Sie sind auf Seite 1von 40






Rights were not granted to include this image in electronic media.

Please refer to the printed report.



. . . . .T. . H. . .E. . . . .
PROJECT PARTICIPANTS Organizational affiliations for identification purposes only; affiliations current as of 2006

George J. Agich, PhD *Arnold Farley, PhD Margaret O’Kane, MHA

Bowling Green State University Centers for Medicare and Medicaid National Committee for Quality
Bowling Green, OH Services Assurance
Baltimore, MD Washington, DC
Mary Ann Baily, PhD
The Hastings Center *J. Michael Fitzmaurice, PhD Paul Schyve, MD
Garrison, NY Agency for Healthcare Research and Joint Commission on Accreditation
Quality of Healthcare Organizations
Paul Batalden, MD Rockville, MD Oakbrook Terrace, IL
Dartmouth Medical School
Hanover, NH *Ellen Fox, MD Theodore Speroff, PhD
National Center for Ethics in Vanderbilt University Medical Center
Nancy Berlinger, PhD, MDiv Health Care, Veterans Health Nashville, TN
The Hastings Center Administration
Garrison, NY Washington, DC Sean Tunis, MD, MSc
*Melissa Bottrell, MPH, PhD Brent James, MD, MSTAT San Francisco, CA
National Center for Ethics in Intermountain Institute for Health
Health Care, Veterans Health Care Delivery Research Matthew K. Wynia, MD, MPH
Administration Salt Lake City, UT American Medical Association
Seattle, WA Institute for Ethics
*Stephen F. Jencks, MD, MPH Chicago, IL
*Michael Carome, MD Centers for Medicare and Medicaid
DHHS Office for Human Research Services *Note: The authors of this report are re-
Protections Baltimore, MD sponsible for its content. Statements in
Rockville, MD this report should not be construed as
Bruce Jennings, MA endorsement by the Agency for Health-
David Casarett, MD, MA Center for Humans and Nature care Research and Quality or the U.S De-
Center for Health Equity Research New York, NY partment of Health and Human Services.
and Promotion
Philadelphia VA Medical Center Robert J. Levine, MD The federal government employees list-
Philadelphia, PA Yale University ed above participated in the meeting dis-
New Haven, CT cussions in an advisory capacity. Howev-
Janet Corrigan, PhD er, in addition to the preceding dis-
National Quality Forum Joanne Lynn, MD, MA, MS claimer, it should be further emphasized
Washington, DC The RAND Corporation that the statements in this report should
Arlington, VA not be construed as endorsement by the
Linda Cronenwett, PhD, RN, Department of Veterans Affairs or any
FAAN Karen Maschke, PhD other department of the U.S. govern-
University of North Carolina at The Hastings Center ment, nor do they represent the policies
Chapel Hill, School of Nursing Garrison, NY of any federal departments, agencies, or
Chapel Hill, NC offiices.
Ethel Mitty, EdD, RN
Frank Davidoff, MD, MACP New York University
Institute for Healthcare College of Nursing
Improvement New York, NY
Wethersfield, CT
Robyn Y. Nishimi, PhD This project was funded by the
Nancy Neveloff Dubler, LLB National Quality Forum Agency for Healthcare Research
Montefiore Medical Center Washington, DC and Quality, grant #1R13HS13369.
Bronx, NY



owerful forces of change are at work within the Federal agencies with responsibilities in this area have
American health care system. The public debate disagreed on where the interface between medical research
concerning health care financing and access to in- and QI lies and how it should be handled. (See Box 1 for
surance coverage is intensifying. But below the surface of a particularly dramatic example of such a conflict.) The
the media and policy debate about cost and access, a qui- strict ethical rules of oversight, regulation, and patient
eter but perhaps more significant process of change is consent for human subjects research, including the re-
under way: the transformation of health care management quirement for institutional review board (IRB) approval,
and delivery—indeed, health professional work itself— have significant implications for the feasibility and cost of
through health care quality improvement. pursuing QI activities. More specifically, the mechanism
The innovative, interdisciplinary quality improvement developed to govern ethical conduct in one important
(QI) movement has begun to significantly upgrade deliv- area—human subjects research—could have the perverse,
ery of health care in the United States. Taking its cue from if unintended, consequence of interfering directly with an
reform approaches in other industries, and driven espe- equally important ethical imperative in another area—
cially by studies indicating a shockingly widespread inci- that is, unceasing efforts by health care professionals to
dence of medical errors and a striking lack of consistency make clinical care safer and more effective. The current
in the standard of care patients receive in different facili- state of uncertainty about what is ethically and legally re-
ties and from different practitioners, the QI movement quired to safeguard participants in QI activities has al-
has arrived in health care. Using knowledge gained from ready become a disincentive to engage in QI, making it
the disciplines of medicine, nursing, health care manage- more difficult to bring about the system transformation
ment, and medical and health services research, it at- urgently needed if health care is to be made better and
tempts to mobilize people within the health care system to safer for patients.
work together in a systematic way to improve the care In 2002 The Hastings Center began a project to ad-
they provide. In this work, discipline-specific knowledge dress these issues and to investigate more generally the
is combined with experiential learning and discovery to ethical and value issues that arise in the theory and prac-
make improvements. tice of quality improvement in health care. The project, ti-
Ethical issues arise in QI because attempts to improve tled “The Ethics of Improving Health Care Quality and
the quality of care for some patients may sometimes inad- Safety,” was funded by grant #1R13HS13369 from the
vertently cause harm, or may benefit some at the expense Agency for Healthcare Research and Quality (AHRQ).
of others, or may waste scarce health care resources. Ethi- The Hastings Center project assembled a group of ex-
cal issues also arise because some activities aimed at im- perts from a number of interrelated fields and disciplines
provement have been interpreted as a form of medical re- involved in health care quality improvement, including
search in which patients are used as subjects. If this inter- medicine, nursing, law, social science, health care manage-
pretation is correct, QI would come under the same com- ment, medical editing and publishing, health policy and
plex review and regulatory requirements that have been regulation, health services research, and bioethics. The
set up to govern biomedical and other types of research. project’s goals were: (1) to develop an ethical framework
But is this type of regulation necessary, given what QI in- that can be applied to ethical issues arising in quality im-
volves? Is it the most effective and reasonable way to regu- provement practice, (2) to make practical policy recom-
late QI to ensure that it is carried out in an ethical fash- mendations for quality improvement oversight, and (3) to
ion? These are important questions, both conceptually help promote a constructive dialogue both within and
and practically. Thus far, however, relatively few attempts outside the quality improvement community on pertinent
have been made to address QI from an ethical perspective, ethical issues.
and the interface between research and quality improve- Over a two-year period, the project group met four
ment has not been adequately explored or defined. times for two-day working sessions. Project members and

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S3
1. The Pittsburgh ESRD Case

In October 2000, a nephrologist coauthored OHRP responded that the design of the quality
an article about a project to improve the dialysis improvement project met the definition of
care delivered to patients in Medicare’s End human subjects research and that the CMS de-
Stage Renal Disease (ESRD) program.1 Some termination that the project was exempt from
time after the article appeared, the university at IRB review “was not made in collaboration with
which he held a faculty appointment notified OHRP.” OHRP and CMS have had discussions
him that an audit of faculty publications had and exchanged memoranda as to whether this
identified this project as a quality improvement and other quality improvement activities meet
effort that met the definition of human subjects the regulatory definition of human subjects re-
research but had not undergone IRB review. The search and require IRB review; however, these is-
nephrologist responded that he had participated sues have not yet been resolved. This issue has
in the project as the chair of the local ESRD caused concern among other volunteer physi-
Network’s Medical Review Board, that the cians in the eighteen ESRD Networks about the
ESRD Network conducted the project under implications for ongoing quality improvement
contract to the Center for Medicare and Medi- projects and the possible impact if their conduct
caid Services (CMS), and that the CMS scientif- is found to violate the regulations for protecting
ic officer overseeing the project had told him human research subjects.
that, as a CMS-directed quality improvement
project, it was not subject to oversight by the 1. P.M. Palevsky et al., “Improving Compliance with
university’s IRB. Later, CMS reaffirmed this the Dialysis Prescription as a Strategy to Increase the De-
livered Dose of Hemodialysis: An ESRD Network 4
opinion in a letter to the university. The univer- Quality Improvement Project,” Advances in Renal Re-
sity submitted the dispute to the Department of placement Therapy 7, no. 4, supplement 1 (2000): S21-
Health and Human Service’s Office for Human S30.
Research Protections (OHRP) for review.

other invited guests presented and debated the contacts with persons and organizations interested
findings of their own research and drafts of com- in the ethics of QI.
missioned papers focusing on ethical issues in This report presents the analysis, findings, and
quality improvement. Those papers will be pub- conclusions that emerged from the deliberations
lished in an edited volume in 2007. In addition to of the project participants and staff research con-
our deliberations with project participants, we ducted for this project. The authors are Mary Ann
have also made presentations at professional and Baily, Melissa Bottrell, Joanne Lynn, and Bruce
academic meetings and communicated electroni- Jennings. This report is not a consensus docu-
cally with a larger group interested in health care ment in the sense that each project participant
system improvement by setting up a moderated e- agrees with it in all details, but we have done our
mail listserv. Following the publication of this re- best to present an accurate and faithful reflection
port, The Hastings Center will continue to serve of the thinking of the group as a whole, and this
as a resource in the process of translating the con- analysis certainly would not have been possible
clusions and recommendations of the project into without the benefit of their insights and expertise.
practical policies through the listserv, presenta-
tions at professional and academic meetings, pro- —Mary Ann Baily, Melissa Bottrell,
duction of additional written materials, and direct Joanne Lynn, Bruce Jennings



Q uality improvement in health care takes many meet accepted ethical standards for protection of human
forms, ranging from changes in financing to research subjects. The regulations that are relevant to this
reforms in professional education to invest- report are codified at Title 45 CFR 46 as Department of
ments in new facilities and equipment. In this Health and Human Services, Protection of Human Sub-
report, we are concerned with the form of improvement jects. The core requirements (Subpart A) are that individ-
that occurs through clinical and managerial innovations uals be selected equitably to participate in research; that
and adaptations in the delivery of care. These changes research subjects give full, voluntary written consent; and
have always been an integral part of health care delivery, that an IRB review proposed research and approve it only
but in the past they were often introduced informally and if risks to subjects have been minimized and are reason-
idiosyncratically, without careful attention to all of their able in relation to anticipated benefits.2 There are addi-
effects. In recent years, people in health care have begun tional requirements (Subparts B, C, and D) for research
to use many new formal, explicit methods, some of which that involves especially vulnerable subjects, such as preg-
were developed in other industries, to make the process of nant women, prisoners, and children. The DHHS regula-
continual adjustment more self-reflective and systematic, tions do not cover non-federally funded research, but or-
and thus increase the likelihood that it produces positive ganizations engaged in research are encouraged to provide
change. In other settings, this approach to improving formal assurances to the government that all their human
quality is often referred to as “QI.” subject research will comply with the DHHS regulations
When we use the term “QI” in this report, we mean whatever the funding source. Also, although Subpart A
systematic, data-guided activities designed to bring about explicitly exempts some categories of human subject re-
immediate, positive changes in the delivery of health care
in particular settings. While QI uses a wide variety of
methods, they all involve deliberate actions to improve Although systematic empirical information on
care, guided by data reflecting the effects. Depending on
the activity, QI can look like a type of practical problem the quality of care is limited—which is itself
solving, an evidence-based management style, or the ap-
plication of a theory-driven science of how to bring about surprising—what is available is disturbing.
system change. Introducing QI methods often means en-
couraging people in the clinical care setting to use their search from the regulations, the DHHS Office for
daily experience to identify promising ways to improve Human Research Protections (OHRP) has issued guid-
care, implement changes on a small scale, collect data on ance stating that researchers should not be allowed to de-
the effects of those changes, and assess the results. The cide on their own that their projects are exempt. Many in-
goal is to find interventions that work well, implement stitutions require determination of exempt status to be
them more broadly, and thereby improve clinical practice. made by an IRB. In practice, a substantial proportion of
Alternatively, a QI activity might begin with a review of both federally funded and nonfederally funded human
aggregate data at the patient, provider, clinical unit, or or- subject research, especially biomedical research, must be
ganizational level to identify a clinical or management submitted to an IRB. (For a brief summary of the DHHS
change that can be expected to improve care. The change regulations, see Box 2.)
is made, the effects are monitored, and conclusions are Discussion of ethical issues raised by QI has tended to
drawn about whether the change should be made perma- center around the legal question, “Is this QI project
nent. QI can also mean collecting data from multiple or- ‘human subjects research’ as defined by federal regulations
ganizations, analyzing it to understand what drives posi- and therefore subject to IRB review and the regulatory re-
tive change, and using the results to design and imple- quirements for informed consent?’’ The case outlined in
ment a strategy to achieve a specific improvement across Box 1 is but one example of many in an ongoing contro-
organizations. At its heart, QI is a form of experiential versy in the literature, in IRBs, and in organizations doing
learning and discovery.1 QI about whether QI is research and therefore covered by
QI work is data-guided, usually involves human par- the regulations.
ticipants, and sometimes uses methods that are also used Framing the issue in this way may not be the best way
in medical research. Thus, it is not surprising that the to proceed, however. The goal is to ensure that QI activi-
issue of its relationship to federal regulations governing re- ties are carried out in an ethical manner and that their
search with humans has arisen. These regulations were put human participants are treated appropriately, however the
in place to ensure that federally funded research projects activities are planned and structured and whatever the ac-

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S5
2. Summary of DHHS Policy for the Protection of Human Subjects (Subparts A-D)

Department of Health and Human Services policy monitoring, when appropriate; provisions are made for
sets out a complex set of provisions for the protection of subject privacy and confidentiality of data, when ap-
human research subjects. It: propriate (45 CFR 46.111)
 extends to research funded by the Department of
 requires IRBs to ensure that additional safeguards
Health and Human Services and to nonfederally fund-
are included in a study when participants will likely be
ed research conducted at or in conjunction with an in-
vulnerable to coercion or undue influence (45 CFR
stitution that agrees through its federal-wide assurance
to extend the regulations to all research regardless of
funding source. Six categories of research activities are
 prohibits institutional officials from approving re-
exempt from the regulations (45 CFR 46.101(b)(1)-
search that has not received IRB approval (45 CFR
 requires prior review and approval of nonexempt re-
 permits IRBs to suspend or terminate research ap-
search by a duly constituted institutional review board
proval when a study is not conducted in accordance
(IRB), and continuing IRB review of ongoing research
with IRB requirements or is associated with unexpect-
(45 CFR 46.103)
ed serious harm to participants (45 CFR 46.113)
 permits expedited review (conducted by IRB chair-
 requires that specific information be provided to re-
person or by one or more experienced reviewers) when
search participants to satisfy the general requirements
certain criteria are met (45 CFR 46.110 and Guidance
for informed consent (45 CFR 26.116)
on the Use of Expedited Review Procedures)
 permits IRBs to approve a consent procedure that
 requires seven items to be satisfied for IRB approval
omits or alters some or all of the elements of informed
of research: risks to subjects are minimized; risks are
consent when certain criteria are met (45 CFR 46.116)
reasonable in relation to anticipated benefits; subjects
are selected equitably; informed consent will be ob-
 requires additional protections for pregnant women,
tained from prospective subjects or their legally autho-
human fetuses and neonates (45 CFR 46.201), prison-
rized representatives; there will be written documenta-
ers (45 CFR 46.301), and children (45 CFR 46.401)
tion of informed consent; the trial will receive safety

tivities are called. Focusing on IRB review and the defini- sponsibility for improving the quality of care. We con-
tion of research may implicitly suggest that IRB review clude that engaging in quality improvement is not purely
would be adequate and that it is the only way to protect discretionary; health professionals, managers, delivery or-
QI participants. This is an overly narrow perspective. In ganizations, patients, and government all have an ethical
this report, we explore the similarities and differences be- responsibility to cooperate with one another to improve
tween research and QI, but rather than focus entirely on the quality of care. Section II compares and contrasts QI
“Is QI research?” and “IRB review, yes or no?” we step and research in terms of each activity’s goals, methods,
back to look at the bigger picture. We ask: “What is the role in the health care system, and impact on human par-
place of QI in the operation of the health care system?” ticipants. We use the principles of research ethics that un-
“What makes a QI activity ethical?” and “What institu- derlie the current system for protecting human research
tional arrangements should be in place to ensure that spe- subjects as a base from which to explore and define the re-
cific QI activities meet ethical requirements?” Only after quirements for ethical conduct of QI activities. We then
that do we return to the question of how QI activities consider the use of IRBs for ethical oversight of the con-
should relate to IRB review and other procedures for the duct of QI and conclude that, for a variety of reasons,
protection of human subjects in research. IRBs as currently constituted are not appropriate for this
purpose. Section III discusses the institutional arrange-
Organization of the Report ments that should be in place to ensure that QI activities
meet ethical requirements. Section IV addresses imple-
S ection I explains why quality improvement is impor-
tant in health care and discusses the role of QI meth-
ods in the management of health care delivery. We then
mentation, and Section V briefly summarizes the conclu-
sions and policy recommendations of the report.
consider the underlying ethical question of who has re-


I. The Place of Quality Improvement in Health Care

any Americans have a rosy picture of the quali- quality problems were seen in preventive, acute, and
ty of their health care system. They believe that chronic care, in the care provided in different kinds of
it provides a generally high standard of care that health care facilities, in care paid for by different kinds of
is continuously updated as new research findings come health insurance, in the care received by different age
out, so that it remains the best in the world. They recog- groups, and in rural and urban settings.9 The studies they
nize, of course, that there are some access problems (espe- reviewed document unnecessary surgery; inappropriate
cially for the uninsured and underinsured) and some use of medications; failure to perform standard screening
below-par facilities and professionals—no health care sys- tests; inadequately controlled asthma, diabetes, and hyper-
tem is perfect—but overall, the sense is that the quality of tension; and significant departures from recommended
care is excellent. Unfortunately, this view is too optimistic. levels of care for patients with cardiovascular disease. A
Recent work by individual scholars and by organizations more recent study examined the care received by a large
such as the Institute of Medicine (IOM) and the Agency sample of adults in twelve cities and documented a variety
for Healthcare Research and Quality (AHRQ) has provid- of quality problems, finding that overall, participants re-
ed detailed, compelling evidence of serious problems with
the quality of American health care.3
Since quality is a complex, multifaceted concept, as- 3. Quality Framework
sessing just how well the American health care system is
doing is not easy. It requires the development of a concep- Quality is defined in the Institute of Medicine re-
tual framework for understanding quality and the transla- port, Crossing the Quality Chasm, as the degree to
tion of that framework into practical measures that can be which health services for individuals and populations
applied in specific contexts. The framework developed by increase the likelihood of desired health outcomes and
the Institute of Medicine in several influential reports de- are consistent with current professional knowledge.1
fines quality as “the degree to which health services for in-
dividuals and populations increase the likelihood of de- The report describes the following characteristics of
sired health outcomes and are consistent with current pro- quality:
fessional knowledge.”4 The core dimensions of quality are
identified as safety, effectiveness, patient-centeredness, and Safety: avoiding injuries to patients from the care
timeliness.5 Two additional aspects of care that cut across that is intended to help them
these dimensions are equity and efficiency.6 (See Box 3 for
definitions of these six quality characteristics.) Effectiveness: providing services based on scientific
The status of efficiency in the quality framework is knowledge to all who could benefit, and refraining
controversial. In Crossing the Quality Chasm, the Institute from providing services to those not likely to benefit
of Medicine Committee on the Quality of Health Care (avoiding underuse and overuse, respectively)
considers it a dimension of quality, but another IOM
committee, charged with translating the IOM’s quality Patient-centeredness: providing care that is respect-
framework into practical measures, does not, arguing that ful of and responsive to individual patient preferences,
efficiency is “an important goal of the health care system needs, and values, and ensuring that patient values
that is related to, but conceptually different from quality guide all clinical decisions
of care.”7 In this report, we work from the position that
increasing efficiency is often a goal of quality improve- Timeliness: reducing waits and sometimes harmful
ment activities, since the developers of QI methods have delays for both those who receive and those who give
traditionally considered efficiency an appropriate QI con- care
cern, and activities involving changes designed to produce
the same quality of care with fewer resources constitute an Equity: providing care that does not vary in quality
important QI category.8 because of personal characteristics such as gender, eth-
Although systematic empirical information on the nicity, geographic location, and socioeconomic status
quality of care is limited—which is itself surprising, given
the size and importance of the health care sector—what is Efficiency: avoiding waste, including waste of
available is disturbing. In a comprehensive review of the equipment, supplies, ideas, and energy
literature on quality of care from 1987 to 1998, Schuster
and colleagues found substantial evidence of overuse, un- 1. Institute of Medicine, Crossing the Quality Chasm (Wash-
deruse, and misuse of care in the United States. Serious ington, D.C.: National Academy Press, 2001).

