Who's the physician in charge? Generalist and specialist jurisdictions in professional practice

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

WHO’S THE PHYSICIAN IN CHARGE?
Generalist and Specialist Jurisdictions in Professional Practice

by
Mariam L. Krikorian

______________________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)

August 2016

I’m grateful to the National Science Foundation for partially funding my dissertation under the
mentorship of my advisor, Dr. Peer Fiss (Doctoral Dissertation Improvement Grant, award #1459996).

ii

DISSERTATION ABSTRACT
To manage complexity, a growing number of organizations take a multidisciplinary approach,
housing a variety of specialist and generalist professionals. Generalists and specialists possess
overlapping knowledge areas and expertise, which complicates the way jurisdictions are claimed
in the daily practice of work. Said differently, work boundaries between generalists and
specialists are permeable and often overlapping, which frequently results in problems with the
division of labor. This dissertation focuses on the healthcare context to highlight the challenges
and mechanisms that underlie the professional division of labor, specifically in the hospital
setting. Through two studies, I show how generalist physicians, called hospitalists, and specialist
physicians enact their respective jurisdictions in daily practice. The first study, uses ethnographic
and survey data from a children’s hospital. In this study, I describe the coordination challenges
that lead to conflict over work boundaries, the factors professionals use to identify their work
boundaries, and the boundary management practices professionals employ to resolve conflicts
over work boundaries. The second study uses survey data collected at a children’s hospital to
understand how hospitalist and specialist physicians make claims to their jurisdictions in relation
to the nature of the tasks (i.e. pediatric diagnoses) they commonly encounter. The first two
studies provide the springboard for future research, including a third study currently underway.
This study uses electronic medical record data from a children’s hospital to analyze the
performance outcomes resulting from the assignment of a hospitalist in charge of patients with
certain diagnoses when a specialist should have been in charge (and vice versa). Altogether, this
research has several theoretical and practical implications that aim to provide a more
comprehensive understanding of the professional division of labor.

iii

TABLE OF CONTENTS
DISSERTATION ABSTRACT .................................................................................................................. ii
LIST OF TABLES ...................................................................................................................................... v
LIST OF FIGURES ................................................................................................................................... vi
ACKNOWLEDGEMENTS ..................................................................................................................... vii
CHAPTER 1 – Introduction ...................................................................................................................... 1
OVERLAPPING EXPERTISE IN THE PROFESSIONAL WORKPLACE ........................................... 2
THE EVER-CHANGING HEALTHCARE WORKFORCE ................................................................... 5
The Affordable Care Act....................................................................................................................... 5
The Hospitalist, a New Medical Profession .......................................................................................... 9
CORE RESEARCH QUESTIONS ......................................................................................................... 14
STRUCTURE OF DISSERTATION...................................................................................................... 15
DATA AND METHODS ....................................................................................................................... 17
Empirical Context ............................................................................................................................... 17
Data Collection ................................................................................................................................... 17
Data Analysis ...................................................................................................................................... 18
THEORETICAL AND PRACTICAL CONTRIBUTIONS ................................................................... 18
CHAPTER 2 - Jurisdictional Negotiation in the Workplace: An Ethnography of Hospitalist and
Specialist Physicians (Phase 1 Study) ...................................................................................................... 21
ABSTRACT ............................................................................................................................................ 22
INTRODUCTION .................................................................................................................................. 23
THEORETICAL BACKGROUND ........................................................................................................ 25
Negotiating Work Boundaries ............................................................................................................ 25
Boundary Work in the Division of Labor ........................................................................................... 26
Jurisdictional Overlap of Generalists and Specialists ......................................................................... 28
METHODS ............................................................................................................................................. 31
Research Setting .................................................................................................................................. 31
Study 1: Ethnography of Hospitalist and Specialist Physicians .......................................................... 33
Study 2: Survey of Hospitalist and Specialist Physicians ................................................................... 36
FINDINGS .............................................................................................................................................. 37
Research Question 1: Contradictions .................................................................................................. 40
Research Question 2: Boundary Identification Factors ...................................................................... 46
Research Question 3: Boundary Management Practices .................................................................... 55
DISCUSSION ......................................................................................................................................... 61
iv

Implications ......................................................................................................................................... 62
Directions for Future Research ........................................................................................................... 66
CHAPTER 3 - The Nature of Work and Professionals’ Jurisdictional Claims (Phase 2 Study) ....... 69
ABSTRACT ............................................................................................................................................ 70
INTRODUCTION .................................................................................................................................. 71
THEORETICAL BACKGROUND ........................................................................................................ 73
Generalist and Specialist Professionals in the Workplace .................................................................. 73
Professional Division of Labor and Knowledge ................................................................................. 75
Nature of Tasks in Identifying Expertise ............................................................................................ 76
DATA AND METHODS ....................................................................................................................... 86
Empirical Context ............................................................................................................................... 86
Data Collection ................................................................................................................................... 87
Respondent Characteristics ................................................................................................................. 88
VARIABLES AND ANALYSIS ............................................................................................................ 90
Dependent Variable ............................................................................................................................ 90
Independent Variables......................................................................................................................... 91
Controls ............................................................................................................................................... 93
RESULTS ............................................................................................................................................... 94
DISCUSSION ......................................................................................................................................... 97
CONCLUSION ..................................................................................................................................... 102
CHAPTER 4 - Conclusion ...................................................................................................................... 104
CONTRIBUTIONS AND NEXT STEPS ............................................................................................. 105
Theoretical and Practical Contributions ............................................................................................ 105
Next Steps: Phase 3 Study ................................................................................................................ 108
FUTURE RESEARCH DIRECTIONS ................................................................................................ 114
REFERENCES ........................................................................................................................................ 117

v

LIST OF TABLES

TABLE 1 - Projected Growth in Service Demand by Care Setting and Source, 2013-2025 ................ 6

