The n e w e ng l a n d j o u r na l of m e dic i n e

s ounding boa r d

Medical Education
Malcolm Cox, M.D., and David M. Irby, Ph.D., Editors

“Continuity” as an Organizing Principle

for Clinical Education Reform
David A. Hirsh, M.D., Barbara Ogur, M.D., George E. Thibault, M.D., and Malcolm Cox, M.D.

If the ultimate purpose of medical education — cal education, one that would incorporate the
to meet the health needs of society — is to be strengths of the present acute care educational
achieved, the primary goal of undergraduate med- model but eliminate the model’s major weakness
ical education should be to produce students who — a lack of connection or continuity among dif-
are broadly skilled in the core competencies that ferent learning experiences.8,9
transcend all disciplines of medicine.1,2 The chal-
lenge is how to accomplish this goal in a clinical ed uc ational continuit y
learning environment fragmented by increasing
specialization and demands for clinical produc- Rooted in the principles of modern learning
tivity and constrained by a prevailing culture in theory,10,11 the notion of educational continuity
which education must compete with research and reflects the progressive professional and personal
clinical practice for medical school resources.3 development required of physicians in training.12
As compared with the dramatic changes that A spirit of “ownership” of the entire curriculum,
have occurred in biomedical science and the prac­ rather than one discipline-specific portion of the
tice of medicine, the fundamental model of clini- curriculum, is a prerequisite for educational con-
cal education in American medical schools has tinuity.13 As applied to the core clerkship year,
changed little since the time of Sir William Osler, educational continuity subsumes two interrelated
a century ago. Students are still assigned to spe- integrating forces: horizontal integration (enhanc-
cialty-specific teams of interns, residents, and ing the development of general competency by
supervising faculty physicians for relatively brief, linking learning experiences between and across
randomly sequenced rotations in acute care hos- clinical specialties) and vertical integration (en-
pitals.4,5 And the core clinical credentialing expe- hancing evidence-based practice by linking ad-
rience continues to be this same series of rota- vances in the biomedical and clinical sciences to
tions, primarily in the third year of the traditional clinical problem solving).
four-year undergraduate curriculum, just as it was Continuity of the learning environment fosters
in Osler’s day. both patient-centeredness and learner-centered-
Although there is no doubt that the hospital ness by establishing more opportunities for con-
environment remains rich in learning opportuni- nections with patients (“continuity of care”); by
ties for medical students and that students need integrating important educational themes across
to learn the skills necessary to succeed in an clinical specialties, focusing on the developmen-
environment in which most of them will spend tally appropriate attainment and assessment of
3 to 8 years of postgraduate training, there is a core clinical competencies, and promoting the
growing sense nationally that the current model connection between science and clinical medicine
is poorly aligned with society’s present and future (“continuity of curriculum”); and by enhancing
health care needs.6,7 This realization has led supervision, role modeling, and mentoring (“con-
many observers to call for a new model of clini- tinuity of supervision”) (Table 1).

858 n engl j med 356;8  www.nejm.org  february 22, 2007

The New England Journal of Medicine

Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 sounding board

Continuity of Care

Interdisciplinary curriculum design and Promotion of core doctoring skills, including

Learner-centered education and assessment

including chronic-disease management

communication and clinical reasoning

Throughout the history of the profession, the

Promotion of a full range of clinical skills,

Promotion of medical colleagueship and

“Horizontal” and “vertical” curriculum Enhanced evidence-based practice and
most powerful motivator for learning has been

Interprofessional understanding and

the sense of deep commitment to patients. Con-

Enhanced pedagogy and learning

necting the student’s desire to serve with his or

Patient-centered health care

her desire to learn has strong support in learning

interdisciplinary values
Enhanced professionalism
theory and has been used effectively for many

Expected Outcomes
years in a wide variety of service learning pro-

lifelong learning

­collaboration
grams in health-related disciplines. However, only
rarely has service learning been part of the core
clerkship experience itself.14
In order to anchor clinical learning in care-
giving, students must have relevant involvement
with patients at the site and time of initial med-

Interdisciplinary summative assessment

Protected time for teaching and faculty

time of initial medical decision making Collaborative, interdisciplinary delivery
Involvement with patients at the site and Longitudinal patient care experiences
ical decision making, ideally before the diagno-

