen- to twenty-
year range (Balas and Boren 2000). And if we can’t keep up now, how
will we respond to the extraordinary advances that will emerge during
this new century?
Another consequence of advances in medical science, technology,
and health care delivery is that people are now living longer. Although
health care is by no means the only factor that affects morbidity and
mortality, innovations in medical science and technology have con-
tributed greatly to the increase in life expectancy. As a result of chang-
ing mortality patterns, those age sixty-Šve and over constitute an
increasingly large number and proportion of the U.S. population. In
1994 this age group accounted for approximately one in eight persons,
or 13 percent of the population (National Center for Health Statistics
1999). In 2030, when the large baby boom cohorts have become elders,
one person in Šve, or 20 percent, is expected to be in the sixty-Šve and
over age group. The very old, of course, are also growing in numbers and
as a proportion of the population.
The increasing likelihood of survival due to scientiŠc and techno-
logic advances will continue to result in larger proportions of people
with continuing morbidity and disability. Chronic conditions are now
the leading cause of illness, disability, and death and account for about
70 percent of care (Hoffman et al. 1996). Almost 100 million people in
the United States have one or more chronic conditions (Robert Wood
Johnson Foundation 1996).
These demographic changes have very important implications for
the organization of the health care delivery system, but we have yet to
address them in any signiŠcant way. In many cases we are organized for
[Richardson] Reconceiving Health Care to Improve Quality 11

and oriented to acute care, while the majority of health care resources are
now being devoted to the treatment of chronic disease. There is a dearth
of clinical programs with the infrastructure to provide the full comple-
ment of services needed by people with heart disease, diabetes, asthma,
or other common chronic conditions. Most people with these conditions
require care from multiple providers and across multiple settings over
long periods. Effective and efŠcient care of this population necessitates a
well-organized program, which would include:

• an interdisciplinary team,
• mechanisms for ongoing communication and coordination of ser-
vices across providers and settings,
• education programs and communication mechanisms directed at
patients, their families, and other informal caregivers,
• a formally organized care process designed to achieve best practice,
and
• the ability to measure both medical care process and patient
outcomes for purposes of quality improvement.

And providing state-of-the-art health care to a mostly chronically ill

population is further complicated by the fact that many have co-morbid
conditions. About 44 percent of those with a chronic condition have
more than one chronic condition (Robert Wood Johnson Foundation
1996).
Our challenge, then, is to move from the highly decentralized, often
cottage industry that we have now to one that can provide excellent ter-
tiary intervention and care, both primary and preventive care, and, in-
creasingly, the necessary “packages” of services that are needed for care
of the chronically ill. We fall particularly short in this last category.
It is true that the health care sector is more complex than other in-
dustries. I think it is probably the most complex industry. There are
large number of conditions and ailments, and in some important re-
spects each patient is unique.
But I would suggest that the heavy focus in health care on the clini-
cal needs of individual patients has in some ways blinded us to some of
the principles that have guided the development of better systems in
other industries. The so-called 80/20 rule should apply to health care.
12 The Tanner Lectures on Human Values

For example, a study at one health maintenance organization found that

78 percent of direct medical costs were attributable to just twenty-Šve
acute and chronic conditions and that three cardiovascular conditions
(ischemic heart disease, hypertension, and congestive heart failure) ac-
counted for 17 percent of these costs (Ray et al. 2000).
The Agency for Health Care Research and Quality in its work on the
Medical Expenditure Panel Survey identiŠes a limited number of “pri-
ority conditions” (Medical Expenditure Panel Survey 2000). These con-
ditions account for a sizable proportion of the national health burden.
They include cancer, diabetes, emphysema, and hypertension. If we
were to make it a priority to develop well-organized care programs based
on “best practices” for this limited number of conditions, we would be
well on our way to addressing many quality problems.
In following the 80/20 approach, we determine what work is rou-
tine and design a simple, standard, and low-cost process to perform
this work efŠciently and reliably. This not only results in safer, higher-
quality, and more reliable health care for common conditions but frees
up resources and the time of highly skilled clinicians to focus on the
more unusual, complex cases. Standardizing care processes doesn’t
mean “one size Šts all.” As I said a moment ago, patients are not the
same. They have a range of preferences, and some have special needs.
In other industries, we use what is called mass customization to stan-
dardize common services needed by many patients, while customizing
or tailoring other aspects of services to respond to particular prefer-
ences and needs.
As we struggle to address this challenge, we must also keep in mind
that the health care industry is changing at an extraordinary rate. Our
efforts to narrow the quality gap will be far more successful if they are
congruent with the ways in which the industry is being transformed by
information technology and consumerism. As the saying goes, it is eas-
ier to ride a horse in the direction it is going.
The Internet places us on the threshold of a change that is reshaping
virtually all aspects of society, including health care delivery. The Inter-
net supports rising consumerism, with greater demands for information
and convenience in all areas of commerce. The effect of these trends on
health care will likely be a fundamental transformation in how services
are organized and delivered and how doctors and patients interact with
each other.
To better understand how information technology can contribute to
[Richardson] Reconceiving Health Care to Improve Quality 13