TABLE 2 - Contradictions Leading to Work Boundary Conflicts ....................................................... 44

TABLE 3 - Boundary Identification Factors and Boundary Management Practices ........................ 48

TABLE 4 - Demographic Characteristics Comparing Specialists and Hospitalists ........................... 88

TABLE 5 - Descriptive Statistics and Correlations ............................................................................... 96

TABLE 6 - Results of Logit Regressions for Assigning Hospitalist Physician by Diagnosis .............. 98

TABLE 7 - Descriptive Statistics for Hospitalist and Specialist Assignment at MCH ..................... 111

vi

LIST OF FIGURES

FIGURE 1 - Projected Primary Care Physician Shortfall, 2013-2025 ................................................... 7

FIGURE 2 - Hospitals, Physicians & Clinics, and Prescription Drug Spending by All Sources of
Funds, 1960-2014 (in USD Billions) ........................................................................................................... 8

FIGURE 3 - Boundary Work of Jurisdictional Negotiation ................................................................. 39

FIGURE 4 - Knowledge Fundamentality and the Use of a Specialist versus Generalist ................... 79

FIGURE 5 - Knowledge Breadth and the Use of a Specialist versus Generalist ................................. 83

FIGURE 6 - Percentage of Diagnoses by Specialty Demonstrating Consensus (Pi ≥ 0.6) Among
Specialists and Hospitalists ...................................................................................................................... 90

FIGURE 7 - Marginal Effects of Task Dimensions on Probability of Selecting a Hospitalist’s
Domain for a Given Diagnosis ................................................................................................................. 99

FIGURE 8 - Percentage of Patient Encounters per Year with Assigned Physician Matching
Consensus Perceptions of Assignment at MCH ................................................................................... 112
vii

ACKNOWLEDGEMENTS
As I wrote this dissertation, I often reflected on my 5 years in the PhD program at USC—
especially, the friends and colleagues that were essential in shaping my path. I’d like to
acknowledge these individuals’ vital impact on me—not only as a scholar, but as a person. First,
I’d like to thank my advisor and committee chair, Peer Fiss, for his unwavering support and
mentorship. Peer encouraged me to pursue my ideas without hesitation, and he challenged me to
think deeply about problems. When giving feedback, Peer asked provocative questions that
moved me toward discovering the theoretical significance of my research. Peer gave me enough
independence, yet was always there when I needed advice. It was very evident that he cared
foremost about my overall wellbeing, and I’m fortunate to have Peer as a cornerstone of my
academic pursuits.
Additionally, I’d like to thank my other committee members. Paul Adler’s creativity and
breadth of knowledge is truly inspirational to me. Paul was always there when I wanted feedback
or just to chat—he makes everyone feel like their success is his priority, and his approach has
meant a lot to me. Paul’s ability to see the fundamental aspects of a problem has taught me a lot
about dissecting my research questions and the issues I want to study. My other committee
member, John Romley, provided an excellent sounding board for my quantitative work, as well
as educated me on available resources as I pursue healthcare-related research. As a health
economist, John sharpened my skills to think broadly about the healthcare context and the policy
implications of my research. Together, Peer, Paul, and John formed an invaluable committee that
honed my academic goals and taught me much about the practice of research. I am extremely
grateful for each of their contributions.
viii

I’m also thankful to the individuals at my field site for their willingness to let me into
their world and openness to learn alongside me. The physicians at Metropolitan Children’s
Hospital (represented with pseudonyms in this dissertation) were always willing to explain their
experiences to me and answer my questions. After spending over 3 years with many of these
individuals, I was struck by how deeply they cared about their work and was moved by their
struggles to provide the best possible care to their patients. As a result, in writing this
dissertation, my hope was to both communicate the difficulties faced by medical professionals,
and also their resolve to do whatever they can for their patients, despite numerous hindrances.
Many friends I made at USC played a crucial role during my graduate studies. Vern
Glaser was hugely impactful—his advice was always filled with wisdom and sincerity. Vern’s
friendship from my first year in the program onward served as a stimulus for many of the
opportunities I pursued, and our conversations about research and life have been incredibly
formative. Jennifer Candipan has also been a light throughout this process, as so many of our
conversations put things in perspective. Our breaks from work were some of the best I’ve ever
taken, especially our bilingual karaoke nights. Other friends, including Derek Harmon and
Heejin Wu, have played a critical role in offering their feedback and engaging in deep,
generative discussions. Additionally, numerous USC faculty have contributed significant time
and effort toward my development. Paul Lichterman inspired me to pursue qualitative research
and provided a springboard for my studies in his year-long methods course. Kyle Mayer devoted
much time to share insights about practical ways to bear fruit in an academic career. Nandini
Rajagopalan, as a highly successful female professor, inspired me immensely and constantly
encouraged me to aim big. Tom Cummings always made sure to communicate his support and
do what he could to set me up for the road ahead. Several other former and current USC faculty
ix