Longitudinal student oversight and

Ability to identify, track, and follow

Continuous formative assessment

sis is made, and be able to follow patients for

­patients across care venues

the duration of an illness episode (and beyond),

Operational Requirements
ideally across care venues. The critical thinking
involved in making a diagnosis compels students
to value history taking, physical examination, ra-

management

development
and grading

­assessment
tional diagnostic testing, and differential diag-

integration
nostic reasoning. By the same token, students
of care
should follow patients long enough to observe
the course of the illness and the patient’s expe-
rience of the illness, and they should witness the

Application of biomedical science to clinical

Developmentally appropriate, competency-

effects of their management decisions. Continu- Acquisition of relevant competencies in a

caregivers engaged in a transparent

Community of learners, educators, and
and during the full course of illness

ity of care also provides opportunities for teach- structured, developmental fashion
Custom-designed patient enrollment

Faculty coaching, role modeling, and

­dialogue about patient care and
ers to custom-design patient enrollment to meet
overarching educational goals and fine-tune co-
horts of patients as the learning experience un-
folds over time.
based assessment
problem solving
Specific Objectives

medical science

­mentorship
Continuity of Curriculum
To support the progression of a learner’s values,
Table 1. Goals and Expected Outcomes of Educational Continuity.

attitudes, knowledge, and skills, each component

of a curriculum should have a rational, considered
relationship with all others. A developmentally
Learning through patient connection,

progressive curriculum has three major aspects.

promote foundational knowledge
Learning in an integrated fashion to

Learning from close and serial con-

First, there is a rational sequential order that facili-

nection with the most able

tates learning, with certain types of knowledge

and skills serving as the foundation for subse-
caring, and advocacy

quent learning. Skills that are notably different,

and clinical skills

but equally complex, may still be most appropri-

ately taught in a particular order. For example,
educators

knowledge of anatomy and pathophysiology facili-

tates the taking of a medical history. Second, more
Goals

complex tasks should follow some degree of

achievement in the performance of more simple
but related tasks. Thus, one learns to construct a
Supervision
Curriculum
Continuity

simple problem list before learning to develop a

complicated differential diagnosis. Similarly, grap-
Care

pling with complex therapeutic decisions makes

n engl j med 356;8  www.nejm.org  february 22, 2007 859

The New England Journal of Medicine
Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

little sense for a student who has yet to master in which faculty members have personal respon-
rudimentary diagnostic decision making. Third, sibility for overseeing their own students, pro-
the curriculum should be implemented in a learn- vide students with the emotional comfort to take
er-centered manner, such that a given student’s intellectual risks in their learning. At the same
learning is tailored to his or her particular evolv- time, trusting relationships and shared goals fos-
ing (i.e., developmentally appropriate) needs. ter coaching, promote effective feedback, and en-
Optimally, the core clinical clerkship curricu- hance clinical performance.
lum should be designed to emphasize themes At a minimum, clerkship directors and clini-
central to doctoring (professionalism, commu- cal teachers should collaborate across disciplines
nication, and teamwork) and continuing scientif­ to design, implement, and oversee the entire clerk-
ic literacy (evidence-based decision making). It ship year and should have joint responsibility
should ensure the exposure of students to a pre- with inpatient attending physicians and continu-
defined set of clinical syndromes and diseases, ity preceptors for student supervision, mentoring,
thereby promoting both context-specific clinical and assessment. Because of their particular effec-
reasoning and the acquisition of cross-disciplin- tiveness in teaching the fundamentals of clinical
ary competencies. It should consolidate and ex- reasoning and the psychosocial aspects of care,
pand fundamental insights into the mechanisms experienced clinician–educators (“master clini-
of disease in individual patients and populations. cians”) rather than inexperienced faculty or resi-
And it should assess particular knowledge, skills, dents should have the most prominent education-
and attitudes at times most suited to ascertain al and supervisory roles.17,18
competence and ensure developmental progres-
sion. Diverse and repeated formative assessments b arrier s to ed uc ational
of student performance are also important. Espe- continuit y
cially when embedded in integrated educational
programs, multimethod assessment enhances Although the concept of educational continuity
both the evaluation process itself and the students’ provides a powerful organizing principle for clin-
learning trajectory.15 ical education reform, the barriers to changing
All these requirements are greatly facilitated the manner in which the traditional core clinical
by collaborative, interdisciplinary ownership of clerkship experience is organized should not be
the clinical curriculum. Collaboration across clin- underestimated (Table 2). Promoting innovation
ical departments ensures a horizontally integrated on the basis of either educational theory or nas-
curriculum, with an emphasis on core competen- cent outcomes data alone will require visionary
cy development. Likewise, collaboration between leadership, innovative resource management, and
clinical and basic science departments (vertical careful attention to learning, cultural, and regula-
integration) ensures that the core clinical clerk- tory issues.
ship experience is built on and deliberately con- Relatively few academic medical centers can
nected to the basic biomedical and epidemiologic easily deliver meaningful experiences in continu-
sciences, thereby promoting continuing scientific ity of care. In most such centers, investment in
literacy. In sum, interdisciplinary governance pro- ambulatory care facilities is insufficient to match
vides a platform for promoting general compe- the care needs of patients, let alone support learn-
tencies; exploring the pertinence of basic, transla- ing. And although investments in information
tional, and clinical science to medical practice; technology are increasing, with few exceptions,
and incorporating all manner of biomedical scien- patient information systems and mechanisms to
tists into the clinical learning environment itself. identify, track, and follow patients across sites of
care remain rudimentary. For the most part, care
Continuity of Supervision is delivered in a discipline-specific fashion, and
Establishing connections between faculty, other although all academic medical centers aspire to
caregivers, and students and among faculty across be truly patient-centric organizations, few of them
disciplines is critical to forming a productive actually meet this goal.
learning community.16 Students and longitudinal Achieving the full potential of integration will
preceptors share the professional intimacy of dual not be easy. Departmental boundaries are notori-
responsibility for patient care. Such relationships, ously difficult to breach, and coordinating teach-