quality improvement, our IOM Committee convened a special work-

shop. The participants identiŠed Šve key areas where information tech-
nology can make a difference.
The Šrst area is translating science into practice. Through more ef-
fective use of the Internet, we can help providers gain better access to the
medical science base. The Internet has opened up many new opportuni-
ties to make evidence, both primary publications and secondary analy-
ses, more accessible to clinicians. The efforts of the National Library of
Medicine, through Medline, are particularly promising. Medline con-
tains more than 9 million citations and abstracts to articles drawn
mainly from professional journals (Miller et al. 2000). In June 1997 the
Library of Medicine made Medline on the Web available free of charge.
Usage jumped about ten-fold, to 75 million searches annually (Lindberg
and Humphreys 1998).
Second, information technology also facilitates consumer access to
health information. Patients and their families will be far more effective
caregivers (and team members) if they are knowledgeable about their
health conditions, options for treatment, and expected outcomes. Some
77 million Americans retrieve health-related information annually
(Morrison 1999), but the volume of health-related information can be
overwhelming. There are some 61,000 web sites that contain informa-
tion on breast cancer and about 40,000 for diabetes (Boodman 1999;
National Research Council 2000). This information is of varying qual-
ity—some is incorrect and some is misleading. In 1998 the Library of
Medicine started Medline Plus, a web site for consumers. Medline Plus
includes information on more than 300 health topics and also contains
links to reputable web sites maintained by professional associations and
other government agencies.
The third area is the collection and sharing of clinical information.
Perhaps the single most important contribution of information technol-
ogy will be to supplement paper medical records, which are so often
illegible or incomplete. Handwritten orders are a major source of med-
ical errors. Paper records are often unavailable, which contributes to
many unnecessary services, especially repeat laboratory and radiology
services. I realize there are important conŠdentiality and data security
issues to be resolved, but we can and must work our way through them.
We shouldn’t deny patients the very important beneŠts of automated
clinical information, which include improved quality, safety, conve-
nience, and efŠciency.
14 The Tanner Lectures on Human Values

Fourth, information technology can help reduce errors by standard-

izing and automating certain decisions and by identifying errors before
they occur—errors such as adverse drug interactions. Computerized
drug prescribing has great potential to have a positive impact on dosing
calculations and scheduling, drug selection, screening for interactions,
monitoring and documenting adverse side effects, and other areas. Yet
comprehensive medication order entry systems have been implemented
in only a limited number of health care settings.
Fifth, information technology can change the way individuals re-
ceive care and interact with their providers. Instead of a $65 ofŠce visit
and a half-day off work, a ten-minute e-mail communication could
meet the patient’s needs. Similarly, patients will be able to go online to
get test results, inform their physicians about how they are doing, par-
ticipate in interactive disease management services, and receive after-
care instructions. Touch and face-to-face interaction will always be
important. The essence of high-quality health care is a “healing rela-
tionship.” But in many instances face-to-face encounters are neither
needed nor wanted by the patient or clinician.
Of course, there are challenges we must confront to take advantage
of the many beneŠcial applications that information technology (IT) has
to offer. First and foremost are privacy considerations. The public have
been given little information to help them evaluate the many beneŠts of
IT, while at the same time they have heard potential horror stories that
can come from automated personal health information. There is a very
real need to open the dialogue and inform public debate in this area. For
an interesting example of how public concerns about privacy can halt ef-
forts to advance our use of IT, we need only look back two years, when
the Department of Health and Human Services halted plans to establish
a unique patient identiŠer in response to public outcry over potential
violations of medical privacy (Goldman 1998).
Second, the lack of national standards for the collection, coding, and
sharing of data is also viewed by many as an impediment to moving for-
ward. The efforts of the National Committee for Vital and Health Sta-
tistics are very important in moving us forward in this area. Progress has
been slow.
Third, signiŠcant Šnancial investments in IT will be needed—far
greater than the current investments being made by most health care
organizations. Capital will be required to purchase and install new tech-
nology. The installation of new computerized systems often produces
[Richardson] Reconceiving Health Care to Improve Quality 15