were greatly insightful and offered guidance whenever I asked: Kelly Patterson, Shon Hiatt,
Feng Zhu, Florenta Teodoridis, Victor Bennett, Sun Park, Nan Jia, and Lori Yue. And of course
the USC staff—in particular Martha Maimone, Queenie Taylor, Jennifer Lim, and Michelle
Lee—each continually offered help and ensured all logistics ran smoothly.
Last, but most definitely not least, I’d like to thank my family and friends for building me
up during this time. My close friends provided strong support, especially Stephen Monte, Ali
Meisel, Carolyn Rumpeltes, and Astrid Wake. Each of these individuals have been an
exceptional respite throughout the years, and their talents and curiosity constantly invigorated
me. Above all, I’m extremely fortunate to have my family always by my side during the
dissertation process. My dad, Kapriel Krikorian, motivated me with his relentless determination,
enthusiasm to solve problems, and our thought-provoking conversations. He instilled in me not
just a love for learning, but also for teaching what I know. My mom, Mary Krikorian, continually
offered a listening ear and encouragement to seek God in everything. I’ve been inspired by her
incredible compassion paired with an unshakable strength. Both of my parents have sacrificed so
much, and in no way could I ever fully express my gratitude for all they have done. My sister,
Annie Krikorian, has also constantly been there for me, showing her support at many of my
conference presentations and throughout life in general. I’m grateful for Annie’s thoughtful
advice, friendship, and willingness to go on adventures with me around the world. Also a huge
thank you to my cousin, Ash Krikorian, for giving his time whenever I asked, always striving to
be helpful, and making me laugh like no other New Yorker could.
It is undeniable that this dissertation is largely the product of my interactions and
relationships with such incredible people. I’m moved by these individuals, and look forward to
the journey ahead.
x

I dedicate this dissertation to my sister, Annie, and my parents, Kapriel and Mary.
With much love and admiration for each of you.

xi

Whatever you do in life, surround yourself with smart people who will argue with you.
– John Wooden

1

CHAPTER 1 – Introduction

2

OVERLAPPING EXPERTISE IN THE PROFESSIONAL WORKPLACE
The organizational context of professional work has changed markedly over the last few
decades. Complexity is arguably a major driver of such changes, which include the introduction
of new roles (Currie, Lockett, Finn, Martin, & Waring, 2012), shifts in professional power to
bureaucratic control (Leicht & Fennell, 1997), and the emphasis on professional collaboration
(Okhuysen and Bechky, 2009). Professional organizations encounter increasing complexity
through a growing body of scientific and technological advances. Often, organizations manage
complexity by taking a multidisciplinary approach, employing a variety of specialist and
generalist professionals, with the aim of reducing costs, increasing efficiency, and providing
higher quality services (Brock, 2006; Johns, Laubscher, & Malone, 2011). However,
multidisciplinary professional environments experience challenges to effective collaboration,
since professionals with similar expertise may differ in the way they interpret and apply their
jurisdictions in daily practice.
Professional jurisdiction is simply the linkage between a profession and its work (Abbott,
1988). Recent scholars would add that jurisdictions entail not just a linkage of a profession to its
work, but also the social identities of the relevant professionals (Ashcraft, 2012). Professional
jurisdiction represents a domain of expertise whereby the professional can demonstrate control
over a particular knowledge area. The more organized a profession is, the stronger their
jurisdictional claims (Hughes, 1963). Jurisdictions often represent an exclusive claim, which
creates a system of interdependence between professionals (Abbott, 1988). However, exclusivity
can sometimes shift as one type of professional invades the domain of another. Jurisdictional
conflict generally is conceived in the workplace, then extends before the public mind, and finally
into the realm of law (Abbott, 1988, p. 139). Although settlements of jurisdictions concerning
3

client differentiation largely occurs in the workplace, they rarely depend on strict institutional
standards, since actors can be carriers of institutional meanings, but at the same time their
interpretations can be representations of individual agency (Zilber, 2002).
In the work environment, when jurisdictional contests occur, roles must be actively
interpreted and negotiated between different professional collaborators, and cannot necessarily
be formally assigned (Rawson, 2002). The medium of exchange is typically abstract knowledge
which professionals negotiate for territorial control (Abbott, 1988). Thus, the more abstract
knowledge a particular profession controls, the more powerful they are relative to other
professions. Other means by which professionals gain power are through a larger association
membership base and a more sizable professional workforce (Abbott, 1988; McGregor, 2010).
The logic here is that the greater the number of professionals in an association, the greater their
capability to marshal support and gain power over other weaker professions. In addition, a larger
labor supply of a particular profession increases the likelihood of dominance over a client base.
When professionals contest jurisdictions, a simultaneous process of negotiation occurs in
which professionals collectively construct identities (Ashcraft, 2012). Negotiation refers to the
practices of individuals who depend on the cooperation of others in order accomplish their goals
(L. L. Thompson, Wang, & Gunia, 2010). In the context of professionals, negotiation involves
establishing their practice through a strategic identity project, in which there is a concerted effort
to persuade multiple audiences to accept a particular answer to the question, “what is this line of
work?” and other answers related to how complex or valuable the work is (Ashcraft, 2012).
Since identities develop because of interoccupational competition over expertise,
professions develop in relation to each other. However, interprofessional work diminishes
professional power, since individual autonomy weakens in collaborative settings (Webb, 1992).
4