860 n engl j med 356;8  www.nejm.org  february 22, 2007

The New England Journal of Medicine

Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 sounding board

ing time across clerkships and between basic

Table 2. Potential Barriers to Educational Continuity.
science and clinical departments will be challeng-
ing. Reaching agreement on learning objectives, Continuity Barriers
curriculum content and delivery methods, and Care Underinvestment in ambulatory care infrastructure
common assessment and grading systems will Underinvestment in information technology infrastructure
be possible only with clearly delineated objectives Traditional academic medical center organization and culture
and willingness to compromise. At first sight, Curriculum Departmental boundaries and culture
Lack of agreement on educational and assessment strategies
planning, start-up, and ongoing administrative Administrative costs
costs — including student and faculty schedul- Inflexible accreditation and other regulatory standards
ing across departments and departmental teach- Supervision Incremental faculty teaching effort
ing effort and funds-flow analyses — may appear Lack of recognition and academic advancement
to be prohibitive, especially in the absence of Narrowness of faculty expertise
Lack of interdisciplinary teaching models
mission-based budgeting. New curricular models Insufficient competency-based evaluation instruments
must take into account already established local,
state, and national standards, and regulatory re-
lief will have to be obtained from the appropri- proach to curriculum design and management
ate accrediting and licensing bodies. and to better deal with so-called orphan topics,
The barriers to effective supervision and men- medical schools have begun to assume more cen-
toring are diverse. In the absence of mission- tralized control of the clerkship year. Over the
based budgeting, the cost of incremental faculty past decade, this shift in governance has allowed
teaching and supervision — including both new for the development of a variety of new models
responsibilities and some of those currently as- of clinical clerkships, many of which have incor-
sumed by residents — may appear to be prohibi- porated elements of educational continuity into
tive. Teaching time and faculty availability are the overall learning experience.
limited by demands for increased clinical and re- Some schools have developed interdisciplinary
search productivity. Regardless of cost and avail- “intersessions” or “interclerkships” (courses, gen-
ability, freeing faculty for more extensive educa- erally of about a week’s duration, interposed be-
tional responsibilities will be difficult as long as tween sequential clerkships) (Fig. 1B)19 and longi-
educational effort and excellence continue to be tudinal “themes” or “threads” (courses that link
undervalued in academic advancement. Perhaps similar content between clerkships) (Fig. 1C).20
most important, not all clinicians have the requi- Both models provide opportunities for interdisci-
site background or skills to incorporate innova- plinary curriculum design and management. How-
tions in basic science or evidence-based practice ever, short of major curricular revisions (such as
into their teaching, and not all basic scientists consolidating core clerkship objectives or extend-
are comfortable with participatory teaching in a ing the duration of the overall experience), time
clinical environment. Finally, the relative lack of limitations curtail the ability of either approach
well-validated interdisciplinary teaching models to reach its full potential.
and competency-based evaluation instruments Many of these new clerkship experiences have
may have a negative effect on teaching and as- used small-group, problem-based learning, which
sessment. although a natural locus for interdisciplinary
teaching,21 had not previously been used in the
ne w model s of clinic al clinical curriculum as commonly as in the pre-
clerk ship s clinical curriculum.22 In England, at the Univer-
sity of Manchester, modules of thematically orga-
Promoting educational continuity is complicated nized, problem-based material are now being
by the traditional division of the core clinical taught alongside traditional discipline-specific
clerkship experience into a disconnected series of “attachments” (clerkships).23 Semistructured in-
independently governed, discipline-specific, ran- terviews of Manchester graduates have indicated
domly ordered, sequential blocks (Fig. 1A), each significant gains in dealing with clinical uncer-
characterized by largely ad hoc patient assign- tainty, knowing their personal limits, and asking
ments and poorly coordinated learning objectives. for help when these limits are exceeded.24
To provide opportunities for a more collective ap- A variant of problem-based learning has been