temporary disruptions in the delivery of patient care. Considerable spe-

cialized training and education will be needed to help the workforce
adapt to a new environment. These capital decisions must be made in an
environment where beneŠts are hard to quantify. Unlike capital invest-
ments in new medical technology, which immediately generate rev-
enues under our predominantly fee-for-service system of payment, IT
investments to automate clinical data have only an indirect effect on the
bottom line.
Lastly, not all patients will take advantage of the opportunities af-
forded by information technology. There will be a need to operate the
“old” and the “new” delivery systems in parallel.
It is not possible to foresee all the new organizations, forces, tech-
nologies, needs, and relationships that will develop in the health care
system over the coming decade. The IOM Committee is attempting to
specify some general aims that the system should try to achieve and what
might be called some rules for the road. Our framework is based on an
understanding of systems that can self-organize to achieve a shared pur-
pose and improve by adhering to well-thought-out general rules.
Our report on twenty-Šrst-century health care will be released early
this spring, so this is still a work in progress; but I can share with you
some of our thinking. First, let me again emphasize that fundamental
change is needed. The American people should get a much higher re-
turn on their investment in health care than they currently do.
The IOM Committee has identiŠed six Aims for Improvement—six
dimensions of quality where, we believe, today’s health system func-
tions at far lower levels than it should. Health care should be:

• Safe: avoiding injuries to patients from the care that is intended

to help them.
• Effective: providing services based on scientiŠc knowledge to all
who could beneŠt and refraining from providing services not
likely to beneŠt (avoiding overuse and underuse).
• Patient-centered: providing care that is respectful and responsive
to individual patient preferences, needs, and values; and assuring
that patient values guide all clinical decisions.
• Timely: reducing waits and sometimes harmful delays for both
those who receive care and those who give care.
• EfŠcient: avoiding waste, including waste of equipment, supplies,
ideas, and energy; and
16 The Tanner Lectures on Human Values

• Equitable: providing care that does not vary in quality because

of personal characteristics such as gender, ethnicity, geographic
location, or socioeconomic status.

A health care system that achieves major gains in these six dimen-
sions would be a great beneŠt for patients.
The culture of medicine and the roles of physicians, nurses, and
other clinicians and the clinician/patient relationship are all likely to
change substantially over the coming decade.
To help guide this transition, we have been formulating ten “sim-
ple” rules in our IOM Committee work.
First, care should be based on continuous healing relationships. In
today’s health system the product of health care is the visit or hospital
episode. In the future the product of the health care Šeld will be the
“healing relationship.” In the current framework of health care delivery,
interaction and relationship are regarded more as a toll on health care
than as one of its goals or products. The system today often acts as if in-
teractions and relationships were an added burden for the real care
process. In the twenty-Šrst-century health system, interaction is not the
price of care, it is care. In practical terms, care based on continuous heal-
ing relationships means that the health care system should be respon-
sive at all times. It also means that care should be provided in many
forms, including over the Internet and by telephone, not just through
face-to-face visits.
Second is customization based on patient needs and values. I’ve al-
ready mentioned the concept of “mass customization” in other indus-
tries. In health care, customization means that the system of care should
be designed to meet the most common types of needs but have the capa-
bility to respond to individual patient preferences and choices.
Third is the patient as the source of control. Patients should be given
the necessary information and opportunity to exercise the degree of con-
trol they choose over health care decisions affecting them. Throughout
most of the twentieth century the physician has served as the principal
care provider, exercising a great deal of authority and autonomy. In the
twenty-Šrst century the role of the physician, in addition to providing
highly technical services, will more likely be that of a “care partner” and
coach. The notion is sometimes called “patient-centered care,” and it
means that control over care and choices resides more in the patient. To
[Richardson] Reconceiving Health Care to Improve Quality 17