For many industries, including healthcare, interprofessional relations are unavoidable—effective
healthcare delivery depends greatly on the quality of interprofessional teams and coordination.
Sometimes, such teams are successful by implementing more formalized guidelines, such as care
pathways for treatment of various diagnoses, to enhance coordination and provide clearer
boundaries between professional workers (Deneckere et al., 2013). Other times, implementation
of a coordination role, such as an “integrator,” becomes crucial to managing interprofessional
practice (Lawrence & Lorsch, 1967b).
The social organization of professionals in the workplace hinges upon the division of
labor (Abbott, 1988). The division of labor in a work system can be described along two primary
dimensions that are rooted in cognitive and behavioral aspects (Gulati, Puranam, & Tushman,
2012; Sinha & Van de Ven, 2005). The first is a horizontal component in which work is
distributed across many organizational units within one organization or multiple organizations.
The second is a vertical component in which resources, knowledge and authority reside in one
level of an organization, or may be partitioned among many hierarchical levels. Both of these
components of the division of labor in a work system are premised on structural rules and norms
of doing work, interpersonal relationships among organizational members, timeframe for
processing activities, and goals of the work system, including those of its subsystem units
(Dougherty, 2001; Lawrence & Lorsch, 1967a). The horizontal and vertical dimensions of the
division of labor pose a variety of challenges for the coordination of work across different
specialty units, as well as for identifying and demarcating task boundaries within and across
hierarchical levels (Sinha and Van de Ven, 2005). Such problems can become especially
pronounced in multidisciplinary work settings comprised of professionals with potentially
overlapping knowledge bases and skillsets.
5

Much of the scholarship on professional boundaries focuses on interactions between two
occupational groups, rather than multiple occupational groups (Barrett, Oborn, Orlikowski, &
Yates, 2012). Further, it is unclear how today’s more collaborative professional settings yield
new identities and understandings of jurisdictional domains (Adler, Kwon, & Heckscher, 2008).
This is especially important since collaborative environments are particularly vulnerable to
breakdowns in the performance of work as professionals unwaveringly adhere to familiar
perceptions of their work boundaries, potentially creating an inflexible environment (Ashforth,
Kreiner, & Fugate, 2000). I now explore these issues of overlapping jurisdictions and the
division of labor in the healthcare context.
THE EVER-CHANGING HEALTHCARE WORKFORCE
The Affordable Care Act
Since the rise of modern medicine in the 19
th
century, the medical field has experienced
many changes. Some of these changes are a result of technological advances and scientific
discovery; others, a product of market forces and policy reforms. Such factors have the power to
transform the structure of the entire field, reshaping the nature of work and the way medical
professionals practice. One of these recent changes is the policy mandate, the Affordable Care
Act (ACA), which was signed into law in March of 2010. The ACA is said to be the most
significant regulatory act to reshape the U.S. healthcare system since Medicare and Medicaid
was passed in 1965 (Vicini & Stempel, 2013). The ACA uses subsidies, guidelines, and
insurance exchanges, in an attempt to make healthcare more affordable to all and to increase
cost-effectiveness as well as quality of healthcare.
After passage of the ACA, the U.S. healthcare system underwent a series of sweeping,
unprecedented changes. Some of these changes included an emphasis on preventative and
6

primary care, and directives against denying individuals insurance coverage based on pre-
existing conditions. Arguably, the most important of these changes is the expansion of private
and public insurance coverage through Health Insurance Exchanges (McDonough, 2011). The
percentage of uninsured decreased from 22.3 percent in 2010 to 12.9 percent in the first nine
months of 2015 (National Health Interview Survey Early Release Estimates, February 2016).
With over 20 million formerly uninsured Americans now entitled to seek medical services, one
consequence of these changes has been a shortage of care providers across a variety of
professions. The shift in demand is primarily a result of the passage of the ACA, as well as the
demographics of a more aging population. Table 1 shows the projected growth in demand for
health services resulting from both the ACA and demographic changes across various settings,
including office visits, outpatient visits, emergency visits, and hospital days. For each setting, a
double digit growth is expected from 2013 to 2025.

TABLE 1 - Projected Growth in Service Demand by Care Setting and Source, 2013-2025

Source: IHS Global Insight Report Prepared for the Association of American Medical Colleges (March 2015)

However, despite the growth in demand, IHS Global Insight reported future estimates of
physician shortages, both amongst generalists (primary care) and specialists. Specifically looking
at primary care in Figure 1, the projected physician shortfall between 2013 and 2025 is anywhere
from 12,500 to 31,100 primary care physicians.
Care Setting
Growth from
Changing
Demographics
Growth from
ACA Coverage
Expansion
Office visits 14% 4%
Outpatient visits 15% 2%
Emergency visits 12% 0%
Hospital inpatient days 23% 1%
7

Another unintended consequence of the ACA is the rampant consolidation of not only
healthcare providers, but also managed care companies. The trend reflects the efforts of both
parties to develop the scale and size necessary to succeed under the changes the ACA demands.
These demands include pressure to reduce costs, as well as goals to meet efficiency and quality
of care standards. As a result, both healthcare providers and managed care organizations are
building heft in order to negotiate against each other. Thus, notable aspect of the consolidation is
the increasing number of hospital deals. According to Irving Levin Associates, the pace of deals
in 2015 is the fastest since the ACA passed in 2010 (Mathews, “Healthcare Providers, Insurers
Supersize,” The Wall Street Journal 2015).