n engl j med 356;8  www.nejm.org  february 22, 2007 861

The New England Journal of Medicine
Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Figure 1. Clerkship Organization.

A Sequential Each panel illustrates a different organizational mod-
Discipline-specific
el, with specialties represented by different colors.
For simplicity, only four of the six traditional core
clerkship specialties (internal medicine, obstetrics
B Sequential and gynecology, neurology, pediatrics, psychiatry,
Intersessions and surgery) are shown. Surgery generally includes
formal exposure to anesthesiology. Many schools
now include family and community medicine in the
core clerkship experience as well; a smaller number
C Sequential also include emergency medicine and radiology. The
Longitudinal themes
models are named for their dominant organizational
characteristic (sequential, longitudinal, or mixed). In-
terdisciplinary curricular governance and longitudinal
D Sequential care experiences greatly enhance educational continu-
Limited integration ity. Panel A represents the traditional clerkship year,
a randomly sequenced series of discipline-specific,
inpatient block rotations. This ­organization is not
conducive to continuity of care or supervision and al-
E Sequential lows for only limited curricular con­tinuity. “Interses-
Block ambulatory
sion” courses (gray bars) between two or more disci-
pline-specific blocks provide the opportunity for
interdisciplinary teaching of selected core competen-
F Sequential cies or other material (Panel B). Core material can
Longitudinal ambulatory also be presented as longitudinal “themes” or
“threads” bridging two or more discipline-specific
clerkships (Panel C). The sequential model in Panel
D, in which two related discipline-specific blocks are
G Sequential combined, ­offers opportunities for interdisciplinary
Recurring ambulatory
patient care experiences in the particular disciplines
involved. Panels E and F represent relatively common
variants of the traditional clerkship model: the addi-
H Mixed tion of an ambulatory care experience to the still
dominant sequential, inpatient model. Ambulatory
care experiences can be structured as discrete blocks
(Panel E, darker shades in the first and third blocks)
or in a longitudinal fashion (Panel F, purple block)
I Longitudinal
and may be organized by one or more disciplines.
Integrated
Panel G illustrates an as-yet-untried model for retain-
ing discipline-specific immersion ex­periences and con-
tinuity by alternating inpatient ex­periences with recur-
ring outpatient rotations (purple block); this may be a
way of introducing experiences in continuity of care
used at the University of Dundee in Scotland to
without resorting to a strictly longitudinal curriculum
integrate content across the entire curriculum.25 structure. Panel H is one of many potential mixed
Task-based learning uses the clinical experience models in which discipline-specific sequential clerk-
itself, rather than “paper” cases, to generate ex- ships are retained (with their time allotment reduced),
amples
AUTHOR: ofHirsh/Cox
a series of predetermined
RETAKE 1sttasks, with and educational continuity is provided by longitudinal
ICM
ambulatory care experiences, longitudinal mentoring
REG F the students
FIGURE: 1 of 1 themselves responsible2nd for finding
3rd and assessment, and a longitudinal interdisciplinary
CASE opportunities to explore theseRevised tasks as they move curriculum. In Panel I, the clerkships are organized in
Line
EMail through a discipline-specific, 4-C
ARTIST: tv sequential
SIZE curricu- a parallel rather than sequential fashion. For example,
H/T H/T 22p3
Enon lum. Task-based learning is credited with enhanc- each week of the clerkship year might contain experi-
Combo
ing the AUTHOR,
transferPLEASE
of basicNOTE:science knowledge to the ences in all (or most) of the traditional disciplines. In
Figure has been redrawn and type has been reset. this model, students follow patients longitudinally
clinical years as well as providing
Please check carefully.
an opportunity
across some or all care venues (including across disci-
for integration of core content across clinical dis- plines), and the members of the faculty assume collec-
ciplines without the need to
JOB: 35608 create
ISSUE: interdisciplin-
02-22-07 tive owner­ship of the entire clerkship experience.
ary teaching teams.26