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S7
ceived only a little over half of recommended care.10 There could have been made somewhat differently in that set-
is also ample evidence of equity problems: the quality of ting, and that have been made differently in other settings.
care varies in ways systematically correlated with ethnicity, Within this universe of change-related activity is a sub-
geographic location, and socioeconomic status.11 set of deliberate efforts to make positive changes in the deliv-
To help remedy the lack of information and make it ery of clinical care. Many of these “designed changes” come
easier to monitor quality over time, the IOM and AHRQ about by “just doing it”; someone decides that a change in
have developed a comprehensive set of quality measures, clinical practice or organizational arrangements seems like
and in 2003, AHRQ published the first in a projected se- a good idea and the change is made. Such innovation and
ries of periodic national quality reports. The report pro- adaptation is an intrinsic part of clinical and managerial
vided a baseline of quality information, including evi- practice. Health care practitioners must tailor a general
dence on the tremendous variation in quality of care standard of practice to individual patients, relying on their
across geographic areas and across individual care set- clinical knowledge, their unique knowledge of each pa-
tings.12 The 2004 and 2005 National Healthcare Quality tient, and the characteristics and capabilities of the local
Reports found improvement in some measures, but deteri- context in which care is being delivered. Managers must
oration on others, and concluded that the gap between the exercise judgment about how resources in a particular set-
best possible care and actual care remained large.13 AHRQ ting should be organized to carry out treatment plans for
also publishes a companion series, the annual National the patients being served.
Healthcare Disparities Report. The 2003, 2004, and 2005 When this innovation and adaptation is undertaken in
reports provided baseline and follow-up information on a systematic, data-guided way, it becomes what we are call-
the pervasive racial, ethnic, and socioeconomic disparities ing QI. The category of QI includes a wide variety of ac-
in health care access and quality in the United States.14 Fi- tivities and cannot be defined by any one method or pro-
nally, for an international comparison, in a recent survey, cedure.
one-third of patients with health problems in the United QI is closely related to clinical practice, and in fact,
States reported experiencing medical, medication, or test much of QI is simply good clinical care combined with
errors, the highest rate for the six countries included in the systematic, experiential learning. Individual practitioners
survey (the others were Australia, Canada, Germany, New are constantly learning by doing and taking steps to im-
Zealand, and the United Kingdom).15 prove their own practice. Physicians may develop personal
The more we learn about current quality, the clearer it templates for entering information in medical records and
is that safe, effective, patient-centered, timely, equitable, for tracking the lab tests they have ordered; surgeons may
and efficient health care will not happen automatically. work on their surgical techniques to reduce the time pa-
Rather, it requires and will continue to require systematic tients spend anesthetized; nurses may refine their skin care
and self-conscious management of health care delivery ex- methods to reduce the incidence of bedsores.
pressly directed at improving care. This means not only Very quickly, however, the effort of an individual clini-
the implementation of specific changes, but also the trans- cian to improve his or her practice becomes an issue for
formation of the culture of health care delivery into a cul- others working in the same environment. In health care
ture that is committed to continuous quality improve- facilities today, most clinical care is delivered to patients
ment. on a team basis, and the ability of the team to deliver good
care depends on the characteristics of the administrative
QI Activities: Part of Normal Health Care infrastructure and procedures that are in place. In other
Operations words, both patients and providers are part of systems of
care. A key insight of those promoting QI methods is that

E fforts to improve quality take place in a health care en-

vironment that is continually experiencing change.
Some of this change occurs in response to fluctuations in
quality and safety are largely systems issues, and under-
standing the interdependencies and relationships within
systems is at the heart of QI work.16 In systems, manage-
market conditions as, for example, the age distribution of ment practices may be as significant to good outcomes as
the population evolves, new diseases emerge, or careers in the clinical practices of individual care providers, so learn-
health care are considered more or less attractive. Some of ing by doing in management is also important. Many QI
the change occurs in response to scientific progress as, for activities involve clinicians cooperating with each other
example, a known disease is better understood, a new drug and with management and support staff around improve-
is developed, or new diagnostic technology becomes avail- ments in scheduling procedures, encounter forms, med-
able. Managing this constant change is a core responsibil- ication handling, patient flow within and across clinical
ity of the people and the organizations that make up the departments, communication within patient care teams,
health care system. The specific patterns and processes of record-keeping, and other administrative procedures.
care in one clinical setting are the product of thousands of QI methods include a variety of tools for motivating
small and large decisions about handling change that and structuring the cooperation, changing the process or
system, monitoring what happens, and evaluating the


change. In addition to local “learning by doing,” QI in- medicine. With scientific progress has come better under-
volves deliberate applications in local settings of new standing of bodily processes and more effective treatments
knowledge developed elsewhere. The goal of many QI ac- for illness. As these treatments have been embodied in new
tivities is to introduce a new clinical practice based on an technology, the provision of health care has become a
advance in medical science or a new administrative prac- more complex process, involving an array of organizations
tice based on managerial experience or social science re- and specialized professionals working together in complex
search. Moreover, QI activities often use the science of or- systems of care.
ganizational change to design strategies for changing be- Nevertheless, even as medical practice has changed, the
havior in the local setting—for example, to get adherence moral and ethical aspects of medicine have remained cen-
to a new practice guideline or a new patient safety proto- tral to the understanding of what it means to be a member
col. of a health profession.18 Since health outcomes are inher-
In sum, QI is a particular form of the clinical and man- ently uncertain and vary under different treatment op-
agerial innovation and adaptation that has always been an tions, health care decisions are complicated. Patients find
integral part of health care operations. To the process of it difficult to understand their choices, especially when
change, it adds systematic, self-reflective design, monitor- uncertainty and the risk of adverse events are involved,
ing, data analysis, and assessment of the effects of changes and they must rely on physicians and other health care
to ensure that they are in fact beneficial, and it feeds this providers for help. Given the importance of health to well-
information back into the delivery setting. It uses new being, patients are vulnerable to exploitation in this de-
knowledge developed in the medical and social sciences to pendent role.
help identify positive changes and design practical strate- Therefore, society continues to expect physicians to ac-
gies to implement them in the local setting. In the process, cept a special ethical responsibility to serve each patient’s
the QI practitioner implicitly or explicitly uses each at- interest. As other health professionals and health care or-
tempt to improve care to deepen his or her understanding
of the content and process of bringing about positive
change in that setting. Ideally, over time, the successful use Improvements in the quality of health care
of QI methods helps to transform the culture of an orga-
nization into one in which everyone is committed to con- will not happen automatically. Systematic and
tinuous quality improvement and has the skills to partici-
pate in it. Thus, QI has the potential to make a substantial self-conscious management of health care
contribution to solving the quality problems in American
health care. delivery is needed.
The Ethics of Improving Health Care Quality
ganizations have played increasingly important roles in

B efore we can discuss the ethical issues that arise when

QI methods are used in health care, we must address
the underlying question of who is responsible for improv-
care delivery, analogous ethical responsibilities to serve the
patient’s interest have been recognized for them.19
The understanding of these ethical responsibilities has
ing the quality of care. Health care is of unusual conse- sharpened as health care has evolved from a simple pa-
quence because of the role it plays in relieving suffering, tient-physician interaction to today’s complex interactions
preventing premature death, restoring function, increasing between patients and caregiving teams and organizations.
opportunity, providing vital information about an individ- It is more evident now than in years past that the ethical
ual’s condition, and giving evidence of a community’s mu- imperative to “serve the patient’s interests” does not mean
tual empathy and compassion. As a result, health care ac- ignoring the interests of every patient except the one pre-
cess, quality, and cost have always been matters of societal sent at the moment. Since health care organizations serve
ethical concern.17 groups of patients, clearly they must manage the process
Since antiquity and across different cultures, healers of care with policies and procedures that balance the needs
have been revered as persons with special knowledge of the of all the patients they serve. Characterizing this as “we
mysteries of life and death, and the relationship between treat populations now, not patients” or “the patient’s inter-
patient and physician has held deep moral and religious est must now be subordinated to the population’s interest”
significance. Physicians, nurses, and other health profes- is misleading, however. Physicians have always had multi-
sionals have been understood to have a special ethical re- ple patients in their practices, and hospital-based nurses
sponsibility to serve the interests of their patients, includ- have always managed care for multiple patients simultane-
ing the responsibility to maintain and continually strive to ously. They have thus had to consider the interests of all
improve the quality of the care they provide. their patients in allocating personal resources of time and
Over time, the concept of medicine as a profession has energy to their practices. Moreover, it is in an individual
taken hold and science has replaced religion as the basis of patient’s interest to have health care providers follow or-

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S9
derly processes that function well for people in different meet the quality expectations of patients, health care pro-
situations. After all, one could be in one of those situa- fessionals and managers must have the cooperation of the
tions oneself sometime. patients themselves, because modern health care delivery
It is abundantly clear, therefore, that health profession- is a collective enterprise. Providing safe, effective, patient-
als and health care organizations have an ethical responsi- centered, timely, efficient, equitable, and up-to-date care
bility to serve the interests of patients, and patients cer- is a health care organization’s mission.22 QI methods allow
tainly have an interest in the quality of health care. But the organization to pursue its mission in a systematic,
what is the nature of this interest? What level of quality do data-guided way. To be effective, these methods require at
patients want? What level are they entitled to? In the case least a minimum level of cooperation from patients, such
of an ordinary commodity, these questions would usually as permission for the collection and use of data on their
be answered through a market process. Buyer preferences health conditions, treatments received, and outcomes. In
and budgets would interact with seller production costs to other words, ongoing activities to maintain and improve
produce the products that buyers wanted and could af- the quality of care, including activities using QI methods,
ford—and there would probably be an array of quality are an integral part of the normal operations of the orga-
levels. nization. This means that someone seeking care from a
But given health care’s complexity, people cannot easi- health care organization cannot insist on the freedom to
ly make informed assessments of quality and develop sen- opt out completely from efforts to improve the quality of
sible and stable preferences for their health care. More- care in that organization without jeopardizing the very
over, even if they could, they would have difficulty getting benefits he or she seeks. In fact, it is in the best interest of
them implemented, because often they are not paying di- patients to cooperate with QI activities and even to seek
rectly for their care. Since future health status is uncertain, out the health care organizations that are the most com-
people need insurance to make sure they will be able to af- mitted to QI.
ford the care they want if they get sick. In addition, like As an ethical matter, the responsibility of patients to
most societies, the United States recognizes a societal eth- cooperate in QI activities is justified by the benefits each
ical obligation to provide at least some basic health care patient receives because of the cooperation of the others in
without regard to ability to pay.20 Third-party payment the collective enterprise. To reap the benefits of such a sys-
adds a social dimension to decision-making. When re- tem without participating in it—to be a “free-rider”—
sources are pooled, whether through private insurance or would be unfair.23 A patient’s responsibility to cooperate
through publicly funded social insurance, collective deci- is, of course, subject to a standard of reasonableness,
sions must be made about how the pooled resources which presumes that adequate protections against individ-
should be spent. These decisions play a key role in deter- ual harm and violation of rights are in place. For example,
mining both the quality and the quantity of care to which patients can reasonably expect to have the confidentiality
patients are entitled. of their personal health information protected, and to
Within the framework of these collective decisions, have the opportunity to choose whether to participate in
health professionals and health care managers have an eth- a QI activity that exposes them to more than minimal risk
ical responsibility to meet socially determined quality ex- compared to routine medical care. We will address the na-
pectations, which may take the form of a threshold of ture of these protections and what constitutes minimal
minimally acceptable care with a range of permissible additional risk over and above that found in routine med-
quality variation above it. They have an ethical (and often ical care in greater detail later in this report.
a legal) responsibility to cooperate in activities to improve In sum, health professionals, health care organizations,
the performance of the teams and the organizations they and patients have an ethical responsibility to cooperate in
belong to in order to meet those expectations. They also maintaining and improving the quality of health care. In
have an ethical responsibility to cooperate in societal ef- its traditional role of protector of the health and safety of
forts to improve the overall health care system to accord its citizens, government also has ethical responsibilities
with those expectations. These obligations follow directly with respect to the quality of care. In particular, it should
from their underlying professional responsibility to serve help to clarify the content of the ethical and legal respon-
the interests of patients. sibilities of health care professionals and organizations to-
The responsibility to improve has been stated in terms ward their patients and to ensure that the responsibilities
of meeting current expectations. That is, it is a responsi- are met.
bility to reach at least a threshold standard of quality.
What makes it a responsibility for ongoing improvement is
the understanding that the societal expectation is one of
continuous progress in medicine—progress that produces
its benefits when it is incorporated into medical practice.
As part of the system of care, patients also have a re-
sponsibility to participate in quality improvement.21 To


II. QI and Research: Similarities and Differences

lthough QI is closely related to clinical and man- exploited in the scientist’s search for insight—or more
agerial practice, it also has much in common with pragmatically, the search for funding, publication, and ca-
research. QI uses the kind of reasoning that is in- reer advancement. To counter these pressures to use peo-
herent in the scientific method, it involves systematic in- ple wrongly to further the ends of other people and justi-
vestigations of working hypotheses about how a process fy the research, the subject’s decision to participate in the
might be improved, and it frequently employs qualitative research must be voluntary and fully informed.
and quantitative methods and analytic tools that are also In health-related research, the independent nature of
used in research projects. It is these similarities between QI the enterprise is obvious when the research takes place
activities and research projects that have touched off the away from a health care delivery setting and the human
debate about whether QI activities should be subject to subjects are healthy. When the research takes place in a
the DHHS regulations on human subjects research. clinical setting and involves sick patients who are receiving
How does QI differ from research? The definition of care, research and therapy may be mixed, complicating
research in the DHHS regulations highlights the knowl- matters significantly. Nevertheless, even research on an in-
edge-seeking aspect of research as the element that sepa- tervention that holds out the prospect of direct benefit to
rates it from other activities: “Research means a systemat- the subjects is seen to have a different relationship to its
ic investigation, including research development, testing setting when compared to QI. Research with human sub-
and evaluation, designed to develop or contribute to general- jects in a clinical setting is usually conceived, funded, and
izable knowledge.”24 In this definition, research is designed
to develop new knowledge, not to implement knowledge;
implementation happens later and separately, if it happens QI is an ongoing process undertaken as a
at all. Dissemination of research results, such as through
publication in scientific journals, is thus of fundamental consequence of health care providers’
importance. Allowing research subjects to assume the bur-
dens and risks of research is justified by the expectation of responsibility to serve their patients’
societal benefits from the new knowledge produced; pub-
lication is an important step in conveying the new knowl- interests. This makes it very different
edge to those who can put it into practice and thereby cre-
ate the social benefits. from research.
Although the definition does not make it explicit, the
regulations implicitly reflect a view of research as a knowl- managed as discrete projects, each led by a principal inves-
edge-seeking enterprise that is independent of routine medical tigator who is responsible for the project’s design and con-
care. Opinions differ on whether a society has an obliga- duct. For research in a clinical setting, the resources often
tion to engage in research, including research that will ul- come from outside the organization in which it is carried
timately yield important benefits to human health. But out (from a federal agency, private foundation, or biotech
even if such an obligation exists, there is a presumption in or pharmaceutical corporation, for example); or, if from
research ethics that the research enterprise should rely on inside, from a separate research budget, not from clinical
volunteers.25 Investigators choose to do research, and peo- care resources. Most important, there is significant uncer-
ple should be able to choose whether to be research sub- tainty about whether the intervention is in fact beneficial,
jects. This ethical paradigm emerged in response to re- and the activity is designed to produce generalizable
search that imposed substantial risks on subjects yet of- knowledge about the intervention, not immediately im-
fered them no prospect of direct benefit (such as the Nazi proved care. This usually means that research has a proto-
experiments and the U.S. Public Health Service Study of col that is constructed to minimize the effect of specific
Untreated Syphilis, commonly known as the “Tuskegee local variables and is maintained unchanged during the
syphilis study”), and it builds on the observation that the period of the research. It also means that there is relatively
interests of researchers are inherently tangential to, and little urgency to disseminate the results, and the results
possibly even at odds with, the interests of the subjects. may not be made known to the scientific community for
Since researchers want to produce successful research, they many months, even years, after the research is begun. Fi-
have a strong interest in enrolling subjects in their studies, nally, there is no presumption that the results will be in-
keeping them in, and getting them to conform to study corporated into the local care delivery process, and fre-
rules. People who do not know themselves to be involved quently little or no attention is paid to the challenges and
in research, or who do not have the opportunity to discern opportunities for implementation of the findings in any
the merits of their own involvement, are at risk of being setting at all.

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S11
Research QI Clinical &
Innovation &
Research Adaptation
on QI QI/
on QI


Note: The figures are not drawn to scale.

Figure 1.