FIGURE 1 - Projected Primary Care Physician Shortfall, 2013-2025

Source: IHS Global Insight Report Prepared for the Association of American Medical Colleges (Mar 2015)

8

In support of the ACA, the American Hospital Association (AHA) agreed to funding cuts
amounting to $155 billion, which has put even more impetus on hospitals to reduce expenses.
The potential reduction in hospital expenses could dramatically decrease national healthcare
costs, since hospitals are the most expensive of all types of health services in the U.S. (Figure 2).
However, while healthcare costs are one part of the story, the quality of care provided is the
other. Based on data collected by The Commonwealth Fund, the U.S. has the highest rate of
medical errors compared to other countries. With regard to readmissions, 19.6 percent of all
hospitalized patients are re-hospitalized within 30 days (Jencks, Williams, and Coleman, 2009).
Astoundingly, about half of those discharged patients who were re-hospitalized did not have
medical interventions for transitional or preventative care at the time of their initial discharge,
which could have reduced the patients’ likelihood of re-hospitalization (Jencks, Williams, and
Coleman, 2009).

FIGURE 2 - Hospitals, Physicians & Clinics, and Prescription Drug Spending by All
Sources of Funds, 1960-2014 (in USD Billions)

Source: Kaiser Family Foundation, “Health Spending Explorer” (Dec 07 2015)

9

Paradoxically, in the U.S., the high cost of care does not translate to high quality care
relative to other nations. Data from the Organization of Economic Cooperation and Development
(OECD) illustrates that the U.S. has the highest per capita healthcare expenditure, compared to
the other OECD countries. Notably, the OECD data also indicates that the U.S. has the worst
error record in providing patient care, as compared to the other OECD countries. How can the
most costly healthcare system be among the lowest in quality? This contradiction serves as a
broader motivation for my dissertation, and it is my intent to better understand possible reasons
for the cost and quality disconnect, particularly within the hospital context.
Since the ACA also makes it possible for the formerly-uninsured to be able to afford care,
the increasing number of patients seeking medical attention will certainly compound the current
issue facing the medical field: a shortage of both generalist and specialist physicians. Hospitals
are not only faced with the challenge of containing costs and increasing quality, but also treating
a changing demographic of patients now gaining access to care. Because the ACA also will
reduce reimbursements, it is speculated that these issues will result in 15 percent more
institutional providers who will be bankrupt by 2019 (Bessler, 2012).
The Hospitalist, a New Medical Profession
What can hospital leadership do to ensure survival? One solution has been to increasingly
specialize and divide up tasks within the hospital, in an effort to provide more focused patient
care. With advances in medicine resulting in the need for more depth of knowledge, hospitals
have increased specialist roles to keep up with changes in practice that would ultimately improve
care delivery (Burns & Muller, 2008). Hospitals seeking a patient-centered focus to improve the
quality of care, as well as reduce healthcare costs, place greater emphasis on multidisciplinary
care teams comprised of different specialists working together to treat patients (Meguid et al.,
10

2015). However, such settings require someone who coordinates the work of multiple specialists
to ensure high quality care and cost containment.
A possible strategy to address some of the major issues plaguing multidisciplinary
hospital settings— namely cost-effectiveness, coordination to improve the quality of care, and
meeting patient demand— may rest with an interventionist role (Bessler, 2012). According to
Bessler, the interventionist should be one who manages costs and has a mind for efficiency.
Additionally, the interventionist should have first-hand knowledge of medical practice and
innovates to improve the quality of care. Moreover, the interventionist, should be an active
clinician in the hospital with a broader knowledge and skillset, who can attend to the greater
number of patients seeking care. These requirements would appear to qualify the interventionist
as the role of a generalist physician. About 15 years before the ACA was passed, such an
interventionist role was just in its nascent stages. The role is called the hospitalist
1
, a generalist
physician who coordinates patient care in the hospital setting. Hospitalists have been found to be
effective at reducing total costs and charges in the hospital (Coffman & Rundall, 2005; Meltzer
et al., 2002), though findings are inconsistent with regard to hospitalists improving patient
outcomes (Pannick, Beveridge, Wachter, & Sevdalis, 2014; Wachter, Goldman, & Hollander,
2005).
The role of the hospitalist has gradually evolved with the changing healthcare
environment. Since becoming law, the ACA has implemented several measures as a push toward
value-based reporting and purchasing of health services, which has an important influence on the
way physician performance is assessed (Chien & Rosenthal, 2013). These measures have shifted