862 n engl j med 356;8  www.nejm.org  february 22, 2007

The New England Journal of Medicine

Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 sounding board

Where curricular content sufficiently overlaps ample, time was equally divided between tradi-
disciplines (neurology and psychiatry or obstetrics tional, discipline-specific inpatient rotations and
and neonatology, for example), the opportunity ambulatory settings, with ambulatory training
exists to integrate clerkships more fully across being provided in specialty clinics and a year-
disciplinary lines (Fig. 1D). However, although long continuity experience in one of the general-
there are examples of integration in which relat- ist disciplines.37 Weekly tutorials and seminars,
ed clerkships have been grouped together for ad- organized as longitudinal themes and provided
ministrative or scheduling purposes, multidisci- by a constant group of faculty mentors, served to
plinary governance and joint teaching have been bridge individual specialty-specific experiences.
attempted only infrequently and have proved dif- Grade distributions in core clerkships were sim-
ficult to sustain in a subspecialty-dominant prac- ilar, except in psychiatry, in which students in the
tice environment.27-29 integrated track achieved significantly higher
Ambulatory care clerkships are another poten- scores than did students in the traditional cur-
tial locus for interdisciplinary design and man- riculum. They also performed better on a gener-
agement. In recent years, block or longitudinal alist Objective Structured Clinical Examination
ambulatory care clerkships (Fig. 1E and 1F, re- but had a lower mean score on the National Board
spectively) — individually or collectively organized of Medical Examiners’ internal-medicine “shelf”
by departments of family medicine, general inter- exam. A majority of students reported that they
nal medicine, and general pediatrics — have be- would choose the integrated third year again and
come relatively common components of the core would recommend it to others. Similar approach-
clerkship year.30-34 Although single or isolated es are being tested at several teaching hospitals
block experiences are an appropriate forum for associated with Harvard Medical School and the
the follow-up of time-limited disorders, they pro- University of California, San Francisco, as part of
vide little opportunity for exposure to chronic school-wide efforts on medical education reform.
disease management, a major required compe- Longitudinal organization of most or all spe-
tency in the modern practice environment. cialties that are commonly represented in the core
Students appear to benefit from longitudinal clerkships (Fig. 1I) is an emerging but still un-
ambulatory care experiences by developing more common model. Motivated by the need for grad-
effective relationships with patients, gaining in- uates who are interested in practicing in medical­
sight into the psychosocial aspects of care, and ly underserved areas, some schools have created
understanding the longitudinal management of clerkships that place students in longitudinal am-
chronic illness.35 However, this potential is often bulatory care experiences — including primary
degraded by competing inpatient responsibilities care and multidisciplinary group practices — for
and patient-scheduling problems. Recurring am- a significant portion of their clinical training.
bulatory-block rotations devoted exclusively to When measured against regional workforce goals,
generalist community practice, alternating with these programs have been judged to be quite suc-
discipline-specific inpatient blocks (Fig. 1G), cessful.38-43 Students in these variously integrated
might provide an effective solution. Combining longitudinal clerkships have performed as well
both departmentally based and interdisciplinary as their more traditionally trained counterparts
governance models, this intriguing approach — on local and national examinations of clinical
recently suggested to promote continuity in in- competence.44-46
ternal medicine residency education36 — has yet Other schools are testing the feasibility of
to be tested. Applied to undergraduate education, multidisciplinary, cross-site longitudinal integra-
however, it would probably require substantial tion without emphasizing primary care or at-
lengthening of the traditional clerkship year. tempting to steer students toward the generalist
Many permutations of these basic models are disciplines. In a pilot project at Harvard Medical
possible. Any substantial combination of sequen- School and the Cambridge Health Alliance in
tial and longitudinal experiences — so-called Cambridge, Massachusetts, students spend the
mixed models — would allow for some degree of entire third year learning from serial contact with
both discipline-specific immersion and education­ a carefully selected cohort of patients recruited
al continuity (Fig. 1H). In a pilot program at Case from their preceptors’ practices in internal medi-
Western Reserve University in Cleveland, for ex- cine, pediatrics, psychiatry, neurology, and ob-

n engl j med 356;8  www.nejm.org  february 22, 2007 863

The New England Journal of Medicine
Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