be sure, it takes time for patients to be included as partners in care, and

they’ll often need to be coached in this new role. However, in settings
where this has occurred, its value has been shown by medical research.
In sixteen of twenty-one studies published in recent years, patients who
partnered with physicians were more likely to follow treatment advice
(Stewart 1995). This held true for a range of health problems, including
breast cancer, diabetes, hypertension, headaches, and gastrointestinal
disease. This rule is not intended to imply that patients should be forced
to share decision-making, only that it should be possible for them to ex-
ercise the degree of control they wish.
Fourth is shared knowledge and the free šow of information. Pa-
tients should have unfettered access to their own medical information
and to clinical knowledge. This rule goes hand in hand with the preced-
ing ones. It recognizes that information is the key to the patient-clini-
cian relationship and that in most instances the exchange of information
is the essence of the healing relationship.
Fifth is evidence-based decision-making. Patients should receive
care based on the best available scientiŠc knowledge. I’ve already dis-
cussed the importance of moving toward science-based practice. An-
other general rule is best practice based on systematically acquired
evidence. In the twentieth century a commitment to autonomy of clini-
cal decision-making has been a fundamental health care value. In the fu-
ture a commitment to excellence—standardization to the best-known
method, given the patient’s circumstances—should be preeminent.
Sixth is safety as a system property. Threats to patient safety are the
end result of complex causes such as faulty equipment, system design,
and the interplay of human factors such as fatigue, limitations on mem-
ory, and distraction. The way to improve safety is to learn about causes
of error and to use this knowledge to design systems of care to prevent
error when possible, to make visible those errors that do occur (so that
they can be intercepted), and to mitigate the harm done when an error
does reach the patient.
Seventh, transparency is necessary. The health care system should
make information available to patients and their families that allows
them to make informed decisions when selecting a health plan, hospi-
tal, or clinical practice or choosing between alternative treatments.
Eighth is anticipation of needs. The health system should anticipate
patient needs, rather than simply reacting to events. Our current health
care system works largely in a reactive mode. The twenty-Šrst-century
18 The Tanner Lectures on Human Values

system should organize health care to predict and anticipate need based
on knowledge of patients, local conditions, and the natural history of
illness.
Ninth, waste should be continuously reduced. The health system
should not waste resources or patient time. Members of the committee
do not believe that increased value will come by stressing the current
system—that is, by asking people to work harder, faster, and longer.
Rather, increased value will come from systematically developed strate-
gies that focus on the six Aims for Improvement of the health care sys-
tem.
Tenth is cooperation among clinicians. Clinicians and institutions
should actively collaborate and communicate to assure appropriate ex-
change of information and coordination of care. The current system
shows too little cooperation and teamwork. Each discipline and type of
organization tends to defend its authority at the expense of the total sys-
tem’s function—a problem known as suboptimization. Patients suffer
through lost continuity, redundancy, excess costs, and miscommunica-
tion. The new rule asserts that cooperation in patient care is more im-
portant than professional prerogatives and roles.
Although I have used the term “simple rules” to refer to the guiding
principles, our committee recognizes that adhering to these rules will
be very challenging. Some of these rules bump up against professional
norms and behaviors, while others will require major cultural changes
in health care organizations. And there are environmental barriers that
must be overcome as well. Legal liability, payment policies, regulatory
systems, and other external forces that inšuence health care must be
changed to encourage the types of behavior consistent with quality im-
provement.
How do we get from here to there? Changes will be needed at two
levels: the care delivery and the environmental level.
Although we cannot foresee the range of new organizations, rela-
tionships, and technologies that will emerge over the coming decade,
there are certain functions that organizational structures—whether vir-
tual or bricks and mortar—will need to perform. Quality health care
cannot be delivered through a cottage industry any longer. Well-de-
signed care processes that are based on sound clinical and engineering
principles and make the best use of information technology and human
resources are essential.
Health care today is more and more an interaction between the sys-
[Richardson] Reconceiving Health Care to Improve Quality 19