1
Hospital Medicine is actually a specialty of medicine, and hospitalists are considered specialists since they
specialize in inpatient care. However, relative to other specialist physicians in the hospital setting (who are actually
subspecialists, e.g. pediatric cardiologists), hospitalists are considered generalists in the organization and among
their colleagues. For simplicity, I refer to the hospitalists as generalists and the subspecialists as specialists.
11

the paradigm held by some that the value of the hospitalist program is to churn through the
highest volume of patients at the lowest cost, to one which emphasizes providing the highest
quality of care at the lowest cost (McGuire, 2011). In managing the quality-cost balance, the
hospitalist’s role is not devoid of challenges, and many of these surface in their work
environment. One primary challenge will be the focus of this dissertation: the division of labor
between hospitalist and specialist physicians.
In their daily practice, hospitalists and specialists encounter ambiguous work boundaries,
and this is increasingly becoming the case with the trend toward comanagement practices.
Comanagement is the shared responsibility, authority, and accountability between hospitalist and
specialist physicians for the care of hospitalized patients (Glasheen, Siegal, Epstein, Kutner, &
Prochazka, 2008). The implementation of comanagement practices varies from one hospital to
the next. Typically, a comanaging hospitalist follows the patient during their entire hospital stay
until discharge, rather than just select consultations. Comanagement practices are more
widespread than ever, with more than 85 percent of hospital medicine groups performing some
degree of comanagement (Society of Hospital Medicine, 2008).
Comanagement has increased over the last several years for numerous reasons. First,
because they are in the hospital around the clock, hospitalists have been thought of by some as
“glorified house staff,” often being conflated with the role of residents. Consequently, physicians
and other medical professionals who share this perception “dump” grunt work on hospitalists
(Beresford, 2011). Comanagement practices should theoretically equalize the relationship
between specialists and hospitalists, even if a specialist is the physician of record (Cheng, 2012),
and thus, may be an attempt for hospitalists to be viewed on the same level as their physician
colleagues. Another reason for increasing comanagement in hospitals is that the aging population
12

in the U.S. calls for more complex care, because these patients often have more comorbidities.
Complex care typically means numerous specialists are involved, and the comanaging role of a
hospitalist becomes more necessary. Finally, policy passed in 2003 to limit resident work hours
has increased the involvement of hospitalists in patient care, since they are available to provide
generalized hospital care 24 hours per day, seven days per week.
Siegal (2008) identifies several advantages and disadvantages of comanagement. Some
benefits include improved quality of care, patient satisfaction, and collegiality among
professionals. However, comanagement can facilitate disengagement of specialists as they may
perceive a hospitalist is making treatment decisions on their patient. Further, comanagement can
result in fragmented care because of the presence of multiple managers. Also, comanagement
may lead to reduced job satisfaction of hospitalists as they may experience a steadily increasing
workload that results in burnout. While the costs and benefits of comanagement practices may be
apparent, there are few studies that explore what successful comanagement practices actually
look like (Mayer, Skinner, & Freed, 2009; Stucky & Kuelbs, 2007).
Comanagement is an important concern for a variety of reasons, poignantly illustrated in
a case study published recently in the New England Journal of Medicine (Stavert & Lott, 2013).
The authors of this article discuss a patient who was admitted into the intensive care unit of a
hospital, and during the course of his 11-day stay, over 40 physicians participated in his care.
However, no one was clearly assigned as the physician ultimately in charge of overseeing the
diagnosis and treatment of the patient, which resulted in several issues. First, there were
breakdowns in communication between the medical professionals on the care team, and often a
duplication of work, namely the performance of tests and procedures in providing care.
Additionally, the patient did not receive timely and effective care, and his condition continued to
13

deteriorate. Another major issue that emerged is what the Stavert and Lott attribute to the
“bystander effect,” namely that each of the physicians thought that another physician was
making decisions on the patient, when in fact no one was. Decisive action was finally taken
when the patient experienced an acute event during the course of his stay. However, until that
point, the lack of timely treatment and the involvement of superfluous medical staff led to
escalating costs and produced a hefty expenditure on needless resources. Of greatest
consequence, the absence of a physician clearly assigned as the one in charge meant that no one
was primarily held accountable, and the quality and safety of care provided to this patient was
sacrificed. As this example shows, who is in charge matters in ensuring high quality care while
avoiding unnecessary costs.
Several actions can be taken to improve comanagement practices (Beresford, 2011):
clarifying a shared vision; establishing mutual goals and expectations; deeply understanding a
given hospital’s patient population; matching staffing resources with hospital caseload; and
identifying value propositions from both the hospitalist and specialist perspectives. However,
even if hospitals implement such improvements to enhance comanagement practices, what are
factors that can undermine these strategies? Specifically, what do we know about professional
practice in the multidisciplinary workplace to inform our understanding of effective
comanagement practices?
This dissertation turns to organizational theory for possible answers. Jurisdictional
overlap means that work boundaries may be ambiguous between different types of professionals,
and understanding the way work boundaries are drawn is critical in the hospitalist’s setting, as
several studies show inconclusive evidence on the effectiveness of these professionals as
coordinators of other specialists (Goodrich et al, 2012; Lindenauer et al, 2007). One possibility
14

for the inconsistent findings concerning hospitalists’ effectiveness as comanagers and
coordinators may be challenges arising from the division of labor between hospitalists and
specialists. To explore the challenges of the division of labor between professionals with
overlapping expertise, my dissertation examines the interactions and perceptions of hospitalist
and specialist physicians in their daily practice. The workplace provides a setting in which we
can understand how the division of labor is established and also the way a set of negotiated,
seemingly normative rules of professional jurisdiction are actually enacted (Leicht, Fennell, and
Witkowski, 1995). With these considerations in mind, the broader objective of my research is to
understand how the division of labor works between professionals with overlapping jurisdictions,
in particular generalists and specialists.
CORE RESEARCH QUESTIONS
In light of the issues I outlined in the healthcare context, I use organizational theory to
address my overarching research question, which is as follows: How do professionals with
overlapping expertise enact their jurisdictions in the daily practice of work? I focus on generalist
and specialist practices as a representation of professionals with overlapping expertise. In order
to delve into my broader research question, my dissertation hones two specific questions:
1. What are coordination challenges in collaborative professional environments that lead to
conflict over work boundaries?
2. How do professionals with overlapping expertise identify work boundaries?
3. How do professionals with overlapping expertise resolve conflicts over work boundaries?
4. How do generalist and specialist professionals claim their jurisdictions in relation to the
attributes of tasks they commonly perform?
15