stetrics and gynecology.47 Each patient is followed a health care system second to none.53 Medical
across all venues of care, including outpatient education reform is one important means to
specialty and subspecialty clinics, the inpatient this end.
setting, and rehabilitative, nursing home, and Although considerable heterogeneity of clini-
home care. Special arrangements facilitate expo- cal education is ultimately likely, and even desir-
sure to patients in the emergency department able, the essential features of a new paradigm for
and a full spectrum of general surgical care. the 21st century must include a substantive re-
Weekly case-based tutorials on fundamental top- thinking of the relationships among patients,
ics that seek to integrate basic and clinical sci- students, and teachers and most especially the
ence, simulation exercises, electronic records, and environment in which this relationship either
mentored educational portfolios further empha- prospers or falters. An emphasis on continuity
size the interdisciplinary and personalized nature of care, curriculum, and supervision provides a
of the curriculum.48,49 solid foundation for maintaining and enhancing
Outcomes data have been encouraging, al- an even more fundamental continuity: the conti-
though the population of participating students nuity of idealism. Students enter medical school
is as yet small.50 Students see patients far more highly idealistic, with core values of altruism,
frequently before a diagnosis is made and after empathy, humanism, and service. However, de-
discharge from the hospital and are supervised spite being cornerstones of professionalism, val-
by experienced faculty, rather than residents, to a ues such as excellence in communication, cultur-
much greater extent. In tests of knowledge and al competence, and attention to social justice,
clinical skills, these students perform as well as actually erode during training.54,55
or better than their more traditionally trained Attention to educational continuity has the
counterparts. At the same time, clinical precep- potential to forestall such erosion. Continuity of
tors and tutorial facilitators are enthusiastic about care provides students with relevant, extended,
teaching, some for the first time after many years and serial contact with patients, physician precep-
of frustration. tors, and other health care professionals. The
goals of students and patients are aligned, and
conclusions students become natural advocates for their pa-
tients’ interests and needs. Continuity of curric-
Only time will tell whether any of these new ulum creates space for self-reflective practice,
clerkship models will have enduring value or conceptual integration, and critical thinking,
whether yet others will need to emerge. Whatever without which learning becomes task-based and
the model, the clinical environment must be made heuristic. As students wrestle with complex pro-
more receptive to professional development, and fessional issues, a cohesive curriculum provides
learning must be embedded in caring for patients. both a conceptual framework and a practical fo-
Just as patient-centeredness and improvements in rum for explicit learning and development. Mean-
health care quality are becoming the overarching ingful clinical experiences and continuity of su-
metrics of the health care delivery system, so too pervision support students’ ability to know all
should learner-centeredness and improvements they can about their patients and their conditions,
in educational quality become the proximate met- from the basic science underlying the pathophys-
rics of the medical education system. iology to the family and community in which the
The concept of educational continuity — driv- patient lives. Continuity of supervision also pro-
en by collaborative, interdisciplinary governance vides the luxury of intergenerational, iterative
— provides a sufficiently broad framework to dialogue grounded in practice about values, pro-
accommodate the development and evaluation of fessionalism, and lifelong learning. In this way,
a wide variety of new models of clinical education. the entire learning community nurtures and
Any model of clinical education that emphasizes maintains a spirit of idealism — idealism that
the complex cross-disciplinary skills of doctoring will surely be translated into enhanced learning,
rather than preparing students solely for disci- greater patient satisfaction, and more efficient
pline-specific inpatient practice will present sub- and effective medical care.
stantial financial, organizational, and cultural No potential conflict of interest relevant to this article was re-
difficulties,51,52 but the American public deserves ported.