In contrast, QI is an integral part of the ongoing man- able insights for a QI practitioner in another setting.
agement of the system for delivering clinical care, not an in- Moreover, QI activities that have theory-based designs in-
dependent, knowledge-seeking enterprise. QI practition- evitably become examples that in the long run tend to in-
ers design QI activities to bring about immediate im- crease or decrease the level of confidence in the theories
provements in care, relying on theory and evidence from on which they are based. With careful, systematic report-
research and practical experience to identify changes that ing, the insights from QI are of use to people in other set-
are very likely to be beneficial. QI activities take place in a tings, just as case reports on individual patients are useful
particular localized health care setting, their design is ex- (although they too are not research). Given this, QI prac-
pected to incorporate the specific features of the setting, titioners should be encouraged to share information
they are led by people who work in that setting, and they about their QI activities with others in the health care sys-
incorporate rapid feedback of results to bring about posi- tem.
tive change for the patients in that setting. Instead of a Furthermore, although QI and research are conceptu-
fixed protocol implemented for a time period that may ally distinct, in practice they are sometimes combined in
last for years, QI methods often require repeated modifi- one activity. In other words, some systematic, data-guided
cations in the initial protocol as experience accumulates activities designed to bring about immediate local change
over time and as the desired changes engage the local are also “investigations designed to develop or contribute
structures, processes, patterns, habits, and traditions. The to generalizable knowledge”; they are both QI and re-
term “continuous quality improvement” (CQI), used al- search.
most interchangeably with the term QI, highlights the At this point, a diagram is useful to illustrate the rela-
fact that QI is not so much the implementation of dis- tionships among the activities under discussion. In Figure
crete projects as it is an ongoing process of continual, self- 1, the set “Clinical and Managerial Innovation and Adap-
conscious change, undertaken as a natural consequence of tation” is shown as a large oval and consists of activities
health care providers’ ethical responsibility to serve the in- designed to bring about immediate local improvements in
terests of their patients. This makes it a very different kind clinical and managerial practice. The set “QI” is shown as
of endeavor from research and generates the prima facie a subset of that oval; it consists of clinical and managerial
case for questioning whether the public’s interests would innovation and adaptation activities that are designed and
be best advanced by subjecting QI to exactly the same carried out in a systematic, data-guided way.
process of review and evaluation that has been designed “Research” is the large circle and consists of systematic
for discrete research projects. investigations designed to develop or contribute to gener-
QI and research are thus both conceptually and practi- alizable knowledge. The “Research” set includes basic and
cally distinct and play different roles in health care. Nev- applied medical research, of course. It also includes other
ertheless, in the process of making change, QI does yield categories of research with a potential impact on health
information about what works and the way in which care quality, such as epidemiological research, health ser-
change can come about. When the results of QI activities vices research, management research, and educational re-
in various settings are looked at together, patterns may search. The activities that are both QI and research are
emerge—and even a single QI endeavor may yield valu-


shown in the diagram as “QI/Research”—the overlapping of burdens that may fall on participants as a result of a QI
section of the “QI” set and the “Research” set. activity.
One important research category, “Research on QI,” is Researchers have a recognized ethical responsibility to
explicitly depicted in Figure 1. It consists of systematic in- ensure that the human subjects of their research are ap-
vestigations designed to produce generalizable knowledge propriately protected. The people using QI methods to
relevant to the design and implementation of QI activities. manage the quality of care also have an ethical responsibil-
The application of evidence-based medicine requires the ity to participants in their QI activities. The need to meet
generation of new knowledge on the behavior of systems, ethical standards creates a potential need for explicit struc-
and research on QI contributes to this knowledge by help- tures and oversight to ensure that research and QI are
ing to answer questions such as “What are the principles practiced appropriately. It does not necessarily follow,
of change?” “How do these principles work within differ- however, that the standards and oversight should take the
ent organizational contexts?” and “How can one spread same form for both activities.
desired change across an organization or between organi- To address this issue, we must first consider the gener-
zations?” al principles and values that are at stake. We begin with
Research on QI can be independent of the QI activities the principles of research ethics that form the foundation
it studies. For example, an investigator could do a retro- for the current system for protecting human research sub-
spective study of QI activities carried out in different or- jects. The discussion is organized around seven ethical re-
ganizations with the aim of testing a hypothesis about the quirements: social or scientific value, scientific validity, fair
effects of organizational characteristics on results. Alterna- subject selection, favorable risk/benefit ratio, respect for
tively, QI and research on QI can be combined in a single potential and enrolled subjects, informed consent, and in-
activity designed to produce both immediate local change dependent review.27 We discuss the justifications given for
and generalizable knowledge about the process of change. each requirement in the context of research and then ask
For example, a health care organization with multiple de- whether the same reasoning applies to QI work. (The re-
livery sites could conduct an activity in which the sites are sults of the latter portion of this analysis are summarized
divided into two groups, a different strategy is used in in Box 5 near the end of this section.)
each group to introduce a new practice, and the results of
the strategies are compared, with elements included in the Social or scientific value: Is the research worth doing? Do the
activity’s design to facilitate generalization of the results to potential gains from doing it justify the resources spent and the
other organizations. In Figure 1, the set of such hybrid ac- risks imposed?28
tivities is the region of overlap between the sets “Research To be ethical, research must be worth doing, because
on QI” and “QI,” and is labeled “QI/Research on QI.”26 there is an ethical obligation to use scarce resources re-
sponsibly. Also, it would be wrong to expose human sub-
Protection of Human Participants in QI and jects to risk without social or scientific benefit. This means
Research that researchers have an ethical obligation to share the
knowledge gained from research with others, through

B oth research and QI can adversely affect the people

who participate in them. Today’s human research sub-
ject protection system was inspired by research projects
publication in peer-reviewed journals or by other means,
so that the benefit can actually be achieved.
Obviously, QI should also be worth doing for the same
that offered subjects no direct benefit and exposed them to reasons. The potential gains must justify the resources
substantial harm. In QI, however, the changes made in the spent on the activity and any risks imposed. The primary
process of delivering care are expected to be improve- gains from QI are the benefits from the local improve-
ments, and given the serious quality and safety problems ments that result. Broader social benefits are also possible
in health care, patients are often at greater risk if a current when reports of QI activities in different settings generate
practice is allowed to continue than they are if a QI activ- insights about the nature and process of improvement. To
ity goes forward. Nevertheless, any change may have un- achieve the social benefits, those who conduct QI activi-
expected negative consequences, and even the data collec- ties should be willing to share information about them
tion and monitoring that makes the change a QI activity with others in the health care system.
may itself impose burdens on the QI participants. This requirement is thus the same for QI and research.
Burdens can take the form of direct physical harms, Its application is different, however—and in some ways
mental and psychological harms, “hassle” harms such as more difficult—in the QI context. To establish the worth
time consumed in completing surveys or submitting to of a research project, one must assess the net benefit to so-
extra clinical or administrative procedures, or harms relat- ciety from the knowledge it might produce. Usually, this
ed to loss of privacy and confidentiality. Often, there is amounts to ensuring that the research question is appro-
uncertainty about whether a harm will occur and how se- priate and the research design adequate. To establish the
vere it will be if it does occur. In the rest of this report, we worth of a QI activity, one must assess the expected net ef-
will use the broad term “risk” to refer to the various kinds fects of the proposed process of change on present and fu-

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S13
ture patients in the local setting, as well as any social ben- risks in the usual work situation. For example, in the un-
efits that may result from sharing the insights gained. likely event that a QI activity exposes workers to extra ra-
Forming an estimate of the impact of the local change can diation or toxic chemicals, or invades privacy by collecting
be a complex professional and management task that re- information employers are not normally entitled to have
quires a detailed understanding of the local system of care (as might happen with surveys of the use of alcohol, ciga-
delivery. rettes, or illegal drugs outside the workplace), the potential
One must also assess the potential risks to those who effects on workers should be included in the assessment of
participate in the QI activity itself. The chief difficulty QI-associated risk. On the other hand, one should not in-
here lies in the identification of the appropriate baseline clude the harm to a worker’s economic security that might
from which to measure the risks to participants attribut- result if a QI activity reveals that the worker is incompe-
able to the QI activity. In a freestanding research project, a tent or the organization can provide quality care with
person is either getting an intervention or not, and the fewer workers. A health care employer has a moral and
baseline for measuring harm and risk is ordinary life with legal right—and responsibility—to ensure that employees
no intervention. In a clinical treatment research project, are competent and conscientious, and the operations of
the fact that the subject would be receiving clinical care in the organization are efficient. The collection of informa-
any case complicates the assessment, but typically, the re- tion on individual and system performance is a feature of
search has a control group receiving a carefully specified normal working conditions, and the risk that it will pro-
standard treatment that is the appropriate baseline from duce information that leads to adverse job actions is part
which to measure the risks to subjects receiving the un- of the baseline job risk.
proven treatment.
In the QI context, the risks related to extra visits, ques- Scientific validity: Is the research methodologically sound—
tionnaires, or procedures needed for data collection and for example, is it properly structured to achieve its goals?
monitoring can be distinguished and attributed to the QI This requirement is also justified by the scarcity of re-
activity. Although these burdens are usually small, they sources and the need to avoid exposing subjects to possible
should be weighed in deciding on the merits of proceed- harm without benefit, and it also applies to both research
ing. But how should one assess the net risks to the partic- and QI. The interpretation of “scientific validity” and
ipant associated with the proposed change itself? Here it is “methodologically sound” must be adjusted to QI, howev-
important to note that the care delivery process to which a er. Since the primary goal of QI is local improvement, QI’s
patient is entitled is not specified in all its details; in fact, methods and the local knowledge they generate can be
it is always changing. Even in a single clinical setting, the quite different from the methods designed to enhance the
attributes of the care patients receive can vary from day to opportunity to gain generalizable knowledge through re-
day, depending on the availability of staff, the number of search.29 QI methods should be chosen on the basis of
patients scheduled, and other factors. Managers and clini- knowledge of the context, the requirements for connect-
cians have the professional discretion to make changes in ing evidence to the context, and the requirements of actu-
care within certain (fuzzy) limits related to the range of ally “making something happen” in the local setting while
variation in existing health care delivery. Over time, they balancing the importance of the improvement sought and
are actually required to make changes in order to remain in the clarity needed in the assessment of changes against the
compliance with evolving minimum standards of care and practical costs of monitoring the effects. A randomized
to avoid malpractice liability. For many of the changes in- controlled trial seeks to eliminate context—key to QI—as
troduced through QI methods, the difference for patients a variable and thus would usually be inappropriate for the
between being in or out of the QI activity may be no kinds of questions and changes that QI addresses, while
greater than the existing variation in the patterns of care changes that are already known to be worth their costs and
across organizations, or within the same organization. readily implemented should in general simply go for-
Moreover, the methods of QI guide practitioners to use ward.30
theory, evidence, and practical experience systematically in
order to identify and implement changes that are very Fair subject selection: Are the subjects of the research selected
likely to be beneficial. We are not suggesting that it is im- so that vulnerable individuals are not targeted to bear the
possible to recognize that a specific change has the poten- risks of the projects while the rich and socially powerful are fa-
tial to cause significant harm. Rather, we are noting that vored for receiving the benefits?
patients cannot expect that services will only be presented In the selection of research subjects, there should be
in one way, or that those services can be specified fully in similar treatment of equals, and both the burdens and
advance. This fact is relevant in the assessment of the risk benefits of the research should be distributed in a fair
attributable to inclusion in a QI activity. manner. But, to quote Emanuel and colleagues: “This
A related issue arises for staff members, the other major does not mean that individual subjects and members of
group of human participants in QI activities. For them, groups from which they are selected must directly benefit
the appropriate baseline for measuring QI effects is the from each clinical research project or that people who are


marginalized, stigmatized, powerless, or poor should never The other two requirements do not apply to QI as stat-
be included.”31 ed. The next section on informed consent explains why
This requirement applies to both research and QI. Fair permitting withdrawal from QI activities does not always
subject selection is particularly relevant to organizations apply. It also argues that QI participants should receive
conducting entire programs of QI activities. The choice of general information about QI efforts and should be able
objective and the design of the activity will sometimes to learn more about specific projects if they are interested,
have implications for the fairness of the distribution of but a QI practitioner is not required to provide informa-
benefits and risks across patients. For example, the distrib- tion automatically about a QI activity’s existence or results
ution of benefits and risks may be affected by choosing to to the participants.
do QI projects on treatment of heart disease rather than
treatment of diabetes or asthma, or introducing improve- Informed consent: Do the subjects receive information about
ments in some parts of an organization and using the the purpose of the research, its procedures, potential risk, ben-
other parts as comparison groups. Such distributional ef- efits, and alternatives, so that each subject understands the in-
fects should be considered in evaluating the ethical accept- formation and can make a voluntary decision whether to en-
ability of QI. In particular, people who are currently dis- roll and continue to participate?
advantaged in the health care system should not have to This requirement differs for QI and research. In princi-
bear a disproportionate share of the burden of improving ple, research participation is optional for all parties,32 but
the system, and in fact should be given priority in the dis- we have argued that this is not so for QI. Continuous
tribution of the benefits of QI activities. quality improvement is part of the mission of health care

Favorable risk-benefit ratio: Is the research designed to mini-

mize the risks and maximize the potential benefits? Are the Because QI is an essential part of normal
risks to an individual human subject proportionate to the
benefits to the subject and to society? health care, it is necessary—and
This requirement applies to both research and QI. It
relates to the requirement of social or scientific value, but acceptable—to have consent to receive
focuses directly on the participant’s situation, rather than
the activity’s overall net benefit. An affirmative effort health care include consent to a reasonable
should be made to design both research and QI activities
to minimize the risks and maximize the benefits to the level of cooperation with QI activities.
participants and society. For example, an activity may have
a risk/benefit relationship within an acceptable range, but professionals and health care managers. They owe it to
if the relationship could be improved with minimal their patients to be constantly trying to improve their
effort/cost, both researchers and QI practitioners are ethi- practice; QI is an important tool for doing this; and there-
cally obliged to do so. fore, participating in QI is not completely optional for
them. Since quality improvement is a critical ingredient in
Respect for potential and enrolled subjects: Is respect for sub- the creation of the benefits a patient seeks from health
jects demonstrated by the following actions? care, QI is not completely optional for patients, either.
In practical terms, this means that specific informed
 protecting the privacy of individuals and maintaining
consent is not required every time a human participant is
the confidentiality of private information
included in a QI activity. Because QI is an essential part of
normal health care operations, it is necessary—and ethi-
 maintaining the welfare of the subjects
cally acceptable—to have consent to receive health care in-
clude consent to a reasonable level of cooperation with the
 informing subjects of newly discovered risks or bene-
ongoing QI activities of the organization providing the
fits associated with participation
care. Through public education, people should already be
aware of this before they seek care. When they present
 permitting withdrawal
themselves for care, they should be reminded of the back-
ground expectation and informed in general terms about
 informing subjects of the results
the QI that occurs in the particular setting and how they
The first three of these requirements apply to both re- can obtain more information about individual projects.33
search and QI. Protecting privacy and maintaining confi- By “a reasonable level of cooperation,” we mean that
dentiality is important in QI, as is maintaining the welfare patients may be included without their specific consent in
of the participants and informing them of newly discov- QI activities that involve no more than minimal addition-
ered risks or benefits associated with their participation. al risks to them compared with the normal risks associated
with receiving standard medical care. If a QI activity is as-

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S15
sociated with more than minimal additional risk, patients of protected health information in both clinical care and
must be asked for their specific informed consent before QI. (See Box 4.)
inclusion, and they are free to refuse it. In the latter case, When the human participants in a QI activity are staff,
patients must receive full information about the activity consent requirements play out a little differently. Health
and the risks to them personally when they are asked to professionals, management, and other workers have a gen-
participate, just as for any clinical consent. If later on they eral ethical obligation to cooperate to improve the quality
would like information about the activity’s results, they of care, and a specific contractual obligation to cooperate
should be able to obtain it.34 with their organization’s QI program (since it is part of
Organizations should not be required to distribute de- normal health care operations). The obligation is not un-
tailed information proactively about the results of individ- limited, however. Specific consent is required when the QI
ual QI activities. Most people would find the detail super- represents more than minimal risk to the worker, as mea-
fluous, and the effort would waste their own resources sured from the baseline of normal working conditions.
since they ultimately pay the bill for the health care sys- Therefore, consent would be required for a QI activity
tem. QI practitioners and the organization as a whole that exposed workers to more than minimal additional
should, however, willingly provide information to anyone risk of physical or mental harm compared to their current
who asks about any QI activity, to the extent possible working conditions (exposure to radiation or toxic chemi-
while respecting patients’ and providers’ right to privacy. cals, for instance) or collected information about workers
The most common examples of QI activities that do that was outside the category of information employers are
not require consent are those activities that use routine normally entitled to have about their employees (such as
personal health information to implement a change in the their use of tobacco, alcohol, or illegal drugs outside of the
care process and impose no additional risk on participants workplace). Consent is not required, however, for QI that
beyond that associated with the information use. For ex- is risky to the worker simply because it might generate ev-
ample, a group practice might decide to modify schedul- idence of incompetence on the job or lead to a reduction
ing practices to reduce the time patients spend waiting to in force for efficiency reasons. The formal contracts with
see their physicians when they come for an appointment. workers and the informal workplace expectations should
The QI team might begin by reviewing patient visit reflect the understanding that cooperation with minimal
records to develop data on the types of visits, the average risk QI activities is part of the job. As with patient partic-
time needed for each type, the pattern of visit types over a ipants, confidentiality of personally identifiable informa-
typical month, and so on, and then use the data to devel- tion about workers should be appropriately protected
op new scheduling practices, implement them, monitor within the organization.
the effect on waiting time, and refine them until the de- Does any of this analysis change if those leading the QI
sired result is achieved. If confidentiality is appropriately project publish or otherwise share their process and ac-
protected, then using patient information for QI activities complishments with others? Do participants have to give
without specific consent is ethically acceptable on the specific consent to the publication of QI results? QI prac-
same grounds that its use is acceptable in other normal titioners should be encouraged to share QI results through
health care operations. Since QI is part of the care process, publication and other means, since sharing yields benefits
the people doing QI have the same status as others in the to individual patients and to society as a whole. The peo-
care process. Often this is literally true, since QI teams are ple who receive care from health care organizations pay as
often made up of the people who provide direct care; how- a group for the development of this knowledge, and over a
ever, the rationale is the same when access to protected lifetime, people receive care from many different organiza-
health information is granted to someone more removed, tions. Although some organizations may consider QI re-
such as an epidemiologist abstracting information from sults proprietary information and wish to maintain secre-
medical records for statistical analysis. cy for competitive advantage, patients are better off if
The Privacy Rule of the Health Insurance Portability health care providers cooperate by sharing QI results.
and Accountability Act (HIPAA) has established standards Even though most QI can be carried out ethically with-
for the protection of the confidentiality of personal health out explicit patient consent, published results must be pre-
information in normal health care operations, and has in- sented in a form that preserves patient confidentiality;
cluded QI (but not research) within that category.35 Con- otherwise, the team must have the patients’ specific con-
sent to the use of protected health information for QI ac- sent to publication. Similarly, individual worker informa-
tivities is provided as one item in the list of routine uses tion must be nonidentifiable or worker consent to publi-
the patient agrees to as a condition of treatment. Accord- cation must be obtained, unless there are overriding con-
ing to HIPAA standards, a QI practitioner’s access to pro- tractual or legal rules relating to the provision of informa-
tected health information is limited to the information tion on worker performance.
needed for the activity, persons engaged in QI have the The ethical acceptability of the QI activity itself and
same confidentiality obligations as other health care work- the ethical acceptability of the form in which the results
ers, and the same precautions must be in place for the use are disseminated are conceptually distinct, and the deci-