Focusing on these two questions, the aim of this dissertation is to develop a theoretical and
empirical account of how professional jurisdictions are negotiated in complex work
environments, where expertise often overlaps and it is not always clear who should be in charge.
STRUCTURE OF DISSERTATION
In addition to this introductory chapter, my dissertation consists of two studies that
address my research questions, as well as a concluding chapter. The following summarizes each
of the remaining chapters.
Chapter 2 presents a study (“Phase 1”) in which I examine generalist and specialist
professionals’ interactions as they establish and manage work boundaries that represent their
respective jurisdictions. This study develops a theory of how jurisdictional boundaries are
negotiated in the workplace, particularly when domains of expertise overlap due to greater
complexity. Through ethnographic and survey methods, I investigate how generalist and
specialist physicians at a children’s hospital are assigned patient cases. Such a setting serves as
an extreme case in which work boundaries between generalist and specialist professionals
substantially and frequently overlap. This study describes the interactions and decision-making
processes of generalist physicians (“hospitalists”), who act as generalists that coordinate work
between themselves and specialist physicians. Through detailed observations, I derive a
boundary work model of jurisdictional negotiation that highlights the coordination issues that
lead to work boundary conflicts between generalist and specialist professionals, factors
generalists and specialists rely upon to identify work boundaries, and boundary management
practices professionals use to resolve issues.
In Chapter 3, I conduct a study (“Phase 2”) of generalist and specialist professionals
claims to jurisdictions in relation to the nature of tasks they commonly perform. Four task
16

dimensions are examined, based on prior scholarly work and observational data collected at
MCH: fundamentality, breadth, client contact, and resource use. To carry out this analysis, I
administered a survey to 182 hospitalist and specialist physicians in eight divisions at a
children’s hospital (36 percent response rate). The survey presented the most common pediatric
diagnoses for the various specialties, and asked respondents whether they believe each diagnosis
belongs in the specialist’s domain or in the hospitalist’s domain. Descriptive statistics provide
insight into demographic differences between generalist and specialist professionals (e.g. level of
experience, organizational tenure), as well as how much agreement, or lack thereof, respondents
demonstrated concerning demarcations of their domains of expertise. Results show that
generalists, rather than specialists, were less likely selected to be in charge of tasks with high
fundamentality, greater breadth, and high client contact. No significant results were found for the
resource use category.
My dissertation finishes with Chapter 4, which summarizes key findings of my two
studies, as well as the theoretical and practical implications of this research. Chapter 4 also
introduces a third study (“Phase 3”) that builds on the other two of my dissertation. In Phase 3, I
use electronic medical record data to compare perceptions of whether a hospitalist or specialist
should be in charge from the Phase 2 survey to actual assignments of hospitalists and specialists
in the medical records. Future work entails analyzing quality and cost outcomes associated with
physician assignments, and exploring what happens when actual versus perceived assignments
misalign.
17

DATA AND METHODS
Empirical Context
The field setting of my research is Metropolitan Children’s Hospital (MCH), which is a
large tertiary care medical facility with over 300 beds. The studies I conduct center on generalist
and specialist physicians at MCH. The generalists I study belong to the Hospital Medicine (HM)
division and are called hospitalists. Hospitalists at MCH are general pediatricians acting as
coordinators of patient care who solely work in the hospital-setting and do not have outside
clinical practices. MCH hired its first hospitalist in 2000, and subsequently, the HM division was
established in 2009. Today, there are about 50 active hospitalists at MCH (contract, full-time,
and fellows), which parallels the rapid growth of role nationwide. Hospitalists at MCH serve as
“primary” or “consultant” physicians in 10 specialty divisions in the Department of Pediatrics at
MCH. For a patient admitted into a particular division in the hospital, the Hospitalist may be
assigned as primary physician (or the foremost decision-maker in treating a patient), while the
specialist may be a consultant physician (only provide decision support to the hospitalist on a
required or requested basis); alternatively, the specialist may be assigned as the primary
physician for a particular patient, while the hospitalist may serve a consultant role. Since
hospitalist and specialist physicians’ primary or consultant assignments effectively delineate
jurisdictional domains in daily practice, MCH is an ideal context to study my research questions.
Data Collection
For Phase 1, I collected approximately two years of ethnographic data, in which I spent
over 100 hours observing physician rounds and attending various meetings. After analyzing my
observational data and developing a conceptual model of jurisdictional negotiation, I decided to
go back to my field site to collect additional data, as I noticed that the specialists’ perspective on
18