864 n engl j med 356;8  www.nejm.org  february 22, 2007

The New England Journal of Medicine

Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 sounding board

From Cambridge Hospital, Cambridge, MA (D.A.H., B.O.); 26. Davis MH, Harden RM. Planning and implementing an under­
Harvard Medical School Boston (D.A.H., B.O., G.E.T., M.C.); and graduate medical curriculum: the lessons learned. Med Teach
the Veterans Health Administration, Washington, DC (M.C.). 2003;25:596-608.
27. Speer AJ, Stagnaro-Green A, Elnicki DM. Interdisciplinary
1. Tomorrow’s doctors. London: General Medical Council, clerkships: educational models of the future? Am J Med 1995;
2003. 99:451-3.
2. Educating doctors to provide high quality medical care: 28. Roberts LW, Franchini G, Fielder K. An integrated psychiatry-
a vision for medical education in the United States. Washington, neurology clerkship within a problem-based learning curriculum.
DC: Association of American Medical Colleges, July 2004. Acad Med 1997;72:423-4.
3. Watson RT. Rediscovering the medical school. Acad Med 29. The education of medical students: ten stories of curriculum
2003;78:659-65. change. New York: Milbank Memorial Fund, Association of
4. Christakis NA. The similarity and frequency of proposals to American Medical Colleges, 2000:92-114.
reform US medical education: constant concerns. JAMA 1995; 30. Irby DM. Teaching and learning in ambulatory care setting:
274:706-11. a thematic review of the literature. Acad Med 1995;70:898-931.
5. Whitcomb ME. Redesigning clinical education: a major chal- 31. Hunt CE, Kallenberg GA, Whitcomb ME. Trends in clinical
lenge for academic health centers. Acad Med 2005;80:615-6. education of medical students: implications for pediatrics. Arch
6. Ludmerer KM. Time to heal: American medical education Pediatr Adolesc Med 1999;153:297-302.
from the turn of the century to the era of managed care. New 32. Peters AS, Feins A, Rubin R, Seward S, Schnaidt K, Fletcher
York: Oxford University Press, 1999. RH. The longitudinal primary care clerkship at Harvard Medical
7. Greiner AC, Knebel E, eds. Health professions education: School. Acad Med 2001;76:484-8.
a bridge to quality. Washington, DC: National Academies Press, 33. Carney PA, Pipas CF, Eliassen MS, et al. An analysis of stu-
2003. dents’ clinical experiences in an integrated primary care clerk-
8. Christakis DA, Feudtner C. Temporary matters: the ethical ship. Acad Med 2002;77:681-7.
consequences of transient social relationships in medical training. 34. Pipas CF, Peltier DA, Fall LH, et al. Collaborating to inte-
JAMA 1997;278:739-43. grate curriculum in primary care medical education: successes
9. Glick TH, Moore GT. Time to learn: the outlook for renewal and challenges from three US medical schools. Fam Med 2004;
of patient-centred education in the digital age. Med Educ 2001; 36:Suppl:S126-S132.
35:505-9. 35. Ogrinc G, Mutha S, Irby DM. Evidence for longitudinal ambu-
10. Bransford JD, Brown AL, Cocking RR, Donovan MS, Pellegri- latory care rotations: a review of the literature. Acad Med 2002;
no JW, eds. How people learn: brain, mind, experience, and school. 77:688-93.
Expanded ed. Washington, DC: National Academy Press, 2000. 36. Weinberger SE, Smith LG, Collier VU. Redesigning training
11. Norman G. Research in clinical reasoning: past history and for internal medicine. Ann Intern Med 2006;144:927-33.
current trends. Med Educ 2005;39:418-27. 37. Lewin LO, Papp KK, Hodder SL, et al. Performance of third-
12. Armstrong E, Parsa-Parsi R. How can physician’s learning year primary-care-track students in an integrated curriculum at
styles drive educational planning? Acad Med 2005;80:680-4. Case Western Reserve University. Acad Med 1999;74:Suppl:S82-
13. Armstrong EG, Mackey M, Spear SJ. Medical education as a S89.
process management problem. Acad Med 2004;79:721-8. 38. Verby JE, Newell JP, Andresen SA, Swentko WM. Changing
14. Elam CL, Sauer MJ, Stratton TD, Skelton J, Crocker D, Musick the medical school curriculum to improve patient access to pri-
DW. Service learning in the medical curriculum: developing and mary care. JAMA 1991;266:110-3.
evaluating an elective experience. Teach Learn Med 2003;15:194- 39. Hansen LA, Talley RC. South Dakota’s third-year program of
203. integrated clerkships in ambulatory-care settings. Acad Med
15. van der Vleuten CPM, Schuwirth LWT. Assessing professional 1992;67:817-9.
competence: from methods to programmes. Med Educ 2005;39: 40. Worley P, Silagy C, Prideaux, Newble D, Jones A. The parallel
309-13. rural community curriculum: an integrated clinical curriculum
16. Kenny NP, Mann KV, MacLeod H. Role modeling in physi- based in rural general practice. Med Educ 2000;34:558-65.
cians’ professional formation: reconsidering an essential but un- 41. Anderson AS, Martell JV. Comparing sequential clerkships
tapped educational strategy. Acad Med 2003;78:1203-10. and a longitudinal clerkship for third-year medical students.
17. Fishbein RH. Professionalism and the ‘master clinician’ — Acad Med 1994;69:418-9.
an early learning experience. J Eval Clin Pract 2000;6:241-3. 42. Ko M, Edelstein RA, Heslin KC, et al. Impact of the Univer-
18. Beckman TJ. Lessons learned from a peer review of bedside sity of California, Los Angeles/Charles R. Drew University Medi-
teaching. Acad Med 2004;79:343-6. cal Education Program on medical students’ intentions to prac-
19. Fenton C, Loeser H, Cooke M. Intersessions: covering the tice in underserved areas. Acad Med 2005;80:803-8.
bases in the clinical year. Acad Med 2002;77:1159. 43. Hurt MM, Harris JO. Founding a new College of Medicine at
20. Magrane D, Ephgrave K, Jacobs MD, Rusch R. Weaving wom­ Florida State University. Acad Med 2005;80:973-9.
en’s health across clinical clerkships. Acad Med 2000;75:1066-70. 44. Ramsey PG, Coombs JB, Hunt DD, Marshall SD, Wenrich MD.
21. Wood DF. Problem-based learning. BMJ 2003;326:328-30. From concept to culture: the WWAMI program at the University
22. Foley RP, Polson A, Vance JM. Review of the literature on PBL of Washington School of Medicine. Acad Med 2001;76:765-75.
in the clinical setting. Teach Learn Med 1997;9:4-9. 45. Oswald N, Alderson T, Jones S. Evaluating primary care as a
23. O’Neill PA, Morris J, Baxter C-M. Evaluation of an integrated base for medical education: the report of the Cambridge Com-
curriculum using problem-based learning in a clinical environ- munity-based Clinical Course. Med Educ 2001;35:782-8.
ment: the Manchester experience. Med Educ 2000;34:222-30. 46. Halaas GW. The Rural Physician Associate Program: success-
24. O’Neill PA, Jones A, Willis SC, McArdle PJ. Does a new under- ful outcomes in primary care and rural practice. Rural Remote
graduate curriculum based on Tomorrow’s Doctors prepare Health 2005;5:453.
house officers better for their first post? A qualitative study of 47. Hirsh D, Gutteron W, Batalden M, Krupat E, Pelletier S, Ogur
the views of pre-registration house officers using critical incidents. B. The Harvard Medical School Cambridge integrated clerkship.
Med Educ 2003;37:1100-8. J Gen Intern Med 2006;21:Suppl 4:180. abstract.
25. Harden RM, Crosby F, Davis MH, Howie PW, Struthers AD. 48. Cohen PA, Hirsh DA, Ogur BR, Elvin DM, Antman AM, Bor
Task-based learning: the answer to integration and problem- DH. A novel year-long learner centered tutorial for third-year
based learning in the clinical years. Med Educ 2000;34:391-7. medical students. J Gen Intern Med 2005;20:Suppl 1:26-7. abstract.