4. Summary of the Privacy Rule of the
Health Insurance Portability and Accountability Act
45 CFR 160 and 45 CFR 164, Subparts A and E

 defines health information as “any information, • PHI will be disclosed to business associates provid-
whether oral or recorded in any form or medium that: ed certain arrangements in place (45 CFR 164.502 (e))
(1) is created or received by a health care provider, health • the disclosure is required by law, authorized by law
plan, public health authority, employer, life insurer, for public health purposes, or required by the FDA for
school or university, or health care clearinghouse; and (2) adverse event reporting (45 CFR 164.512)
relates to the past, present, or future physical or mental
• an institutional review board (IRB) or privacy board
health or condition of an individual; the provision of
waives or alters the authorization requirement to use or
health care to an individual; or the past, present, or fu-
disclose PHI for research, which is possible when certain
ture payment for the provision of health care to an indi-
conditions are met (45 CFR 164.512(i))
vidual” (45 CFR 160.103)
• a researcher satisfies certain conditions to use PHI
 defines protected health information (PHI) as “indi- for review preparatory to research (45 CFR
vidually identifiable health information that, subject to 164.512(i)(1)(ii))
exceptions, is (i) transmitted by electronic media; (ii)
• a researcher satisfies certain conditions for research
maintained in any medium described in the definition of
solely on information of decedents (45 CFR
electronic media; (iii) transmitted or maintained in any
other form or medium” (45 CFR 164.501)
• the PHI has been deidentified in accordance with
 extends to a defined set of covered entities that trans- specific Privacy Rule standards (which makes the data no
mit health information electronically in connection with longer PHI [45 CFR 164.514(a)-(c)])
certain defined HIPAA transactions (45 CFR 160.13):1
• the covered entity releases the information in the
• health plans form of a limited data set that includes the removal of
• health care clearinghouses certain identifiers and a data use agreement between the
• health care providers researcher and the covered entity (45 CFR 164.514(e))
• nonemployee business associates of covered entities
 QI (but not research) is included in health care oper-
• researchers who are health care providers engaging
ations and specifically exempted from the above require-
in any of the covered electronic transactions
ment for project-specific written permission (“authoriza-
 prohibits a covered entity from using or disclosing tion”)
PHI about an individual without the individual’s specif-
ic written permission (“authorization”) unless: 1. For additional information see U.S. Department of Health
and Human Services, OCR Privacy Brief, Summary of the HIPAA
• PHI will be used/disclosed for treatment, payment, Privacy Rule, at
and health care operations (45 CFR 164.506(c))

sion about whether to publish and in what form may rea- takes judgment calls, and conflicts of interest may distort
sonably be made after the results are in. Therefore, it is ac- the researcher’s judgment (consciously or unconsciously),
ceptable and may be desirable to separate ethical assess- potentially causing harm to participants or taking away
ment of the activity itself and ethical review of the form in their opportunity for voluntary consent. Review can pro-
which the results are disseminated. Thus, organizations tect against such distortion and help potential research
might well have different procedures for ensuring ethical subjects have confidence in the honesty and integrity of
implementation of QI and ethical dissemination of the the offer to volunteer to participate. In the United States,
findings. this ethical review takes the form of review of research pro-
tocols and relevant documents by the institution’s IRB.
Independent review: Is there review of the design of the re- In the discussion of the first six categories of ethical re-
search, its proposed subject population, and its risk-benefit quirements for human research subject protection, we
ratio by individuals unaffiliated with the project? have concluded that QI activities with human participants
In research, the purpose of ethical review by individu- should meet similar ethical requirements, but the require-
als unaffiliated with the proposed research activities is to ments must be interpreted and applied somewhat differ-
ensure that the researcher has understood the require- ently. (See Box 5.) As with research, some considerations
ments for ethical research and applied them properly in might justify review of QI activities to ensure that the re-
developing the proposed activities. Applying requirements quirements for ethical conduct of QI are met. Review

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S17
might be justified, for example, if the QI might pose sub- ments. The research protocol specifies the goals of the pro-
stantial risks or waste substantial resources, and if review ject, the subject population, methodology, and time peri-
by a nonparticipant could raise those issues and stop the od for the research. After reviewing the documents, the
activity. As QI becomes a desirable skill and care systems IRB either approves the research project or requires the
invest in QI budgets, the leaders of QI might find them- principal investigator to make changes to the protocol, the
selves invested in QI work that has become unresponsive associated consent form, or other documents until all are
to the real needs of patients and families—a conflict of in- acceptable. Then the project begins and continues un-
terest that would echo that which gave rise to research re- changed unless the PI obtains IRB approval for modifica-
view. Furthermore, the focus on QI is relatively new, and tions. In addition, the PI is required to report any unan-
the practices that support it are still in flux. Thus the ticipated problems involving risks to subjects or others to
boundaries between QI and research and between QI and the IRB. Although the research is carried on under the
routine treatment are not well understood. Review by a auspices of the institution and receives continuing IRB re-
nonparticipant would be useful if a project that should view (at least once per year), the PI takes full responsibili-
have counted as research is not being managed as such, or ty for its day-to-day conduct.38
if a project that offers no particular opportunity for im- Most QI activities are unlike the activities that IRBs
provement over usual practice is being touted as QI. routinely approve and monitor. Instead of a fixed protocol
Since these arguments justify some review, at least for with fixed goal, methodology, population, and time peri-
some projects, the question is what form it should take. od, QI activities ordinarily entail frequent adjustments in
Should QI also receive IRB review, or would some other the intervention, the measurement, and even the goal over
approach to ethical oversight be more appropriate? In the time as experience accumulates. Even in cases where a QI
next section, we will discuss why we have concluded that activity produces insights useful in other settings as well as
IRB review is not ethically required and would not even local change, the activity is not a knowledge-seeking en-
be generally beneficial or effective for QI activities. terprise that is independent of ongoing clinical care. It is
designed to produce local change and is closely linked to
IRB Review of QI: Why Not? the normal operations of the institution in which it takes
place. QI occurs in the context of the professional obliga-

M uch of the concern expressed about IRB review of

QI has focused on the time and effort associated
with the preparation of documents for submission to an
tions of the health professionals involved, and the people
who initiate or lead a QI activity are often part of the team
that provides care. All parties affected by the QI activity
IRB and the time required for the IRB to complete its re- function within a supervisory and management structure
view.36 Researchers have been dissatisfied with these as- that already bears responsibility for clinical care, including
pects of IRB review for some time, and improving IRB ef- its quality and safety. As we have shown above, the ethical
ficiency is an acknowledged policy goal.37 This issue is requirements for QI are in some ways similar to those for
even more important for QI than for research, since the research, but they must be modified to the particular local
people who initiate QI are already employed in the deliv- setting and the QI goal of immediate improvement. In
ery of care, and the resources used for QI are part of the particular, the desirability of the change itself and the risks
health care system’s cost structure. Even with a more effi- to participants often depend heavily on the characteristics
cient IRB process, many valuable QI projects would be and reactions of the system in which the QI activity oc-
unable to muster the necessary resources and simply curs. This means that assessing the activity’s potential ef-
would not be done if IRB review were required. For those fects and understanding how risks can be minimized re-
QI activities that did go forward, the process would im- quires intimate knowledge of that system.
pose high transaction costs on improvements that are Given these characteristics of QI and its role in health
often small in scale and represent little burden or risk to care, the procedures used to ensure its ethical conduct
participants. This would be a poor use of scarce resources must allow QI activities to remain flexible and fully inte-
and would in itself be a kind of harm to the patients, grated into the ongoing management of care delivery.
whose resources are thereby wasted. Since the current IRB review process does not readily
There is a more basic problem, however: the current allow this, requiring QI to undergo IRB review would cre-
structure of IRB review is inappropriate for QI. To explain ate perverse incentives for professionals and organizations
why, we will describe the process and discuss its limita- trying to improve care. Change is already an intrinsic fea-
tions in the QI context. The IRB’s role is to ensure that re- ture of health care delivery. Solving the quality and safety
search carried out under an institution’s auspices complies problems in American health care will require many
with federal regulations and conforms to accepted ethical changes—and these changes are more likely to be genuine
principles. Usually, the IRB is part of the institution in improvements if done in a systematic, data-guided way,
which the research is done, but it is always independent of using QI methods. If the “price” of making change in this
the research itself. The IRB meets at intervals to review re- way is a cumbersome, costly process of review, however,
search protocols, consent forms, and other relevant docu-


managers may opt to make change without using QI that improve quality and lower cost at the same time.
methods. Worse, they may simply leave things as they are. There is ample evidence of waste in the system, in the
Of course, not all clinical and managerial changes are sense of care that is unnecessary, harmful, or inefficiently
ethically acceptable. The underlying questions here are: produced and delivered. Reallocating the resources devot-
What standards should be applied to determine whether a ed to this care to beneficial uses improves care without in-
change is ethical? How should the agents of change in creasing cost.40
health care be held accountable for their actions? Current- Nevertheless, sometimes a tradeoff between quality and
ly, to the extent that accountability for clinical and man- cost is unavoidable. Managers must make changes for
agerial change occurs, it is through the system of account- budgetary reasons, and the changes may reduce the quali-
ability that has been established for all ongoing clinical ty or quantity of care for some people. Is it always unethi-
care. Admittedly, the clinical accountability system has cal to make patients worse off than they are under existing
shortcomings, and concern about these shortcomings fuels practices—or even to expose them to a small chance of
the concern for the welfare of QI participants. Neverthe- being worse off—without their explicit consent? The an-
less, desirable procedures for ethical oversight of QI should swer is no. Patients have an interest in maintaining and
encourage, not discourage, the collection and evaluation improving quality, but they also have a stake in ethical
of information on the consequences of any change that stewardship of the resources devoted to health care. When
could harm someone, and should avoid acting to slow the changes must be made and budgets are tight, sometimes it
pace of improvement. These ends cannot be accomplished would be unethical not to reduce the quality or quantity of
by exporting QI into the IRB-based research review sys- care for some people. For example, suppose a costly new
tem. diagnostic test becomes available that allows diagnosis of a
Instead, in our view, ethical review to protect human deadly disease at a treatable stage. The medical staff mem-
participants in QI should be selectively imported into the bers of a nonprofit health plan think the plan should begin
accountability system for clinical care. By this, we mean using it immediately in patients at risk for the condition.
that ethical oversight of QI should be fully integrated into
the routine management and supervision of health care
operations.39 The responsibility for ensuring that this inte- The procedures used to ensure that QI is
gration occurs and results in ethically conducted QI
should belong to the entities that already have responsibil- conducted ethically must allow QI activities
ity for the quality of care provided by individual profes-
sionals and health care organizations. to be flexible and integrated into the
Protection of human participants in QI belongs in the
clinical accountability system because ethical decision- management of care delivery. The current IRB
making is an intrinsic part of everyday clinical manage-
ment. Clinical and managerial decisions in health care review process does not allow that.
often—perhaps always—have ethical dimensions. Sending
QI to an IRB for ethical review implies that protection of How should the plan find the resources to cover the test-
participants is an ethical question that is separate from the ing and the effective treatment it makes possible? The plan
other ethical questions that arise in managing patient could raise premiums, free up resources by producing the
care—or even worse, that ethical decision-making in gen- same care more efficiently, reallocate resources by reducing
eral is something to outsource. When an activity is as the quality of care slightly for plan members in a domain
closely tied to clinical practice as QI is, the responsibilities where the current quality level is well above the acceptable
for its conduct should remain closely aligned. The special threshold level, or do some combination of the three. Each
circumstances of widespread conflict of interest and risks alternative has ethical implications, all should be on the
of wrongful use of human subjects without consent that table, and sometimes, reducing the quality of care in an-
generated the strong requirements of separate ethical re- other domain is the solution that best meets ethical re-
view do not apply to QI. quirements. Similar examples could be constructed for a
Some might challenge the last statement by arguing hospital, a group practice, or an individual physician.
that change in health care is inherently biased toward One might acknowledge that change must be man-
making patients worse off for the sake of “the bottom aged, that decisions about change have ethical implica-
line.” From this perspective, QI is often just a cover for tions, and that a decision to trade quality for cost can be
change motivated by concern for profits, not patients. The ethically acceptable, yet still argue that, given the current
underlying issue here is that of the relationship of cost state of quality and safety in health care, these decision-
containment to quality improvement, and the relationship making processes cannot be trusted. Many now agree that
of both to professional and organizational ethics. To re- a transformation in the organizational culture of health
spond to the challenge, one might point out that in today’s care is needed to improve the quality of care. This is exact-
health care system, it is often possible to make changes ly where QI methods make their contribution, however.

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S19
Optimal oversight of change requires learning how to as- cepting “value for money” as a quality characteristic has
sess the dangers of specific changes to patients, and QI is a practical advantages. In considering the ethics of using QI
systematic approach to improving decisions about change. methods to evaluate a designed change, it is best to avoid
By providing evidence on the effects of changes, QI meth- creating a bright-line distinction between efforts to im-
ods are valuable tools in holding health care providers ac- prove quality and efforts to reduce cost. When assessing
countable for the care they deliver and can be used to re- quality, people often disagree about what weights to apply
spond effectively to past failures to live up to quality stan- to different dimensions of care. If several dimensions
dards. We conclude that to the extent that health care change at once, some may consider the result higher qual-
managers have a conflict of interest in managing cost-re- ity care and others lower quality care. Thus, when a pro-
ducing change in health care, QI methods are not the ject reduces cost and also changes quality characteristics,
problem, but an integral part of the solution. there may be no consensus on whether the change is qual-
Our discussion has brought us back to the question ity improvement that also lowers cost or cost containment
raised in Section I of this report: should increasing effi- achieved at the expense of quality. Either way, manage-
ciency by providing the same quality of care at lower cost ment should know what the consequences of the change
count as an improvement in quality? One way to answer is are, and QI methods can be used to find out.
to say: “No, it is a good thing, but it is cost containment, Note that the methods used in QI are also useful for
not quality improvement.” Another way to answer is to the study of changes that everyone agrees are not good for
say: “Yes, ‘value for money’ is a quality characteristic of patients but that are made anyway: for example, changes
health care.” Both positions have conceptual merit, but ac- forced by budget deficits, such as cuts in nursing staff or

5. Ethical Requirements for the Protection of Human Participants in QI Activities

Social or scientific value Informed consent

The gains from a QI activity should justify the re- Patients should give background consent to inclusion
sources spent and the risks imposed on participants. in minimal risk QI activities as part of consent to receive
Scientific validity Patients should be asked for informed consent to be
A QI activity should be methodologically sound— included in a specific QI activity if the activity imposes
properly structured to achieve its goals. more than minimal risk.
The risk-harm ratio for patients is measured relative
Fair subject selection to the risk associated with receiving standard health care.
Participants should be selected to achieve a fair distri- Workers (employees or nonemployee professionals
bution of the burdens and benefits of QI. who provide care within an organization) are expected to
participate in minimal risk QI activities as part of their
Favorable risk/benefit ratio job responsibilities.
A QI activity should be designed to minimize risks Workers should be asked for their informed consent
and maximize potential benefits, and to ensure that risks to inclusion in a QI activity that imposes more than
to an individual human participant are proportionate to minimal risk.
benefits to the participant and to society. The risk to workers is measured relative to the risk as-
sociated with the usual work situation and does not in-
Respect for participants clude any risk to economic security that might result if a
A QI activity should be designed to protect the priva- QI activity reveals that the worker is incompetent or that
cy of participants through confidentiality. the organization can provide quality care with fewer
Participants in a QI activity should receive informa- workers.
tion about findings from the activity that are clinically
relevant for their own care. Independent review
All patients and workers in a care delivery setting Accountability for the ethical conduct of QI should
should receive basic information about the program of be integrated into the system of accountability for clini-
QI activities. cal care. Each QI activity should receive the kind of eth-
QI results should be freely shared with others in the ical review and supervision that is appropriate to its level
health care system, with participant confidentiality pro- of potential risk and project worth.
tected by putting results into nonidentifiable form or ob-
taining specific consent to sharing.


elimination of evening clinic hours. In fact, it may be es- pass” compared to research? No. We believe that review
pecially important to understand the effects of purely can be cheaper, less bureaucratically frustrating, and—
cost-driven changes, since their effects may be more far- most important—at least as likely to ensure ethical con-
reaching and more negative than anticipated. When such duct of QI if designing and implementing it is seen as a
changes are made in an incremental, data-guided, moni- core management function.
tored way, management can understand better what is ac- This approach will be most effective if it occurs as part
tually at stake and more easily minimize the negative im- of a systematic transformation of the existing clinical ac-
pact. Obviously, these activities are not quality improve- countability system. The transformation should include
ment and should not be referred to as QI, since doing so clarification of the ethical responsibility of clinicians and
would foster cynicism about the whole QI enterprise. managers to take the lead in improving the quality of care,
Nevertheless, the practices that provide ethical review of ongoing guidance on the requirements for the ethical use
QI should be designed to encourage rather than discour- of QI methods, and change in professional and internal
age the use of data collection and monitoring methods to organizational cultures to make continuous quality im-
minimize harm in these situations. provement routine. Established external clinical care ac-
In sum, we argue that ethical review of QI activities for countability mechanisms should then be used to make
purposes of protection of human subjects should be inte- sure that professionals and organizations have proper su-
grated into the accountability system for managing the pervision of the conduct of QI, including the extent to
quality of care rather than delegated to the IRB-based re- which ethical requirements are met and whether human
search review system. Is this a proposal to give QI a “free participants are adequately protected.