tem and a person who needs help from that system. To be sure, the
physician plays a critical role, but his/her effectiveness is increasingly
determined by the characteristics of the system within which practice
takes place.
Changes will also be needed in the environment of care. Current
payment policies do not adequately encourage or support the provision
of quality health care. Although payment is not the only factor that in-
šuences provider and patient behavior, it is a very important one. Too
little attention has been paid to the careful analysis and alignment of
payment incentives with quality improvement. The current health care
environment is replete with examples of payment policies that work
against the efforts of clinicians, health care executives, and others to im-
prove quality. For example, a safety improvement initiative that reduces
adverse drug events may also reduce payments for physician visits or
shift hospital patients into Diagnosis Related Group categories that are
less complicated and generate less revenue. Similarly, under current
visit-based payment systems, clinicians have little incentive to commu-
nicate with patients through e-mail.
There will also need to be changes in health professional education
and training programs. The traditional emphasis in clinical education,
particularly medical education, has been on teaching a “core of knowl-
edge,” much of it focusing on the basic mechanisms of disease and
patho-physiologic principles. Given the expansiveness and dynamic na-
ture of the science-base in health care, this approach should be expanded
to include knowledge management as a means to support clinical deci-
sion-making. Similarly, as more care is provided in teams, more oppor-
tunities for interdisciplinary training should take place.
In order for innovative programs to šourish, our regulatory environ-
ments will also need to adapt. In general, regulation in this country can
be characterized as a dense patchwork of federal and state requirements
that are slow to change. One of the key regulatory issues affecting the
workforce and how it is used is licensure and scope of practice acts, im-
plemented at the state level. One effect of these acts is to deŠne how the
health care workforce is deployed. Although scope of practice acts are
motivated by the desire to establish minimum standards to ensure the
safety of patients, they also can make it difŠcult to use alternative ap-
proaches to care delivery, such as telemedicine, e-visits, nonphysician
providers, and multidisciplinary teams.
These are but a few examples of some of the far-reaching changes
20 The Tanner Lectures on Human Values

that will be necessary. In short, the need for leadership has never been
greater—organizational leadership, physician/clinician leadership, and
community participation. The transformation of the health care system
will not be an easy process. But the potential beneŠts are tremendous.
This Šeld has changed over time as the society has changed and as its
capabilities have developed. It was only a century ago that care got com-
plex enough, capital-intensive enough, and successful enough to war-
rant the middle class even wanting to be in a hospital. Physicians who
volunteered in indigent clinics at hospitals applied for privileges. What
was the privilege? To admit their private patients to the hospital and
provide them care.
Today we’re at the dawn of a new era of complexity, capital require-
ments, and potential effectiveness. The privilege for us in this new era
will be to support development of new sets of organizational and Šnan-
cial capacities, just as we did during the last century, to make the most
of these opportunities for all of our people.

References

Balas, E. Andrew, and Suzanne A. Boren. 2000. Managing Clinical Knowledge

for Health Care Improvement. In Yearbook of Medical Informatics, pp. 65–70.
Bethesda, Md.: National Library of Medicine.
Boodman, Sandra G. 1999. Medical Web Sites Can Steer You Wrong: Study
Finds Erroneous and Misleading Information on Many Pages Dedicated to
a Rare Cancer. Washington Post, August 10, Health-Z07.
Chassin, Mark R. 1997. Assessing Strategies for Quality Improvement. Health
Affairs 16, no. 3: 151–61.
———. 1998. Is Health Care Ready for Six Sigma Quality? Milbank Quarterly
76, no. 4: 575–91.
Chassin, Mark R., Robert W. Galvin, and the National Roundtable on Health
Care Quality. 1998. The Urgent Need to Improve Health Care Quality.
Journal of the American Medical Association (JAMA) 280, no. 11: 1000–1005.
Classen, David C., Stanley L. Pestonik, Scott Evans, and John P. Burke. 1991.
Computerized Surveillance of Adverse Drug Events in Hospital Patients.
JAMA 266, no. 20: 2847–51.
Goldman, Janlori. 1998. Protecting Privacy to Improve Health Care. Health
Affairs 17, no. 6: 47–60.
Hoffman, Catherine, Dorothy P. Rice, and Hai-Yen Sung. 1996. Persons with
Chronic Conditions: Their Prevalence and Costs. JAMA 276, no. 18:
1473–79.
[Richardson] Reconceiving Health Care to Improve Quality 21