n engl j med 356;8  www.nejm.org  february 22, 2007 865

The New England Journal of Medicine
Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
 49. Elvin DM, Hirsh DA, Ogur BR, Cohen PR, Bor DH. Teaching 53. Why not the best? Results from a national scorecard on U.S.
translation: formally incorporating basic science into a third- health system performance. New York: Commonwealth Fund
year medical school curriculum reinforces lessons learned. J Gen Commission on a High Performance Health System, 2006.
Intern Med 2005;20:Suppl 1:38. abstract. 54. Hafferty FW. Beyond curriculum reform: confronting medi-
50. Ogur B, Hirsh D, Krupat E, Bor D. The Harvard Medical cine’s hidden curriculum. Acad Med 1998;73:403-7.
School/Cambridge integrated clerkship: a pilot multidiscipli- 55. Lempp H, Seale C. The hidden curriculum in undergraduate
nary, longitudinal clerkship. Acad Med (in press). medical education: qualitative study of medical students’ percep-
51. Whitcomb ME. What does it mean to be a physician? Acad tions of teaching. BMJ 2004;329:770-3.
Med 2005;80:1077-8. Copyright © 2007 Massachusetts Medical Society.
52. Solyom AE. Improving the health of the public requires
changes in medical education. Acad Med 2005;80:1089-93.

866 n engl j med 356;8  www.nejm.org  february 22, 2007

The New England Journal of Medicine

Downloaded from nejm.org at ST LOUIS UNIVERSITY on November 4, 2011. For personal use only. No other uses without permission.
Copyright © 2007 Massachusetts Medical Society. All rights reserved.