III. Institutional Arrangements Needed to Ensure

Ethical Conduct of QI

o ensure that QI meets ethical requirements, there government regulators, contracting health plans, and the
must be a set of regular procedures, understood public.
obligations, and clear standards to hold health care Within the clinical accountability system, the arrange-
professionals and organizations accountable. ments for ensuring that QI is conducted ethically should
Many organizations and policies within the health care be designed to be flexible. The specific practices should be
system intertwine to create the existing system of account- able to vary across organizations to accommodate varia-
ability for the quality and safety of health care.41 First, of tion in the local environment and in the QI activities con-
course, is the widely understood social contract, articulat- ducted. The interaction between the characteristics of a
ed in statements of professional ethics, which calls on QI activity and the characteristics of the local setting is
practitioners to be allegiant to the needs of their patients. critical in determining whether the activity is ethical and
Formal accountability organizations include state licensing what kind of review and supervision it should receive. In
bodies, private sector accrediting bodies, the Centers for general, if participants are at more than minimal risk,
Medicare and Medicaid Services, and certification and there must be more clarity about review and more super-
credentialing organizations. Managed care organizations vision. However, for a given level of risk, an organization
and large employers often impose quality performance re- whose employees are very familiar with QI methods and
quirements on the professionals and organizations with ethical requirements can allow QI to be done with a lower
whom they contract. Medical staff in hospitals and group level of supervision than an organization whose employees
practices undertake responsibility for credentialing their have lesser skills. A large organization may choose to have
members and awarding staff privileges. Many hospitals a specialized QI oversight system that a small organization
now support clinical ethics committees. Under certain cir- cannot afford; however, the small organization may not
cumstances, health care organizations are subject to the ac- need any formal review if its QI activities are simpler and
countability requirements included in the HIPAA Privacy its size allows everyone in the organization to know what
Rule and the DHHS regulations for the protection of is going on. For example, an HMO with multiple delivery
human research subjects. The malpractice litigation sys- sites might have formal procedures, while a small nursing
tem allows patients to pursue compensation from health home might simply require an informal review by the
care providers for harms caused by negligence in the deliv- nurse executive.
ery of care. Within this accountability system, the various The practices should be able to vary over time. Since
parties may hold each other accountable over different QI is part of a process of continual change in health care,
content areas. For example, a hospital may evaluate the ac- it would be a mistake to impose an ethical oversight struc-
tivities of its physicians, but then in turn be accountable to ture that cannot respond quickly and appropriately to

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S21
changing conditions. If the QI regulatory procedures erect explicit promises about the kind of care they will receive.
artificial barriers to new ways to organize care delivery and Sometimes the reasons are not directly related to ethical is-
reimbursement in health care, they may prevent desirable sues but to other organizational concerns. For example,
change, or even create perverse incentives that encourage the organization may be seeking to maintain market posi-
undesirable change. tion and reputation, remain financially solvent, avoid law-
The practices should be designed to ensure that ethical suits, or prevent bad publicity. The ethics-related oversight
oversight produces its benefits at affordable cost. The ad- of QI, including the oversight of the protection of human
ministrative complexity of doing QI should be as low as participants, must be incorporated into the overall system
possible so that people employed in direct care are not dis- of management and supervision of health care delivery.
couraged from initiating QI projects. This consideration In discussing how this can be done, it is useful to dis-
reinforces the conceptual arguments for building ethical tinguish three broad categories of oversight: professional
oversight of QI into the ongoing management of care de- responsibility for QI; local management review and super-
livery. vision of QI; and QI-IRB review for QI that is also human
We recommend that the primary responsibility for the subjects research. In addition, there should be procedures
ethical conduct of QI be lodged in individual organiza- in place for ethical oversight of QI activities that are car-
tions, where it should be seen as a normal clinical care ried out through collaborations across organizations.
obligation and integrated into normal supervision and
management, with the organizations’ leaders responsible Professional responsibility for QI
for seeing that the integration occurs and is effective. The Some QI activities are so closely tied to clinical practice
organization doing QI could be, for example, a practice of that no additional oversight beyond normal management
one or a few physicians and support personnel, a large supervision and professional ethics is needed. Activities
multispecialty physician group practice, a hospital, a nurs- appropriate for this category are of minimal risk, in the
ing home, a hospice, a staff model HMO, or a more loose- sense that confidentiality is protected and no additional
ly configured managed care plan. It could also be a hospi- risk or harm is imposed on patients compared to that ex-
tal chain, a nursing home chain, or an integrated, compre- pected from clinical care in the absence of the activity.
hensive multisite managed care organization like Kaiser or (And often, a priori evidence exists that suggests that sub-
Intermountain Health Care. It could be the Veterans stantial benefits are likely for the patients or staff in-
Health Administration. Whatever the organization is, it volved.) Therefore, specific informed consent is not re-
must have a set of practices in place to ensure that QI ac- quired. The activities are simple in design, so there is no
tivities are done, and that they are done in conformity need for methodological review. The effects of the activi-
with ethical requirements. The organization must also ties are very local, in the sense that their success or failure
have a procedure in place to provide basic information to will have no repercussions on other parts of the organiza-
patients and staff about the organization’s QI activities. tion. The activities do not use significant additional orga-
We call this internal accountability for the ethical conduct nizational resources beyond the usual resources for clinical
of QI. care.
We recommend that there also be procedures to ensure This category includes most of the routine perfor-
that individual health care organizations and their leaders mance-enhancing improvement activities undertaken by
are held accountable for having well-designed, functioning individual health professionals. Here are two examples of
internal accountability practices in place to carry out their it:
QI-related responsibilities. We call this external account-
 A doctor wants a test order tracking system that
ability for the ethical conduct of QI, and we argue below
makes it easy for him to remember what tests have been
that it should be integrated into the overall system of ac-
ordered for a patient, access the test results, and make
countability for the quality and safety of clinical care.
timely and appropriate adjustments in the patient’s care.
He develops his own patient tracking system, monitors
Internal Accountability for Ethical Conduct
of QI
its performance, and refines it over time in response to
his observations until he is satisfied with it.

A ctivities in a health care facility require management

and supervision for many reasons besides the protec-
tion of the human participants in QI. Sometimes the rea-
 Several nurses find that they are each using a different
method to perform a routine nursing procedure. The
methods are within standard nursing practice and are
sons are directly related to other ethical issues in health
used interchangeably in various clinical settings; howev-
care delivery. For example, management and supervision
er, the nurses wonder whether the methods differ in the
are required to ensure that an organization fulfils its ethi-
time they take and the effect they have on the patients.
cal responsibilities to employees, ethical responsibilities of
The nurses try each method in sequence over a few
stewardship of the organization’s resources, and ethical re-
months, keep track of how long the procedure takes
sponsibilities to patients who have been given implicit or


with each method, and ask patients whether there is any review by supervisors to verify that it meets ethical re-
difference in comfort. At the end of this QI activity, they quirements. Usually, the activity could simply go forward
conclude that one method works better than the others within the ordinary supervisory framework.
in their local setting and all decide to use it. Other activities in this category may need explicit re-
view and supervision by someone other than the people
Activities in this category do not receive any added engaged in them. When a QI activity takes more re-
oversight other than discussion among the participants, sources, has the potential to impose more disruption or
and they may not even be written down in any formal risks, or makes use of unusually sensitive protected health
manner; however, the clinicians who engage in them are information, supervisors and immediate managers should
responsible for protecting the participants by following the be aware and should take the opportunity to be sure that
ethical requirements of their profession and their organi- the endeavor is prudent and that the participants are
zation. It is worth noting that the monitoring initiated by brought in on a fair and appropriately informed basis.
the clinicians carrying out the QI activity ensures that pa- Again, two examples:
tients’ interests are protected. If anything were going awry,
 After doctors develop patient test order tracking sys-
the altered process would quickly be abandoned.
tems, they ask the nurses who care for their patients to
use them. The nurses point out that using a dozen dif-
Local management review and supervision of QI
ferent doctors’ systems would create complexity for
This category consists of activities designed to improve
them and lead to mistakes. They recommend that test
care in the local setting that require at least some monitor-
order tracking be made into a QI activity to develop a
ing by management. The minimum management element
common tracking system incorporating the best features
in this category is a method for tracking the QI activities
of the individual systems. The QI activity is then carried
in progress. In a small organization, this could be as sim-
out under the supervision of the management level that
ple as an informal discussion with a QI manager or other
designated executive staff member. In a larger organiza-
tion, it could take the form of standardized reporting Organizations should establish specialized
about QI activities to a quality management staff member.
To establish a physical record of QI activities, a simple Quality Improvement IRBs to review activities
registration process might be used. Such a process could
also serve as an educational tool to help people doing QI that are both QI and research on human
think through the design of the activity and recognize the
ethically important aspects, and as a screen to identify ac- subjects.
tivities with attributes that indicate the need for more
elaborate review and approval. In a small organization,
covers all parts of the organization affected by the
registration could take the form of answers to a set of
change, with one supervisory task being review of the ac-
structured questions kept in a locally available file. QI ac-
tivity’s design for conformity with ethical requirements.
tivity registration that met the requirements outlined by
the questions in the registration form would constitute
 An activity that imposes no more than minimal addi-
approval to perform the activity. In a large organization,
tional risks on patients does not require specific in-
registering and informing management might be accom-
formed consent on ethical grounds, but sometimes an
plished through an interactive, Web-based registration
informal consent to additional data gathering or minor
form. The initiator of the activity would enter a short de-
annoyances is appropriate. “We’re working on improv-
scription on the form and answer a set of questions. As
ing the way we move patients through the radiology de-
each question is answered, the individual would be direct-
partment. Would you mind filling in a brief survey
ed to carry out the activity or to obtain a specific form of
about your experience during today’s visit?” For such QI
oversight, depending on activity characteristics and poten-
activities, management may wish to introduce measures
tial for risks. After completion, the form would be stored
to ensure that any written surveys and patient interviews
in a database to be used as a tool for ongoing monitoring
are properly structured and that the activities are coordi-
of QI activities, as a source of information about past and
nated so that even minor burdens are spread fairly across
present QI activities for members of the organization, and
as a basis for systematic organizational reflection on the
patterns and outcomes of its improvement work. As a QI activity changes along the dimensions of in-
For some activities, registration is sufficient. If a regis- creasing levels of organizational resource use, methodolog-
tered QI activity uses few organizational resources, impos- ical complexity, potential health risk, burden of participa-
es minimal additional risks on patients besides the use of tion, and confidentiality risk, more intense review and
protected health information, and has confidentiality oversight becomes necessary in order to ensure that the QI
measures in place, it would be unusual to require further practitioners understand the ethical requirements and that

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S23

Exempt Human
HuSR Research
(HuSR) Activities Involving
Human Data Sources

Note: The figures are not drawn to scale.

Figure 2.

the QI activity meets them. When QI activities have sub- committee could be the organization’s IRB or the QI-IRB
stantial resource or staffing implications, major changes in described in the section below. An outside perspective
the providers’ or patients’ expectations, or more than min- would also be useful in cases where an organization’s staff
imal additional risk to patients compared to ordinary clin- members lack the technical skills to assess the benefits and
ical care, then they require active approval by management burdens of the activity; in such a case, the organization
above the level of the immediate managers involved with- might contract with a group such as a quality improve-
in the organization. The registration process outlined ment organization.
above could identify activities with characteristics that We believe that an organization should develop its own
make additional management approval necessary. De- approach to bringing the QI activities in this category into
pending on an organization’s structure, form, or size, ap- conformity with ethical standards. QI methods themselves
proval could be provided by a quality management staff attach considerable importance to identifying the compo-
member, QI director, facility manager, or other staff mem- nents of the existing system that must be involved in a QI
ber who is schooled in QI methods and the ethical re- activity in order for it to be done well. In each organiza-
quirements for QI. Clinical leaders most responsible for tion, the categories that managers use will be defined by
the patient population affected (a department chair or combinations of attributes, some relevant to ensuring eth-
nurse manager, for example) should also be part of the ical conduct of QI and others relevant to other superviso-
process of approval. ry goals. A wide variety of QI management configurations
Activities that pose more than minimal risks compared are likely to emerge. As long as these different configura-
to those in ordinary clinical care require specific informed tions are effective in ensuring that QI activities meet ethi-
consent and explicit review by at least one person who is cal requirements, they should be acceptable.
not directing the project and who is knowledgeable about It is worth noting that, from an ethical perspective, ex-
both ethical issues and management perspectives. This is ternal review of individual QI activities is not generally in-
true even if the proposed project also holds the potential herently better than internal review. On the one hand, an
for substantial benefits to the subjects. The possible harms outsider might provide a useful community perspective on
make it desirable to create “distance” between the organi- the acceptability of the risks involved, and for some kinds
zational units that favor the proposed changes and the of risks, this perspective should definitely be sought. On
oversight structure. A patient safety committee or clinical the other hand, an outsider might not understand the in-
policy committee could perform a structured review. In ternal environment well enough to understand the true
any case where an outsider might think the internal nature of the risks and benefits and the relationship of the
process could involve a conflict of interest, it would be ap- activity to the organization’s current standard of care and
propriate to refer the proposed activity to a committee allocation of resources. As we will discuss below, there is a
with representation from outside the organization. This need for external oversight, but it should be primarily fo-


Research QI Clinical &
QI/ Managerial
Research Innovation &
Research Research Adaptation
on QI on QI

Nonexempt QI Involving
QI/HuSR Human Data Sources

Human Subjects
Research (HuSR)

Activities Involving
Human Data Sources

Note: The figures are not drawn to scale.

Figure 3.

cused on review of the structure and functioning of the or- interaction with the individual, or identifiable private in-
ganization’s procedures for ensuring the ethical conduct of formation.” In the figure, the long oval is the set “Activi-
QI, rather than on the review of individual QI activities. ties Involving Human Data Sources,” where a human data
source is defined as “a living individual about whom a per-
QI-IRB review and supervision son conducting the activity obtains data through interven-
As noted earlier, some activities are both QI and re- tion or interaction with the individual, or identifiable pri-
search involving human subjects. The DHHS regulations vate information.” (The awkward term “human data
require any activity meeting the definition of human sub- source” is used rather than “human subject” because, as
jects research to be treated as such even when it is also an- shown, the regulations define a human subject as someone
other kind of activity, such as QI, education, or public about whom data is obtained for research purposes.) The
health surveillance. Therefore, these hybrid activities must overlap between this long oval and the circular Research
comply with the regulations if they are supported by fed- set (defined earlier for Figure 1) is the subset “Human
eral funds or are covered by the assurance many organiza- Subjects Research.” Within this subset, the smaller oval la-
tions give that all their human subjects research will com- beled “Exempt” consists of human subjects research that
ply with the DHHS regulations whatever the funding falls into any of the categories of research that have been
source. officially declared exempt from the regulations. (These in-
Unfortunately, it can be difficult to determine whether clude, for example, some research in educational settings
an activity is only QI or both QI and human subjects re- using normal educational practices; some research involv-
search, given the ambiguity of the phrase “designed to de- ing tests, surveys, interviews, and public behavior observa-
velop or contribute to generalizable knowledge” in the reg- tions; some research using existing data and specimens;
ulatory definition of research. The matter is of some con- and some research by public benefit programs.42) The re-
sequence, given the sanctions that may be imposed on or- maining area is the subset of human subject research ac-
ganizations that violate the human subjects protection reg- tivities covered by the regulations.
ulations. This section considers approaches to review and Figure 3 combines the diagrams in Figures 1 and 2 to
supervision of the activities that belong in this hybrid cat- show the cross-hatched subset “Nonexempt QI/Human
egory. The section that follows discusses approaches to Subjects Research Overlap (Nonexempt QI/HuSR),” the
their identification. set of QI activities that must be reviewed by an IRB and
We begin with a brief discussion of the activities to conform to other regulations, such as the requirements re-
which the regulations apply (Figure 2). The regulations lating to informed consent. Although the figure does not
define a human subject as “a living individual about show it, the boundaries of this subset are currently fuzzy,
whom an investigator (whether professional or student) given the lack of clarity about precisely which QI activities
conducting research obtains data through intervention or also meet the definition of human subjects research.

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S25
How do we ensure that all of the hybrid activities con-
6. Requirements for IRB Membership form to the regulations, especially given the fuzziness of
From 45 CFR 46.107 the boundary? Since the current IRB review process is not
suitable for most QI, simply sending all QI activities to ex-
isting IRBs would not be a good approach. A second ap-
 A minimum of five members with varying back- proach would be to send no QI activities to IRBs, and in-
grounds must sit on the board. stead authorize the development of a robust but separate
internal accountability system. This separate system would
 Members must not all be from one profession. be based on the ethical requirements developed in this re-
port (Box 5) and designed for all types of QI, including
 At least one member must have primary concerns activities that are both QI and human subjects research.
that are “scientific” (a term the regulations do not de- The hybrid activities could then be exempted from the
fine). DHHS regulations but required to submit to this alterna-
tive pathway. Unfortunately, modification of the existing
 At least one member must have primary concerns DHHS regulations to allow this is a project that would
that are “nonscientific” (also left undefined). take years and is not assured of a satisfactory outcome.
A third approach, and the one we recommend, is for
 At least one member must be unaffiliated with the organizations to establish specialized Quality Improve-
institution in which the IRB resides and not an im- ment IRBs (QI-IRBs) to review the hybrid activities. The
mediate family member of a person affiliated with QI-IRB would be a committee that meets the minimum
the institution. regulatory requirements for IRB composition and has
standard operating procedures that conform to IRB regu-
 Members cannot have a conflicting interest in any lations; however, it would design its process specifically to
project that comes before the IRB for initial or con- meet the review and supervision needs of an activity that
tinuing review. is both QI and human subjects research. This approach
seems compatible with the current regulatory framework.
 The IRB may invite individuals with competence Although many IRBs have chosen to operate under rela-
in special areas to assist in the review of issues that re- tively rigid rules and procedures, there is actually substan-
quire expertise beyond or in addition to that available tial flexibility in the regulations. OHRP’s Guidance on
on the IRB, but these individuals may not vote with Written IRB Procedures, July 11, 2002, summarizes the ac-
the IRB. tivities IRBs must carry out (the distribution of materials
on research projects to IRB members, initial and continu-
 The IRB shall be sufficiently qualified through its ing review of the materials, issuance of approval or disap-
members’ experience, expertise, and diversity, includ- proval of the project, review of changes to the research
ing consideration of race, gender, and cultural back- protocol, and so on) but also says:
grounds and sensitivity to such issues as community
OHRP has not developed a model written IRB proce-
attitudes, to promote respect for its advice and coun-
dures document for institutions to adapt because proce-
sel in safeguarding the rights and welfare of human
dures appropriately can vary significantly among institu-
tions as the result of differences in institution size, the
type of research activities, institutional administrative
 The IRB shall be able to ascertain the acceptabili-
practices, number of IRBs, and local and state laws and
ty of proposed research in terms of institutional com-
regulations. For each required element, the written IRB
mitments and regulations, applicable law, and stan-
procedures should provide sufficient step-by-step opera-
dards of professional conduct and practice, and shall
tional details so that an independent observer can un-
therefore include persons knowledgeable in these
derstand how an IRB operates and conducts its major
 If an IRB regularly reviews research that involves a The minimum regulatory requirements for IRB composi-
vulnerable category of subjects, such as children, pris- tion (shown in Box 6) allow IRBs to have as few as five
oners, pregnant women, or handicapped or mentally members and also to incorporate considerable flexibility
disabled persons, consideration shall be given to the into the specification of member qualifications. Thus the
inclusion of one or more individuals who are knowl- specialized IRB—that is, an IRB whose members are cho-
edgeable about and experienced in working with sen to have the expertise needed to review the specific
these subjects. types of research activities usually submitted to that
IRB—is compatible with the regulations.