Institute of Medicine. 1999. Ensuring Quality Cancer Care. Ed. Maria Hewitt
and Joseph V. Simone. Washington, D.C.: National Academy Press.
———. 2000. To Err Is Human: Building a Safer Health System. Ed. Linda T.
Kohn, Janet M. Corrigan, and Molla S. Donaldson. Washington, D.C:
National Academy Press.
———. 2001. Crossing the Quality Chasm: A New Health System for the Twenty-
Šrst Century. Washington, D.C.: National Academy Press.
Leape, Lucian L., Troyen A. Brennan, Nan M. Laird, et al. 1991. The Nature of
Adverse Events in Hospitalized Patients: Results of the Harvard Medical
Practice Study II. JAMA 324, no. 6: 377–84.
Lindberg, Donald A. B., and Betsy L. Humphreys. 1998. Updates Linking Ev-
idence and Experience—Medicine and Health on the Internet: The Good,
the Bad, and the Ugly. JAMA 280, no. 15: 1303–4.
Medical Expenditure Panel Survey. 2000. MEPS HC-006R: 1996 Medical
Conditions. Online. Available at [Link]
(accessed December 7, 2000).
Miller, Naomi, Eve-Marie Lacroix, and Joyce E. B. Backus. 2000. MEDLINE-
plus: Building and Maintaining the National Library of Medicine’s Con-
sumer Health Web Service. JAMA 88, no. 1: 11–17.
Morrison, J. Ian. Healthcare in the New Millennium: The Promise of the In-
ternet. 1999. New York: Presentation at Internet Health Care II: Health
Care in Transition—Preparing for an Interactive Future, October 12.
National Academy for State Health Policy. Forthcoming. State Systems to Track
Adverse Events. Portland, Maine: National Academy for State Health Policy.
National Center for Health Statistics. 1999. Health, United States, 1999: With
Health and Aging Chartbook. Hyattsville, Md.: U.S. Government Printing
OfŠce.
National Institutes of Health. 2000. An Overview. Online. Available at
[Link] (accessed August 11,
2000).
National Research Council. 2000. Networking Health: Prescriptions for the Inter-
net. Washington D.C.: National Academy Press.
Phillips, David P., Nicholas Christenfeld, and Laura M. Glynn. 1998. Increase
in U.S. Medication-Error Deaths between 1983 and 1993. Lancet 351:
643–44.
President’s Advisory Commission on Consumer Protection and Quality in the
Health Care Industry. 1998. Quality First: Better Health Care for All
Americans. Online. Available at [Link]
Šnal/ (accessed September 9, 2000).
Ray, G. Thomas, Tracy Lieu, Bruce Fireman, et al. 2000. The Cost of Health
Conditions in a Health Maintenance Organization. Medical Care Research
and Review 57, no. 1: 92–109.
22 The Tanner Lectures on Human Values

Robert Wood Johnson Foundation. Chronic Care in America: A Twenty-Šrst Cen-

tury Challenge. Princeton, N.J.: Robert Wood Johnson Foundation, 1996.
Available at [Link]
Schuster, Mark A., Elizabeth A. McGlynn, and Robert H. Brook. 1998. How
Good Is the Quality of Health Care in the United States? Milbank Quarterly
76, no. 4: 517–63.
Schuster, Mark A., Elizabeth A. McGlynn, Cung B. Pham, Myles D. Spar, and
Robert H. Brook. 2001. Appendix A: The Quality of Health Care in the
United States: A Review of Articles since 1987. In Institute of Medicine,
Crossing the Quality Chasm: A New Health System for the Twenty-Šrst Century.
Washington, D.C.: National Academy Press.
Soumerai, S. B., T. D. McLaughlin, E. Hertzmark, G. Thibault, and L. Gold-
man. 1997. Adverse Outcomes of Underuse of Beta-Blockers in Elderly
Survivors of Acute Myocardial Infarction. JAMA 277: 115–21.
Stewart, Moira A. 1995. Effective Physician-Patient Communication and
Health Outcomes: A Review. Canadian Medical Association Journal 152, no.
9: 1423–33.
Thomas, Eric J., David M. Studdert, Helen R. Berstin, et al. 2000. Incidence
and Types of Adverse Events and Negligent Care in Utah and Colorado.
Medical Care 38, no. 3: 261–71.