Organizations could make use of the flexibility in the This consent is given with the understanding that confi-
regulations to tailor the membership and process of the dentiality is protected and no identifiable participant in-
QI-IRB to the QI/HuSR activities conducted within their formation from the QI activity is shared outside the orga-
organizational environment. A QI-IRB should be inte- nization without participant authorization. Also, general
grated with the managerial and professional supervisory information about the QI program is available, and it is
structure of the organization and should have members expected that participants in a QI activity will be in-
who are thoroughly familiar with the local environment. formed of any findings that are clinically relevant to their
A QI-IRB should also have members with expertise in the own care. In other words, participants are already aware of
theory and practice of QI and at least one nonaffiliated and have consented to their possible inclusion in minimal
member who can provide an outside perspective. (Of risk QI with their rights and welfare protected. In these
course, all members must also have a basic understanding circumstances, the IRB might reasonably determine that it
of both the ethical requirements for QI and the human re- is unnecessary and even confusing to patients to require
search protection regulations.) With the right member- specific informed consent to a QI project that could be
ship and a flexible process for review and supervision in done ethically without it, simply because the QI activity
place, it should be relatively straightforward for the QI- also contributes to a minimal risk research endeavor.
IRB to ensure that the QI/HuSR activities it reviews are In fact, the IRB could reasonably decide that requiring
designed to meet ethical requirements. As the discussion informed consent would impose additional burdens, since
in Section II summarized in Box 5 indicates, the ethical obtaining specific informed consent has a cost in patient
requirements for QI and human subjects research have time and energy. The costs to the organization might also
much in common. make it impractical. Often, requiring specific informed
The one significant difficulty lies in the strong pre- consent to a hybrid activity would be a poor use of re-
sumption of voluntariness on the part of researcher and sources needed for patient care, as well as a disincentive to
research subject. When human subjects research is com- an organization to add low cost, minimal risk design ele-
bined with an activity in which participation cannot be ments to a QI activity that would allow it to serve a re-
completely optional, problems tend to arise in applying search purpose as well as bring about a local improvement.
informed consent requirements. In the case of QI/HuSR All in all, it seems reasonable for the IRB to interpret the
activities, the proper course is clear if the combined activ- term “not practicable without the waiver or alteration” to
ity presents more than minimal risk; specific informed apply, and thus to permit specific informed consent to be
consent is ethically required for either QI or human sub- waived for most minimal risk QI/HuSR activities in
jects research activities of more than minimal risk. If the health care settings.
hybrid activity presents no more than minimal risk, how- As in the local management review and supervision of
ever, the research regulations and QI ethical requirements the QI category, this category of QI-IRB review and su-
seem to conflict, since specific informed consent is not pervision can—and should—be constructed and used in
ethically required for a QI activity of that degree of risk. different ways in different organizations. The membership
In fact, informed consent is not quite an absolute re- composition and the operating procedures of the QI-IRB
quirement for research under the DHHS regulations. An should be appropriate to the nature of the organization
IRB may waive or alter the specific informed consent re- and the QI activities it conducts.
quirement if these four conditions all hold: (1) the re- An organization that carries out traditional human
search “cannot practicably be carried out” without the subjects research activities as well as QI/HuSR activities
waiver; (2) the subjects’ rights and welfare will not be ad- will probably find it worthwhile to establish both a QI-
versely affected; (3) the research involves no more than IRB and a separate IRB for research. The research IRB
minimal risk; and (4) when appropriate, subjects will be would not need to be as integrated into ongoing manage-
provided with additional pertinent information after par- ment of the organization as the QI-IRB, could meet less
ticipating. Unfortunately, we believe that at this time, re- frequently to review the fixed protocols typical of research,
search IRBs are likely to interpret these requirements nar- and would normally have a different composition in terms
rowly (particularly the “not practicable” requirement) and of scientific and statistical methodological expertise. If
be unwilling to grant waivers of informed consent for there are separate IRBs, there should be excellent lines of
QI/HuSR activities. communication between them to ensure that their mem-
We believe that a waiver would generally be appropri- bers have a basic understanding of the similarities and dif-
ate in cases in which the QI/HuSR activity represents no ferences between the roles of the QI-IRB and the research
more than minimal risk (and especially where strong a pri- IRB within the organization.
ori evidence suggests that substantial benefits to subjects Finally, each organization should have at least one per-
are likely). First, remember that consent to QI is required, son who is trained and authorized to sort activities—to
but takes the form of background consent to inclusion in decide which activities are required to go to the QI-IRB,
routine minimal risk QI activities, given as part of consent which are required to go to the research IRB (if there is
to receive health care from (or work for) the organization. one), which are not covered by or are exempt from the

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S27
DHHS regulations altogether, and which do not fall search but does not involve any people who meet the reg-
under the regulations but should nevertheless go to the ulatory definition of human subjects). The organization
QI-IRB because the organization deems them to need the itself sets the boundary between this category and the cat-
kind of review and supervision the QI-IRB provides. It is egory of professional oversight of QI, and it draws what-
not an efficient use of an IRB or QI-IRB to use it to per- ever lines it needs within the category to sort activities and
form this sorting task. The proposal to designate and train direct them to receive appropriate management and su-
one or more specific persons to perform the task is com- pervision.
patible with the DHHS regulations. Specialized QI-IRB review should eliminate some of
the problems in having QI/HuSR activities reviewed by
Separating QI from Research and Defining the Overlap research IRBs; nevertheless, clarifying which QI activities
We have proposed a specialized QI-IRB to review the are subject to the regulatory requirement of IRB review is
QI activities that are also human subjects research and are still important. Dithering over ambiguous classifications
subject to the DHHS regulations. Now we must address wastes time and tends to produce inconsistent decisions
the vexing question of how to determine whether a QI ac- that undermine respect for the review system. Moreover,
tivity falls into this category. if organizations cannot tell which activities are in the
This report has defined QI as “systematic, data-guided QI/HuSR category, they may overload the QI-IRBs with
activities designed to bring about immediate positive activities that don’t need to go there, just to be safe, when
changes in the delivery of health care in particular set- these activities could receive effective management and
tings.” So understood, QI is just a systematic, data-guid- supervision at lower cost within the purely QI oversight
ed form of the clinical and managerial innovation and system.
adaptation that has always been an integral part of clinical Developing a clear rule for separating QI from
and managerial practice. The fact that QI is a normal QI/HuSR can be seen as a conceptual task or as a practi-
health care operation focused on improving local care has cal regulatory task, but of course it is both. What should
been critical to our argument that ethical review of QI one look for in a sorting rule? The rule should use easily
should be incorporated into the system of accountability observed aspects of an activity to determine whether it be-
for clinical care. longs to the hybrid category—avoiding, for example, re-
We have noted that in the process of making local im- liance solely on the intent of the person initiating the ac-
provements, QI may produce information that is of use to tivity and focusing on concrete elements in the activity’s
people in other settings. The regulations define research as design or context. The rule should be as consistent as pos-
“a systematic investigation, including research develop- sible with the use of the word “research” in both common
ment, testing and evaluation, designed to develop or con- language and the regulatory definition, while openly ac-
tribute to generalizable knowledge.” Depending on one’s knowledging that some arbitrariness is inevitable in inter-
interpretation of the last clause, one might argue either preting the word to devise a practical rule for regulatory
that nearly all QI or that no QI should count as research. purposes. Finally, the rule’s arbitrary lines should be
We are persuaded that neither extreme is appropriate. QI drawn so as to best serve the end goal of protecting
reports can provide useful insights, just as individual pa- human participants—from both the harm that might be
tient case reports, success stories, and cautionary tales pro- caused by the activity and the harm caused by quality and
vide useful insights. Nevertheless, like these examples, safety deficits in the health care system.
most QI activities are not research as understood by those Some aspects of the history and purpose of human re-
who framed the human subjects protection regulations. search protection shape the possibilities. The origins of
Nevertheless, some QI activities are genuine hybrids: sys- the regulation of research to protect human subjects lie in
tematic investigations designed to bring about local im- risky biomedical research and public outrage over inci-
provement and develop generalizable knowledge simulta- dents in which researchers conducted research on patients
neously. Such activities are an extremely useful and cost- and members of the public without their knowledge or
effective way to deal with quality and safety problems, consent, or wasted public or shared resources, or took un-
and they must be encouraged if we want to make real reasonable risks with little prospect of gain. What was be-
progress in improving the American health care system. fore the National Commission in the late 1970s was a se-
Drawing lines separating categories of activities is a dif- ries of scandals and a growing research enterprise that had
ficult task in a regulatory system. We have tried to shape shown itself to allow, at times, threats to the wellbeing of
our proposals for ethical oversight of QI to minimize the its human subjects. In responding to this situation, the
need for bright lines separating categories of activities National Commission saw as the domain of research not
while remaining in accordance with current DHHS regu- just insights that were “generalizable” in that they would
lations. The category called local management and super- be true of some other situations, but facts that are essen-
vision covers QI that requires oversight but is not human tial to the nature of the human body and its reactions to
subjects research (either because it does not meet the reg- drugs and treatments, as well as other facts that would be
ulatory definition of research or because it qualifies as re- expected to be true for all time and in all places. In its re-


port on the provision of health services by the Depart- Below, we give an initial delineation of the divides, dis-
ment of Health, Education, and Welfare, the Commission cussing both the kinds of QI activities that should count
highlighted the importance of audit and improvement ac- presumptively as research, and those that should not. We
tivities and contrasted these activities with research, not- want to underscore that learning about these divides
ing that the practitioners in improvement activities did should entail some deliberate innovation with reporting
not have the conflicts of interest that make human sub- cases in the professional literature and having discussions
jects vulnerable to abuse in research.43 in policy settings. It is not likely that we will have hit upon
Over time, this understanding of research has altered. just the right dividing characteristics, or that we have ar-
The IRB-based human subjects protection system em- ticulated them well enough for implementation without
bodied in the DHHS regulations has been the only sub- case studies.
stantial professional or governmental activity that aims to For the purpose of the DHHS regulations, we propose
protect human subjects. When questions on the scope of that the category of research be made up of activities that
research have been posed to the regulators, virtually any are designed to learn something enduring about the nature
activity that might provide insights of use in another situ- and function of human beings and their environment. Re-
ation has ended up being included in the category of re- searchers might also be clinical service providers, but their
search potentially subject to the regulations. This process research role is to be kept distinct and identifiable to pa-
has engendered long-running controversies over, for ex- tients. Although the word “research” is used in more in-
ample, the status of ethnography, oral history, and report- formal ways, such as “library research for a high school
ing on public programs. The shift toward a very broad in- term paper” or “research by an investigative journalist,” we
terpretation of research is also seen in recent OHRP asser- think this definition is appropriate for implementation of
tions that public health surveillance is generally research, the DHHS regulations. The proposed narrative definition
and likewise in tentative OHRP determinations that re- accords with the regulatory definition but gives voice to
search includes administering surveys to clients as re-
quired under the Government Paperwork Reduction Act,
which establishes practices to limit fraud and to ensure “Research” refers to activities designed to
government contractors’ accountability.
Given the tendency to interpret research broadly and learn something enduring about the nature
the severity of the penalties available to OHRP and to
those who enforce the HIPAA Privacy Rule, organizations and function of human beings and their
have been afraid to develop their own guidelines for sort-
ing activities into the categories of QI, research, and environment. Under this interpretation, most
QI/HuSR. Though many parties know that improvement
activities are increasingly important and that imposition QI is not research.
of IRB requirements is a major barrier to their implemen-
tation, each organization is best served by avoiding any what could be meant by “designed to develop or con-
dispute with OHRP and any risk of penalty under tribute to generalizable knowledge.”
HIPAA. At the same time, OHRP has shown little incli- Under this interpretation, most QI is not research. QI
nation to provide highly specific guidance on how to do is designed to bring about the immediate improvement of
the sorting. care in local settings, and most of the urgently needed QI
Given the great variety and the significant change cur- activity involves changes in practice that are clearly within
rently under way in the kinds of QI and research being the standard of care—often moving from “dangerously
conducted in clinical settings (not just clinical research, substandard” and “barely acceptable” practices to “better”
but organizational, managerial, and health services re- and “best” practices. As a result, QI activities are generally
search), this may be a prudent decision on OHRP’s part. based on knowledge about the enduring “nature and func-
It would be very difficult to produce consistent, practical tion of human beings and their environment,” rather than
guidance at the regulatory agency level. We believe that designed to create new understandings in this regard.
such guidance would be more easily developed by com- Therefore, even though the activities represent “system
bining reflection on the nature of research and the aims of changes,” these changes are within the discretion of clini-
the human research protection system with the practical cians and managers to make. In fact, they have an ethical
experience in managing QI and QI/HuSR activities and sometimes also a legal duty to make such changes.
gained by a variety of organizations. The goal would be an Changes of this type include adoption of evidence-based
interpretation of the definition of human subjects research practice guidelines and best practices issued by respected
that enabled QI and QI/HuSR activities to receive the medical authorities, introduction of process modifications
type of review and supervision needed to ensure ethical designed to reduce the probability of medical error, and
conduct without imposing excessive costs. replacement of a practice that is dangerously below the
standard of care with one that is acceptable.

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S29
Introducing these changes may require adjustments in back is delayed or altered in order to avoid biasing the
practice over and above what is specifically mentioned in interpretation of data
a description of the new practice, and the adjustments are
likely to differ from one organization to another. QI is  Funding from an outside organization with a com-
about “moving fast but checking on progress as you go”— mercial interest in the use of the results
identifying the adjustments that may be required, tracking
what happens as practices change, and being alert for un- As outlined below in the section on strategies for imple-
expected effects—and thus may be the best way to protect mentation, this initial list deserves to be tested in some
patients as the system adjusts to the change. Ensuring that practical settings, aiming to modify it to fit the actual risks
this very specific and local knowledge is developed in a and projects that arise. Organizations with substantial in-
manner that appropriately protects patients is the essence vestment in overlap projects might well be among those
of QI. It is not an activity “designed to develop or con- best situated to develop QI-IRBs, which are described
tribute to generalizable knowledge,” and in fact, the idea above.
that implementing a new evidence-based guideline in a
local delivery setting is a research project requiring IRB re- Ethical Oversight for Collaborations across Organizations
view seems indefensible.44 This type of activity would not Multi-organization QI activities, such as QI collabora-
require review by a QI-IRB or a research IRB; it would tives, often combine QI activities focused on bringing
generally belong in the category of “local management re- about change in individual organizations into a larger pro-
view and supervision of QI.” ject with an overarching goal of sharing information
Nevertheless, some QI activities really do have a re- about the experience of bringing about positive change.
search component in the sense that they are designed to The collaborative structure may provide for the initial
learn something enduring about the nature and function transfer of technical information about a new practice to a
of human beings and their environment in addition to QI team in each organization, group discussion of how
improving care in the local setting. Perhaps a project is de- best to introduce it using QI methods, and periodic group
signed to improve compliance with guidelines regarding a meetings to boost morale and exchange information
particular treatment, but includes design elements that about failures and successes.
allow the formal testing of a few strategies to gain that The initiators of the collaboration should take the re-
compliance, such as having some sites use QI (while oth- quirements for ethical conduct of QI into account in the
ers use other methods). Perhaps a project employs a new design of the entire enterprise. In deciding whether to par-
treatment that has not yet been fully accepted as the stan- ticipate, each organization should subject the QI activity
dard of care, but embeds that treatment in a QI process at the core of the collaboration to its internal QI account-
aimed at bringing about compliance with other aspects of ability process (assigning it to the appropriate category of
care. Projects like these might well require review as re- review and supervision). The aim is to ensure that the ac-
search, as well as review as QI. tivity is consistent with organizational goals and is de-
What characteristics of projects would make it likely signed so that participants within the organization are
that the project is both QI and research and should be re- properly protected.
viewed by a QI-IRB? Here is a starter set. The main additional issue raised by collaborating with
other organizations concerns the sharing of information
 Randomization of patients into different intervention
about the QI activity outside the organization. Care must
groups in order to enhance confidence in differences
be taken to ensure patient confidentiality and conformity
that might be obscured by nonrandom selection (but
to HIPAA rules. The initial transfer of technical informa-
not randomization done to achieve equitable allocation
tion to participants can include assistance in clarifying the
of a scarce resource)
requirements for ethical conduct of the activity, including
the management of protected health information in the
 Testing of issues that are beyond current science and
sharing of results. If patient information might sometimes
experience, such as new treatments
be shared, or if follow-back to identify patients is plausi-
ble, then the affected participants should be bound to
 The involvement in key project roles of researchers
honor confidentiality through a Business Associate Agree-
who have no ongoing commitment to improvement of
ment or an Organized Health Care Arrangement.45
the local care situation (and who may well have conflicts
Typically, QI collaboratives generate substantial insight
of interest with the patients involved), even if others on
about the effective strategies for QI around a given topic,
the team do have professional commitments
but this does not automatically mean that the collabora-
tive project amounts to research involving human subjects
 Delayed or ineffective feedback of data from moni-
and is therefore subject to IRB review. In general, it does
toring the implementation of changes, especially if feed-
not even make it research, since the insights are still tied to
place and time and are acquired while pursuing improve-


ment in the clinical services. Even when the design does (NCQA), which accredits managed care plans, and the
have some characteristics of research (such as a starting Joint Commission on Accreditation of Healthcare Orga-
agreement to implement some key changes in a staggered nizations (JCAHO), which accredits most types of health
fashion so that their effects can be identified), the fact that care organizations, could be modified to add to their re-
the organizations cooperate to share insights about the view of the conduct of QI in organizations a review of the
process of change within each organization would usually extent to which the organization has procedures in place
make any research component an instance of research on to ensure that the QI meets ethical requirements. Eventu-
organizational behavior, not research on human subjects. ally, outcome and process indicators for ethical quality
improvement could be part of a set of regularly collected
External Accountability for quality measures.
Ethical Conduct of QI Public sector mechanisms for accountability include
the federal HIPAA Privacy Rule and other federal and

T he established accountability approaches used by the

health care system to measure, identify, and demon-
strate quality and safety in the delivery of clinical care
state statutory privacy laws that regulate the protection of
identifiable health information. The HIPAA Privacy Rule
requires that “covered entities”—persons and organiza-
should be used to hold health care professionals, man- tions that acquire, use, disclose, or store health informa-
agers, and organizations accountable for the ethical con-
duct of QI.46
A first line of accountability is through the profession- The established approaches used by the
al organizations of those who work in health care—physi-
cians, nurses, health care managers. These organizations health care system to ensure quality and
should inform their members about their professional
obligation to improve quality, identify the basic QI skills safety in clinical care should be used to
they should have, and ensure that they understand and
uphold the standards for ethical conduct of QI. Quality establish accountability for the ethical
improvement can be incorporated into professional
Codes of Ethics. For example, the American Nurse Asso- conduct of QI.
ciation Code of Ethics states that nurses should “partici-
pate in establishing, maintaining, and improving health- tion—protect the individually-identifiable health infor-
care environments and conditions of employment con- mation they create or receive by establishing and adhering
ducive to the provision of quality health care and consis- to privacy protections. These protections include provid-
tent with the values of the profession through individual ing notice to patients about their privacy rights and the
and collective action.” The Code of Ethics of the Ameri- entity’s policies regarding disclosure of identifiable infor-
can Medical Association offers as one of nine core princi- mation, implementing internal privacy policies and pro-
ples the rule that “A physician shall continue to study, cedures, establishing safeguards to protect data privacy,
apply, and advance scientific knowledge,” and on the basis training employees to understand privacy laws, and assist-
of this statement, AMA policy calls on physicians and ing health consumers to exercise their rights under the
their organizations to “(1) strive continuously to improve Privacy Rule. The Privacy Rule allows covered entities to
the quality of health care; (2) encourage the ongoing eval- use protected health information to perform activities de-
uation of continuous quality improvement models; (3) fined as normal health care operations provided specified
promote implementation of effective quality improve- procedures are followed and includes QI in this category.
ment models; and (4) identify useful approaches for as- Research is not included in normal health care operations
sisting physicians in implementing quality improvement and is subject to different standards regarding the use and
procedures in their medical practices and office manage- disclosure of identifiable health information.
ment.”47 Some authors have argued that privacy risks may be
Accountability at the organizational level can be fos- the key potential harm to patients from quality improve-
tered through the accrediting bodies that set standards for ment projects. The broad applicability of the Privacy Rule
health care organizations, assess compliance with those across the health care system could protect patients from
standards, and in some cases focus on the operation and privacy breaches when their data are used for quality im-
effectiveness of internal quality improvement systems. In provement initiatives. The HIPAA Privacy Rule includes
some areas, state and federal governments rely on or rec- provisions for the protection of data from unauthorized
ognize private accreditation to ensure compliance with li- review, requires that only the “minimum necessary” data
censure or regulatory requirements. To encourage ac- be used, and introduces HIPAA privacy boards to moni-
countability for the ethical practice of quality improve- tor the use of protected health information within organi-
ment, the standards of the major private organizations zations. It has been suggested that a HIPAA privacy board
such as the National Committee for Quality Assurance could adequately review QI projects and even research

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S31
projects for which the only significant risk to patients was factor in their choice of provider. This effect is limited,
from a breach of privacy and/or confidentiality. however, since the information available to patients is also
For those QI activities that fall into the QI/HuSR limited, and they rarely have full control over decisions
overlap category, the federal human subject protection about the care that is purchased on their behalf. Market
regulatory system acts as an external accountability mech- forces are most likely to have an effect when they act
anism for the specialized QI-IRBs proposed as part of the through third party purchasers. Health care purchasing
internal accountability structure for the conduct of QI groups and other payers could encourage quality im-
within organizations. provement and its ethical practice by including outcome
Finally, the health care market may enforce a degree of and process standards for the ethical practice of QI in the
accountability through the ability of purchasers to make contracts they make with providers.
the observance of ethical standards in the conduct of QI a

IV. Implementation
 Clarify professional and organizational obligations to-

mproving the quality and safety of American health
care is of vital importance. Health professionals, man- ward QI
agers, health care delivery organizations, patients, and Organizations of health professionals (physicians, nurs-
government all have an ethical responsibility to cooperate es, health care managers, and so on) should inform mem-
with one another in pursuing this goal. In this report, we bers about their professional obligation to improve quali-
have discussed the ethical issues raised by the use of QI ty. They should also identify the basic QI skills their
methods to improve health care. Using the principles of members should have, educate members about standards
research ethics that underlie the current human research for ethical conduct of QI, and even incorporate quality
protection system as a point of departure, we have defined improvement into professional codes of ethics. Organiza-
the requirements for ethical conduct of QI activities and tions of health care provider organizations (hospitals,
then considered the institutional arrangements needed to nursing homes, health plans, and so on) should educate
ensure that QI conforms to them. We have discussed the their members about their quality improvement obliga-
use of IRBs for ethical oversight of the conduct of QI and tions, the need to ensure that their employees have basic
concluded that IRBs as currently constituted are not ap- QI skills, and the standards for ethical conduct of QI.
propriate for this purpose. Instead, we argue that ethical Leaders in professional education should press for greater
review of QI activities for purposes of protection of emphasis on the obligations of health professionals toward
human subjects should be integrated into a trustworthy the quality of care and the development of QI skills in ed-
accountability system for managing the quality of clinical ucational curricula.
We recommend that the primary responsibility for the  Clarify patients’ obligations toward QI
ethical conduct of QI be lodged in individual organiza- When people seek health care, they should be told
tions, where it should be seen as a normal clinical care about the importance of QI activities for maintaining and
obligation and integrated into supervision and manage- improving the quality of their care and informed that con-
ment, with the organizations’ leaders responsible for over- sent to receive care includes consent to a minimum level
seeing the integration. We call this “internal accountabili- of cooperation with ongoing QI. They should be given
ty” for the ethical conduct of QI, and we distinguish three basic information about the organization’s QI program
broad categories of oversight: professional responsibility and told how they can obtain more information if they
for QI; local management review and supervision of QI; want it.
and QI-IRB review for QI that is also human subjects re- Health care organizations should develop educational
search. We recommend that there also be external ac- materials for their patients on the conduct of QI within
countability for the ethical conduct of QI in the form of an organization and on patient rights and responsibilities
procedures to ensure that individual health care organiza- with respect to QI. Organizations, such as JCAHO and
tions have well-designed, functioning internal practices in the Institute for Healthcare Improvement (IHI), and asso-
place to carry out their QI-related responsibilities. These ciations of organizations, such as the American Hospital
procedures should be part of the overall system of holding Association (AHA), might assist with this by promoting
health care providers accountable for the quality and safe- the development and dissemination of model educational
ty of clinical care. materials that individual organizations could adapt for
We have recommended a variety of specific steps by their own use.
which the ethical conduct of QI might be ensured, but
these recommendations bear summarizing.


 Develop guidance on QI methodology tled and the interpretations of the language of the DHHS
QI uses—and should use—a somewhat different range human subjects regulations become clearer. This work
of methodologies from traditional research. Moreover, as a would be greatly enhanced by robust discussion in the
relatively newly rediscovered field of endeavor, QI’s meth- professional literature and at meetings. The topics dis-
ods are in a state of rapid evolution. QI practitioners need cussed would include the criteria used to sort projects, the
ongoing guidance on appropriate methodological stan- merits of particular methods of review and supervision,
dards in order to meet the ethical requirement that QI ac- and actual case studies that illuminate the work.
tivities be properly structured to achieve their goals. The cooperation of OHRP would be very important,
Groups that are already active in developing and promot- especially for efforts to develop practical guidance on the
ing QI and educating people about QI methods, such as classification of activities as QI or QI/HuSR. One might
IHI and AHRQ, are well placed to perform this function. expect that OHRP would be especially anxious about such
a process, since it has been a long practice to extend the
 Develop new models of management and supervision of scope of research broadly. However, keeping OHRP in-
QI and QI/HuSR volved and reflecting upon the actual situations that arise
We have given a rough outline of the internal account- would be helpful in developing sound public policy. Hav-
ability arrangements that should be in place to ensure that ing AHRQ or another agency sponsor some ongoing de-
QI and QI/HuSR activities receive appropriate oversight. scriptive research into the nature of the projects and the
The next step is to translate this rough outline into mod- review approaches that are implemented would help guide
els that will work in real health care settings. We suggest the process. If some adverse occurrences arise during this
that one promising way to move this forward is through process of exploration and consensus building, then a co-
collaborative efforts to develop such models by organiza- ordinated response among the concerned parties would
tions that are leaders in introducing QI methods into their help to ensure that all the issues are carefully weighed and
operations. that overly rigid rulings are not made and transformed
OHRP could take the lead in encouraging some lead- into permanent precedent.
ership organizations to undertake the task, or the leader- As consensus develops on promising models for the
ship organizations could take the initiative themselves. management and supervision of QI and QI/HuSR and on
IHI might be able to play a coordinating role, perhaps rules for sorting activities into the appropriate category,
along the lines of the role it plays in its “Breakthrough IHI, JCAHO, OHRP, the various trade organizations,
Collaboratives.” The goal would be for the individual or- and professional journals should disseminate the results.
ganizations to develop systems of internal management Public Responsibility in Medicine & Research
and supervision for their QI activities, establish QI-IRBs (PRIM&R) could also help to educate IRBs and the re-
for activities that are both QI and human subjects re- search community in general on the similarities and dif-
search, devise practical rules for identifying the activities ferences between QI and research, and the standards for
that should be reviewed by QI-IRBs, and develop QI-IRB ethical conduct of QI. In particular, PRIM&R could help
review processes tailored to the types of QI/HuSR carried the IRB community to understand the developing con-
out in their organizations. As they launched these efforts, sensus on the management of activities that are both QI
the organizations could share their experiences with one and human subjects research, and ensure ongoing consid-
another regularly, engaging in robust discussion of the eth- eration of the issues such hybrid activities raise through its
ical issues they encounter and the practical wisdom they Annual Human Research Protection Program (HRPP)
generate. Ideally, organizations such as OHRP, AHRQ, Forum.
and JCAHO would also be involved in this endeavor.
A federal agency could also undertake to lead an effort  Develop and expand external accountability for QI
to develop practical rules and models, under the exemp- Accrediting bodies such as JCAHO and NCQA have
tion in the DHHS regulations for public benefit pro- already begun to include quality improvement require-
grams. This effort could be instead of, in addition to, or ments in the accreditation process. They should continue
even in collaboration with an effort by private organiza- to develop and expand these efforts to include review of
tions. The obvious candidate would be the Centers for the extent to which organizations have effective mecha-
Medicare and Medicaid Services, since it sponsors QI and nisms in place for managing QI and ensuring that it meets
research and has accepted the mission of improving care ethical requirements in the accreditation process.
for Medicare and Medicaid patients. However, the Health
Resources and Services Administration (HRSA), the De-  Disseminate QI results
partment of Veterans Affairs, and the Department of De- The ethical analysis in this report supports the conclu-
fense could also carry out such efforts under the public sion that the intent to publish (or actual publication) does
benefit agency exemption. not make a QI activity into a research project, nor does it
The consensus of this work should be clearly articulat- establish that IRB review is ethically required. The analy-
ed from time to time, as the categories become more set- sis also implies that QI results can have social value outside

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S33
the setting in which they were developed, and therefore, QI articles would also promote the development and clar-
dissemination by a variety of means should be encour- ification of QI methodological standards.
aged. There has been considerable confusion in the past Editors of medical and other health care-related jour-
about the relationship of publication to IRB review and nals could adopt a standardized format for the reporting
the designation of QI as research; however, at this point, of QI activities for publication, encourage the submission
OHRP and others now seem to be in agreement with of articles on QI methods and results, and become edu-
these conclusions.48 cated about what falls outside of the QI/HuSR range so
The next step is to develop strategies that facilitate the that they do not reject articles that report QI activities just
sharing of QI results. For example, publication of QI re- because they were conducted without informed consent
sults should be encouraged, and the results should be pub- or IRB review. AHRQ, CMS, and other federal agencies
lished in a standardized format so that it is easier to com- could develop and implement ways of providing financial
pare different QI activities and evaluate the extent to support for the conduct of QI, the dissemination of re-
which the conduct of the activities met ethical require- sults, and the development of QI methods, and provide
ments.49 By allowing wider scrutiny of the methods used ongoing guidance on the application of HIPAA privacy
in QI, removal of the current bias against publication of rules to QI activities that involve collaboration across or-

V. Conclusion

 Some projects are correctly counted both as QI and as

lthough the findings and conclusions we have ar-
rived at during the course of our research on ethi- research involving human subjects and should meet the
cal issues and health care quality improvement are requirements for review of protection of human subjects
complex and nuanced, they can be summarized as follows: in research.
 QI is any systematic, data-guided activity that is de-
 Meeting those requirements might be more readily
signed to bring about the immediate improvement of
accomplished with a QI-IRB that met regulations, but
care in a local setting.
whose policies and procedures were also tailored to the
needs and expectations of QI.
 QI is both appropriate and vital to health care.
 Certain issues might trigger the requirement for for-
 QI is marked most distinctly by the prompt feedback
mal review of a proposed QI project: randomized de-
of the effects of deliberate changes to the same care de-
signs, novel treatments, involvement of researchers, de-
livery setting that is making the changes.
layed feedback of monitoring, or external funding.
 QI is intrinsic to health care delivery and obligatory
 Federal agencies and voluntary organizations should
for both professionals and patients.
cooperate in further developing and implementing these
 Though QI is often driven by a priori evidence that
suggests substantial benefits are likely for the patients
Resolving the uncertainty about what is ethically and
and/or staff involved, QI can pose some risks to some
legally required to safeguard participants in QI activities is
essential if QI methods are to be available to help trans-
form the culture of American health care delivery into a
 Not undertaking QI in the face of recognized quality
culture committed to continuous quality improvement.
deficiencies also puts patients at risk.
In this report, we have outlined what must be done to
protect patients’ and staff members’ rights to privacy, in-
 QI should itself be implemented ethically.
formed consent, and justice when carrying out QI activi-
ties. The challenge now is for the regulatory community
 Low-risk QI should generally have the same review
and health care management and professionals to put the
and standards as routine health care delivery.
appropriate standards and practices for QI in place. The
goal—a better and safer health care system—is worth the
 Higher-risk QI should undergo routine and orderly
review within the usual arrangements for clinical super-
vision or by an advisory group.


References 12. Agency for Healthcare Research and 20. We acknowledge that the United
Quality, 2003 National Healthcare Quality States has so far failed to actually fulfill this
1. F. Davidoff and P. Batalden, “Improv- Report (Rockville, Md.: U.S. Department of obligation. Nevertheless, the ongoing debate
ing Improvement: The Educational Chal- Health and Human Services, 2003). about health care reform and the existence
lenge for Medicine,” manuscript submitted of programs like Medicare, Medicaid, State
for publication. 13. Agency for Healthcare Research and
Quality, 2004 National Healthcare Quality Children’s Health Insurance Programs, and
2. Because fourteen other federal depart- Report (Rockville, Md.: U.S. Department of other programs that provide assistance to
ments and agencies have codified Subpart A Health and Human Services, 2004); Agency various categories of the poor are evidence of
in their own departmental regulations, it is for Healthcare Research and Quality, 2005 recognition of an obligation, although there
known as the “Common Rule.” An addi- National Healthcare Quality Report is no consensus about the extent of it or the
tional agency (the Central Intelligence (Rockville, Md.: U.S. Department of public policies that should be used to fulfill
Agency) is required by executive order to Health and Human Services, 2005). it.
comply with all parts of 45 CFR 46. In ad- 21. See N. Dubler et al., “Informed Par-
dition to these human subjects protections, 14. Agency for Healthcare Research and
Quality, 2003 National Healthcare Dispari- ticipation in the QI Process,” background
some federal agencies have issued regula- paper prepared for The Hastings Center
tions specific to human research conducted ties Report (Rockville, Md.: U.S. Depart-
ment of Health and Human Services, project, “The Ethics of Improving Health
under their jurisdiction (for example, the Care Quality and Safety.”
Environmental Protection Agency regarding 2003); Agency for Healthcare Research and
pesticide research), and several states have Quality, 2004 National Healthcare Dispari- 22. See IOM, Crossing the Quality Chasm,
statutes and/or regulations governing re- ties Report (Rockville, Md.: U.S. Depart- 5-6.
search with humans (such as the California ment of Health and Human Services, 23. The argument here contrasts with the
Health and Safety Code). 2004); Agency for Healthcare Research and position in the ethics of QI in which great
Quality, 2005 National Healthcare Dispari- importance is attached to the extent to
3. See M.P. O’Kane, “Do Patients Need ties Report (Rockville, Md.: U.S. Depart-
to be Protected from Quality Improve- which an individual QI participant, or a ma-
ment of Health and Human Services, jority of participants, benefits directly from
ment?” background paper prepared for The 2005).
Hastings Center project, “The Ethics of Im- the specific QI activity. See D. Casarett, J.H.
proving Health Care Quality and Safety.” 15. C. Schoen et al., “Taking the Pulse of Karlawish, and J. Sugarman, “Determining
Health Care Systems: Experiences of Pa- when Quality Improvement Initiatives
4. K.N. Lohr, ed., Medicare: A Strategy for tients with Health Problems in Six Coun- Should Be Considered Research: Proposed
Quality Assurance (Washington, D.C.: Na- tries,” Health Affairs 24 (2005): w509-w525 Criteria and Potential Implications,” Journal
tional Academy Press, 1994), quoted in In- (published online November 3, 2005; of the American Medical Association 283
stitute of Medicine, Crossing the Quality 10.1377/healthaff.w5.509). (2000): 2275-80.
Chasm (Washington D.C.: National Acade-
my Press, 2001), 232; see also Institute of 16. Professional societies have recently 24. See Department of Health and
Medicine, Envisioning the National Health made this clear. See ABIM Foundation, Human Services, “Code of Federal Regula-
Care Quality Report (Washington D.C.: Na- ACP-ASIM Foundation, and the European tions: Protection of Human Subjects,” June
tional Academy Press, 2001). Federation of Internal Medicine, “Medical 23, 2005,
Professionalism in the New Millennium,” humansubjects/guidance/45cfr46.htm#46.
5. See IOM, Crossing the Quality Chasm, Annals of Internal Medicine 136 (2002): 102, section 46.102(d). Emphasis added.
5-6. 243-46; Institute of Medicine, Health Profes- 25. See Dubler et al., “Informed Partici-
6. Ibid. sionals Education: A Bridge to Quality (Wash- pation in the QI Process.”
7. See IOM, Envisioning the National ington D.C.: National Academy Press, 26. Research on the quality improvement
Health Quality Report, 9. The committee 2003); and the description of “the six com- process is sometimes referred to as “QI Re-
notes that “specific causes of inefficiency, petencies” developed by the Accreditation search,” but we prefer the terminology of
such as repeat procedures due to error, Council for Graduate Medical Education “Research on QI” to distinguish this set
overuse, fragmentation of care, and unneces- (ACGME) and endorsed by the American clearly from the set, “QI/Research,” the
sary delays, are included under the appropri- Board of Medical Specialties (ABMS) at: overlapping region of the QI and Research
ate component of quality.” sets.
8. The issue of changes that involve trade- compFull.asp.
27. See E.J. Emanuel, D. Wendler, and
offs between quality and cost—that is, ac- 17. See President’s Commission for the C. Grady, “What Makes Clinical Research
cepting a lower level of quality in exchange Study of Ethical Problems in Medicine, Se- Ethical?” Journal of the American Medical As-
for lower resource costs—is discussed later curing Access to Health Care, Volume One: Re- sociation 283 (2000): 2701-2711. We used
in this report. port (Washington D.C.: U.S. Government the framework of this article because it is a
9. M.A. Schuster, E.A. McGlynn, and Printing Office, 1983). brief, readily accessible, recent synthesis of
R.H. Brook, “How Good Is the Quality of 18. See M. Wynia et al., “Physician Ethics generally accepted ideas in research ethics
Health Care in the United States?” Milbank and Participation in Quality Improvement: and includes all the topics we wish to discuss
Quarterly 76 (1998): 517-63. Renewing a Professional Obligation,” back- in this report. We recognize that it is a varie-
10. E.A. McGlynn et al., “The Quality of ground paper prepared for The Hastings gated list of topics consisting of fundamen-
Health Care Delivered to Adults in the Center project, “The Ethics of Improving tal ethical principles and both substantive
United States,” New England Journal of Health Care Quality and Safety.” and procedural norms. For an extended,
Medicine 348 (2003): 2635. Results based 19. N. Lang, “Healthcare Quality Im- more rigorous discussion and classification
on data collected between 1998 and 2000. provement: A Nursing Perspective,” and G. of these topics, see R.J. Levine, Ethics and
11. B.D. Smedley, A.Y. Stith, and A.R. Agich, “Healthcare Organization Responsi- Regulation of Clinical Research, second edi-
Nelson, eds., Unequal Treatment: Con- bility for Quality Improvement,” back- tion (New Haven, Conn.: Yale University
fronting Racial and Ethnic Disparities in ground papers prepared for The Hastings Press, 1988).
Health Care (Washington, D.C.: National Center project, “The Ethics of Improving 28. The questions in italics with each re-
Academy Press, 2003). Health Care Quality and Safety.” quirement paraphrase the questions posed
for research in Emanuel, Wendler, and

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S35
Grady, “What Makes Clinical Research Eth- “receive and review reports of local, on-site Subjects Regulations Decision Charts,” Sep-
ical?” 2701-2711. adverse events and unanticipated problems tember 24, 2004,
29. See F. Davidoff, “Publication and the involving risks to subjects or others and any ohrp/humansubjects/guidance/decision-
Ethics of Quality Improvement Projects,” other information needed to ensure that its charts.htm.
background paper prepared for The Hast- continuing review is substantive and mean- 43. The National Commission for the
ings Center project, “The Ethics of Improv- ingful. In addition, institutions and IRBs Protection of Human Subjects of Biomed-
ing Health Care Quality and Safety.” may require additional information for con- ical and Behavioral Research, Ethical Guide-
30. For an entertaining discussion related tinuing review at their discretion.” See Of- lines for the Delivery of Health Services by
to this point, see the satirical article by fice for Human Research Protections, DHEW: Report and Recommendations
G.C.S. Smith and J.P. Pell, “Parachute Use “Guidance on Continuing Review,” July 11, (Washington, D.C.: U.S. Government
to Prevent Death and Major Trauma Relat- 2002, Printing Office, 1978).
ed to Gravitational Challenge: Systematic subjects/guidance/contrev2002.htm. 44. Note, however, the survey in P.K.
Review of Randomised Controlled Trials,” 39. One might object that a review car- Lindenauer et al., “The Role of the Institu-
British Medical Journal 327 (2003): 1459- ried out by the organization doing the QI is tional Review Board in Quality Improve-
61, and responses by M. Diaz and D. not an independent review. Note, however, ment: A Survey of Quality Officers, Institu-
Neuhauser, “Pasteur & Parachutes,” Quality that the research ethics requirement is for tional Review Board Chairs, and Journal
& Safety in Health Care 14 (2005): 140-43. review by individuals unaffiliated with the Editors,” The American Journal of Medicine
31. Emanuel, Wendler, and Grady, research project, not review by individuals 113 (2002): 575-77, showing that many
“What Makes Clinical Research Ethical?” unaffiliated with the organization in which chairs of IRBs thought such an activity
2705. the research is carried out. Under current should receive IRB review.
rules, IRBs must have at least one member 45. See K. Lawlor, “OHCAs and Collab-
32. In practice, the DHHS regulations unaffiliated with the organization, but the
allow informed consent to be waived for orative Quality Improvement Projects,”
rest can be affiliated. Moreover, the name background paper prepared for The Hast-
minimal risk research provided that several “institutional review board” reflects the
other stringent requirements are met. Such ings Center project, “The Ethics of Improv-
original concept of IRBs as entities responsi- ing Health Care Quality and Safety.”
waivers are considered incompatible with ble for carrying out an institution’s obliga-
ethical principles by some ethicists, howev- tion to protect the human subjects in the re- 46. Bottrell, “Accountability for the Con-
er. search conducted under its auspices. duct of Quality Improvement Projects.”
33. See Dubler et al., “Informed Partici- Human subjects protection was deliberately 47. See American Nurses Association,
pation in the QI Process.” located at the institutional level so that re- Code of Ethics for Nurses with Interpretative
34. See discussion of “notification” in viewers would have knowledge of local con- Statements, June 30, 2001, http://www.
Levine, Ethics and Regulation of Clinical Re- ditions and the opportunity for ongoing;
search, 178-81. oversight (limited though it is). In other American Medical Association Council on
35. See K. Maschke, “Health Care Qual- words, locating primary responsibility for Ethics and Judicial Affairs, Code of Medical
ity Improvement Activities and the HIPAA the protection of human participants in QI Ethics (Chicago, Ill.: American Medical As-
Privacy Rule,” background paper prepared within the organization doing the QI is not sociation, 2004-2005).
for The Hastings Center project, “The a significant departure from independent 48. Statement by Michael Carome, page
Ethics of Improving Health Care Quality review as it currently occurs in research. 158 of transcript, Secretary’s Advisory Com-
and Safety.” 40. A. Gauthier and M. Serber, A Need to mittee on Human Research Protections
36. See J. Lynn, “When Does Quality Transform the U.S. Health Care System: Im- (SACHRP) Meeting, October 5, 2004,
Improvement Count as Research?” Quality proving Access, Quality, and Efficiency (New h t t p : / / w w w. h h s . g o v / o h r p / s a c h r p /
and Safety in Health Care 13 (2004): 67-70; York: The Commonwealth Fund, 2005). mtgings/mtg10-04/oct0405.txt.
and R.J. Levine, “Institutional Review 41. M. Bottrell, “Accountability for the 49. See Davidoff, “Publication and the
Boards: A Crisis in Confidence,” Annals of Conduct of Quality Improvement Pro- Ethics of Quality Improvement Projects,”
Internal Medicine 134 (2001): 161-63. jects,” background paper prepared for The and F. Davidoff and P. Batalden, “Toward
37. See Institute of Medicine, Responsible Hastings Center project, “The Ethics of Im- Stronger Evidence on Quality Improve-
Research: A Systems Approach to Protecting proving Health Care Quality and Safety.” ment. Draft Publication Guidelines: The
Research Participants (Washington D.C.: 42. The regulations are complicated and Beginning of a Consensus Project,” Quality
National Academy Press, 2002). difficult to summarize briefly or reduce to a & Safety in Health Care 14 (2005): 319-25.
38. The IRB is required to regularly simple diagram. For more details, see Office
monitor the research. For example, it must of Human Research Protections, “Human



ABIM Foundation, ACP-ASIM Foundation, and the Choo, V., “Thin Line Between Research and Audit,”
European Federation of Internal Medicine, “Medical Profes- Lancet 352 (1998): 337-38.
sionalism in the New Millennium,” Annals of Internal Medi-
cine 136 (2002): 243-46. Cretin, S., et al., “Should Patients in Quality-Improve-
ment Activities Have the Same Protections as Participants in
Agency for Healthcare Research and Quality, National Research Studies?” Journal of the American Medical Associa-
Healthcare Quality Report (Rockville, Md.: U.S. Department tion 284 (2000): 1786-88.
of Health and Human Services, 2003).
Davidoff, F., and P. Batalden, “Toward Stronger Evidence
Agency for Healthcare Research and Quality, National on Quality Improvement: Draft Publication Guidelines.
Healthcare Disparities Report (Rockville. Md.: U.S. Depart- The Beginning of a Consensus Project,” Quality & Safety in
ment of Health and Human Services, 2003). Health Care 14 (2005): 319-25.
Agency for Healthcare Research and Quality, National Diamond, L.H., et al., “Quality Improvement Projects:
Health Care Quality Report (Rockville, Md.: U.S. Depart- How Do We Protect Patients’ Rights?” American Journal of
ment of Health and Human Services, 2004). Medical Quality 19 (2004): 25-27.
Agency for Healthcare Research and Quality, National Diaz, M., and D. Neuhauser, “Pasteur and Parachutes:
Healthcare Disparities Report (Rockville, Md.: U.S. Depart- When Statistical Process Control Is Better Than a Random-
ment of Health and Human Services, 2004). ized Controlled Trial,” Quality & Safety in Health Care 14
(2005): 140-43.
Agency for Healthcare Research and Qualit,. National
Healthcare Quality Report (Rockville, Md.: U.S. Department Doezema, D., and M. Hauswald, “Quality Improvement
of Health and Human Services, 2005). or Research: A Distinction without a Difference?” IRB:
Ethics & Human Research 24 (2002): 9-12.
Agency for Healthcare Research and Quality, National
Healthcare Disparities Report (Rockville, Md.: U.S. Depart- Doyal, L., “Preserving Moral Quality in Research, Audit,
ment of Health and Human Services, 2005). and Quality Improvement,” Quality & Safety in Health Care
13 (2004): 11-12.
Amoroso, P.J., and J.P. Middaugh, “Research versus Pub-
lic Health Practice: When Does a Study Require IRB Re- Emanuel, E.J., D. Wendler, and C. Grady, “What Makes
view?” Preventive Medicine 36 (2003): 250-53. Clinical Research Ethical?” Journal of the American Medical
Association 283 (2000): 2701-11.
Bellin, E., and N.N. Dubler, “The Quality Improve-
ment-Research Divide and the Need for External Over- Fox, E., and J.A. Tulsky, “Recommendations for the Eth-
sight,” American Journal of Public Health 91 (2001): 1512- ical Conduct of Quality Improvement,” Journal of Clinical
17. Ethics 16 (2005): 61-71.
Bottrell, M.M., “The Ethics of Quality Improvement: Gauthier, A., and M. Serber, A Need to Transform the
Practitioners’ Perspectives. A Paper Commissioned by the U.S. Health Care System: Improving Access, Quality, and Effi-
Epistemology and Ethics of Quality Improvement Project at ciency (New York: The Commonwealth Fund, 2005).
the RAND Center to Improve Care of the Dying,” 2003, Institute of Medicine, To Err Is Human (Washington,
D.C.: National Academy Press, 2000).
Byers, J.F., and E.D. Aragon, “What Quality Improve-
ment Professionals Need to Know About IRBs,” Journal of Institute of Medicine, Crossing the Quality Chasm: A New
Healthcare Quality 25 (2003): 4-9. Health System for the 21st Century (Washington, D.C.: Na-
tional Academy Press, 2001).
Casarett, D., J.H. Karlawish, and J. Sugarman, “Deter-
mining When Quality Improvement Initiatives Should be Institute of Medicine, Envisioning the National Health
Considered Research: Proposed Criteria and Potential Im- Care Quality Report (Washington, D.C.: National Academy
plications,” Journal of the American Medical Association 283 Press, 2001).
(2000): 2275-80.

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S37
Institute of Medicine, Responsible Research: A Systems Ap- McGlynn, E.A., “Six Challenges in Measuring the Quali-
proach to Protecting Research Participants (Washington, D.C.: ty of Health Care,” Health Affairs 16 (1997): 7-21.
Institute of Medicine, 2002).
National Bioethics Advisory Commission, Ethical and
Institute of Medicine, Health Professionals Education: A Policy Issues in Research Involving Human Participants
Bridge to Quality (Washington D.C.: National Academy (Bethesda, Md.: National Bioethics Advisory Commission,
Press, 2003. 2000).
Jarvis, H., “Differentiating among Research, Evaluation National Commission for the Protection of Human Sub-
and Measures to Assure Quality,” Healthcare Management jects of Biomedical and Behavioral Research, Ethical Guide-
Forum 13 (2000): 34-36. lines for the Delivery of Health Services by DHEW: Report and
Recommendations (Washington, D.C.: U.S. Government
Kofke, W.A., and M.A. Rie, “Research Ethics and Law of Printing Office, 1978).
Healthcare System Quality Improvement: The Conflict of
Cost Containment and Quality,” Critical Care Medicine 31 National Ethics Committee of the Veterans Health Ad-
(2003): S143-52. ministration, Recommendations for the Ethical Conduct of
Quality Improvement (Washington, D.C.: National Center
Larson, E., et al., “A Survey of IRB Process in 68 U.S. for Ethics in Health Care, Veterans Health Administration,
Hospitals,” Journal of Nursing Scholarship 36 (2004): 260- 2002).
Nerenz, D.R., P.K. Stoltz, and J. Jordan, “Quality Im-
Leape, L.L., D.M. Berwick, and D.W. Bates, “What provement and the Need for IRB Review,” Quality Manage-
Practices Will Most Improve Safety? Evidence-Based Medi- ment in Health Care 12 (2003): 159-70.
cine Meets Patient Safety,” Journal of the American Medical
Association 288 (2002): 501-7. Palevsky, P.M., et al., “Improving Compliance with the
Dialysis Prescription as a Strategy to Increase the Delivered
Levine, R.J., Ethics and Regulation of Clinical Research, Dose of Hemodialysis: An ESRD Network 4 Quality Im-
second edition (New Haven, Conn.: Yale University Press, provement Project,” Advances in Renal Replacement Therapy
1988). 7 (2000): 21-30.
Levine, R.J., “Institutional Review Boards: A Crisis in President’s Commission for the Study of Ethical Prob-
Confidence,” Annals of Internal Medicine 134 (2001): 161- lems in Medicine, Securing Access to Health Care, Volume
63. One: Report (Washington, D.C.: U.S. Government Printing
Lindenauer, P.K., et al., “The Role of the Institutional Office, 1983).
Review Board in Quality Improvement: A Survey of Quality Pritchard, I.A., “Searching for ‘Research Involving
Officers, Institutional Review Board Chairs, and Journal Human Subjects’—What Is Examined? What Is Exempt?
Editors,” The American Journal of Medicine 113 (2002): What Is Exasperating?” IRB: Ethics & Human Research 23
575-79. (2001): 5-13.
Lo, B., and M. Groman, “Oversight of Quality Improve- Protection of Human Subjects, Code of Federal Regula-
ment: Focusing on Benefits and Risks,” Archives of Internal tions, 1991, 45 Code of Federal Regulations Part 46: Re-
Medicine 163 (2003):1481-86. vised.
Lohr, K.N., Medicare: A Strategy for Quality Assurance Reinhardt, A.C., and L.N. Ray, “Differentiating Quality
(Washington D.C.: National Academy Press, 1994). Improvement from Research,” Applied Nursing Research 16
Lynn, J., “When Does Quality Improvement Count as (2003): 2-8.
Research? Human Subject Protection and Theories of Sandrick, K., “Ethics: A Unifying Force. The Tavistock
Knowledge,” Quality & Safety in Health Care 13 (2004): 67- Principles Provide a Decisionmaking Framework for Every-
70. one,” Trustee 54 (2001): 24-28.
McGlynn, E.A., et al., “The Quality of Health Care De- Schoen, C., et al., “Taking the Pulse of Health Care Sys-
livered to Adults in the United States,” New England Journal tems: Experiences of Patients with Health Problems in Six
of Medicine 348 (2003): 6-26. Countries,” Health Affairs Web Exclusives 5 (2005): W509-
McGlynn, E.A., et al., “The Quality of Health Care De- 525.
livered to Adults in the United States,” New England Journal Schuster, M.A., E.A. McGlynn, and R.H. Brook, “How
of Medicine 348 (2003): 2635-45. Good Is the Quality of Health Care in the United States?”
Milbank Quarterly 76 (1998): 509, 517-63.


Sharpe, V.A., “Promoting Patient Safety. An Ethical Basis lenge: Systematic Review of Randomised Controlled Trials,”
for Policy Deliberation,” Hastings Center Report 33 (2003): British Medical Journal 327 (2003):1459-61.
Stevens, D.P., “Why New Guidelines for Reporting Im-
Smedley, B.D., et al., Unequal Treatment: Confronting provement Research? And Why Now?” Quality & Safety in
Racial and Ethnic Disparities in Health Care (Washington, Health Care 14 (2005): 314.
D.C.: National Academy Press, 2003).
Stimler, C., “Quality Improvement Projects: Inaction
Smith, G.C., and J.P. Pell, “Parachute Use to Prevent Presents the Greatest Risk,” Archives of Internal Medicine
Death and Major Trauma Related to Gravitational Chal- 163 (2003): 2648-49.


This project had its origins in an exploratory confer- Two “field trips” taken by the project’s principal in-
ence initiated by Joanne Lynn and Melissa Bottrell. With vestigator, Mary Ann Baily, helped to deepen our under-
Bruce Jennings, they developed a proposal to commis- standing of the issues. One was a trip to Dartmouth-
sion four background papers, convene an interdiscipli- Hitchcock Medical Center, made at the invitation of Paul
nary group of experts, and use the papers as the basis Batalden, who thought it would be a good way to learn
for group discussion of the ethical and policy issues more about the range of QI projects. We would like to
raised by the use of quality improvement methods in thank Polly Campion for organizing a day packed full of
health care. The Agency for Healthcare Research and fascinating give-and-take with people engaged in QI;
Quality (AHRQ) funded the proposal, and the conference there were too many to name individually, but we are
took place in December 2001 in conjunction with the In- grateful to all of them. The second was a trip to M.D. An-
stitute for Healthcare Improvement (IHI) Annual Quality derson Cancer Center, where Mary Ann Baily made pre-
Forum. Intense discussions over two days, in breakout sentations about the project to several different groups.
sessions and plenary sessions, clarified the issues but The trip was a wonderful opportunity to engage in in-
made it clear that further work was needed. This gave tense discussion with people about the issues they faced
rise to The Ethics of Improving Health Care Quality and doing QI activities in the context of specialized cancer
Safety Project, which was also funded by AHRQ and by treatment. We are grateful to all the participants and
in-kind contributions by The Hastings Center. would especially like to thank Mano Selvan and Mar-
We deeply appreciate the hard work of the project garet Holm for organizing the visit.
participants (listed by name at the front of this report) We are grateful to AHRQ for providing funds for the
and the authors of the background papers (referred to in project, and we appreciate the assistance of our project
footnotes to this report). The papers, the meeting dis- officers: Elinor Walker, Marge Keyes, and finally, Robert
cussions, the comments on drafts of the report, and the Borotkanics. We thank the RAND Corporation for provid-
many one-on-one conversations we had with partici- ing us with meeting space for our final project meeting,
pants and authors were critical to the development of and the RAND staff for administrative support for the
this document. We would also like to acknowledge the meeting. Finally, we appreciate the efforts of the various
contributions of several people not listed who shared members of The Hastings Center staff who have provid-
their expertise at individual project meetings: Andrea ed support throughout the project. We thank Mary Ann
Kabcenell, Jenny Kitsen, Charles Bosk, and Jason Kar- Hasbrouck, Ann Mellor, Jay Camp, and Vicki Peyton for
lawish. In addition, we had many stimulating discussions their work on meeting arrangements and project admin-
of the issues with others interested in the project, in- istration, Michael Khair and Stacy Sanders for research
cluding Louis Diamond, Paul Palevsky, Michael Rie, and assistance, and Greg Kaebnick, Joyce Griffin, and Nora
Andrew Kofke. Porter for editorial and design assistance for this Special

SPECIAL REPORT / The Ethics of Using QI Methods to Improve Health Care Quality and Safety S39

he Hastings Center addresses fundamental ethical is-

T sues in the areas of health, medicine, and the
environment as they affect individuals, communities, and
Mary Ann Baily, PhD, is associate for ethics and health
policy at The Hastings Center. She is an economist by
training and has written on ethical issues in health care
societies. With a small staff of senior researchers at the rationing, access to care, managed care, newborn genet-
Center and drawing upon an internationally renowned ic screening, and the financing of care for HIV infec-
group of over one hundred elected Fellows for their ex- tion, organ transplantation, and Alzheimer disease.
pertise, The Hastings Center pursues interdisciplinary re-
search and education that includes both theory and prac- Melissa Bottrell, MPH, PhD, works on practical prob-
tice. Founded in 1969 by philosopher Daniel Callahan lems at the intersection of bioethics, health system qual-
and psychoanalyst Willard Gaylin, The Hastings Center ity, and policy. Currently, she is a health services re-
is the oldest independent, nonpartisan, interdisciplinary searcher at the Veteran’s Health Administration Nation-
research institute of its kind in the world. From its earli- al Center for Ethics in Health Care. She has conducted
est days The Hastings Center has understood that the projects on informed consent, facility-level ethics pro-
moral problems arising from rapid advances in medicine grams, end of life care, and quality of geriatric nursing
and biology are set within a broad intellectual and social care in health care facilities such as hospitals and nurs-
context. The Center’s collaborations with policy-makers, ing homes. She has authored numerous publications,
in the private as well as the public sphere, assist them in including three edited books.
analyzing the ethical dimensions of their work.
Bruce Jennings, MA, is director of the Center for Hu-
mans and Nature and senior consultant at The Hastings
Center. He also teaches at the Yale University School of
Public Health. He has authored numerous books and
For copies of this or other Hastings Center Report articles on ethical and social issues in health care and
Special Reports, write or call: public policy.
Subscription Department
The Hastings Center Joanne Lynn, MD, MA, MS, is a geriatrician working
21 Malcolm Gordon Road as Senior Natural Scientist at the RAND Corporation.
Garrison, NY 10524 She has contributed to bioethics literature, including as
(845) 424-4040 staff to the President’s Commission on Ethical Issues in
(845) 424-4545 fax; Medicine; she has led conventional and health services research; and she has now coordinated the work of more than four hundred teams doing quality improvement.

On the cover: Arrow Up & Down, by Alexander Calder, 1972, gouache on paper, 74.9 cm x 110 cm.
© Artists Rights Society, New York. Photo courtesy Art Gallery of Ontario.