physician assignment was relatively lacking in comparison to the hospitalists’. Accordingly, I
administered a survey to 13 specialists and four hospitalists total to further refine the conceptual
model derived from my ethnographic work.
Phase 2 consists of survey data I collected from hospitalist and specialist physicians
asking who should be in charge for various common pediatric diagnoses (i.e. either a hospitalist
or specialist should be assigned as the primary attending). I later asked hospitalist focus groups
to categorize each of the diagnoses in the survey along four dimensions. From there, I analyzed
how the physician respondents of my survey associated diagnoses (task) dimensions with either a
specialist or hospitalist in charge.
Data Analysis
To analyze my ethnographic and open-ended survey data in Phase 1, I used grounded
theory (Glaser & Strauss, 1967). My research questions and theory emerged through an inductive
process. After coding my data through an iterative process, I developed a data structure using
Gioia’s methodology (Gioia, Corley, & Hamilton, 2012). In Phase 2, I used econometric
analysis, specifically logit regression, to analyze the results of the survey of whether a hospitalist
or specialist should be the primary attending for various common pediatric diagnoses.
THEORETICAL AND PRACTICAL CONTRIBUTIONS
This research has several theoretical and practical implications. With regard to theory,
Phase 1 derives a model of boundary work of jurisdictional negotiation in the workplace.
Ethnographic and survey data from Phase 1 reveal rich insights into the microprocesses of
jurisdictional negotiation that occur in the multidisciplinary workplace between professionals
with overlapping expertise. The conceptual model explains coordination problems that may lead
to conflict over work boundaries between professionals (“contradictions”), the factors
19

professionals rely upon in identifying the work boundaries, and also the means by which
professionals resolve work boundary conflicts. In particular, Phase 1 describes the way
generalists and specialists negotiate jurisdictions in day-to-day practice, showing the
intrapersonal and interpersonal aspects of negotiation that emerge when various contradictions in
the workplace arise. This study also provides insights into how new professional roles are
integrated in the organization and how such roles can coexist with and alter the work boundaries
of existing professionals.
Phase 2 describes the way features of the tasks professionals commonly perform serves to
delineate the overlapping domains of expertise between generalists and specialists. This provides
theoretical insight into the way generalist and specialist roles operate in the workplace context.
Much of prior literature provides abstract explanations of the roles of generalists compared to
specialists, while this research illustrates how each type of professional practices relative to the
other. In making these distinctions, Phase 2 provides greater clarity into the division of labor
between collaborating professionals, which is of greater significance as modern organizations
increasingly become multidisciplinary environments.
From a practice perspective, this research offers numerous insights into the streamlining
of the division of labor. Prior scholarly work overlooked practice nuances between generalists
and specialists, while this research provides in-depth explanations and evidence that could
promote more effective practices between professionals with overlapping expertise. The
conceptual model derived in Phase 1 offers an actionable training tool to promote discussion
about work boundary conflicts between professionals, as well as a culture of negotiation by
implementing systems that facilitate work boundary management practices. This model can be
transferrable to a variety of settings. Phase 2 also develops a method in which a survey can be
20

administered to address work boundary conflicts between professionals and identify points of
contestation. Such practices can serve to better direct work flow, clarify boundaries between
professionals, and construct a system of accountability in which professionals when they should
be in charge, as well as what to do when there are jurisdictional conflicts.
In the healthcare context, specifically, the theory developed in my research offers insight
into effective comanagement practices between hospitalists and specialists, especially in terms of
how authority is assigned to physicians that also collaborate in providing patient care. Such
findings have implications for creating greater accountability in medical practice. This work also
expounds upon the role of the hospitalist, its importance in the coordination of care, and its
assimilation into an environment already inhabited by specialist physicians.

21

CHAPTER 2 - Jurisdictional Negotiation in the
Workplace: An Ethnography of Hospitalist and
Specialist Physicians (Phase 1 Study)
22

ABSTRACT
Modern organizations are experiencing greater pressures to specialize, and in response, become
multidisciplinary environments housing a variety of specialist and generalist professionals whose
jurisdictions overlap. Accordingly, this study develops a theory of how professionals with
overlapping expertise identify and manage their jurisdictions in daily practice. Through
ethnographic and survey methods, I investigate how generalist (“hospitalist”) and specialist
physicians at a children’s hospital made decisions about who should be in charge of various
tasks. Using a grounded theory approach, I provide rich descriptions of the coordination
challenges that lead to conflict over work boundaries (contradictions), the factors professionals
rely on to identify work boundaries (intrapersonal negotiation), and the practices professionals
employ to manage work boundaries when conflicts arise (interpersonal negotiation). I derive a
boundary work model of jurisdictional negotiation that has implications for organizations
wishing to streamline the division of labor among collaborating professionals.
23

INTRODUCTION
A fundamental problem of organizing is managing a growing variety of specialized work.
Modern organizations are experiencing mounting pressures toward specialization, largely due to
the expanding repertoire of knowledge and new technologies (Spitz‐ Oener, 2006). As a result,
scholars believe we are entering an “age of hyperspecialization,” in which work is increasingly
compartmentalized to reduce costs and enhance the quality of output (Johns et al., 2011).
Organizations often respond to these pressures by taking a multidisciplinary orientation, as
evidenced for instance by law firms that aim to become a “one-stop-shop” for clients as they
expand operations to multiple locations (Brock, 2006). Accounting, consulting, and engineering
firms have also taken a similar direction (Brock & Powell, 2005). Multidisciplinary
organizations, however, must implement more sophisticated means to enhance collaboration and
coordination between different types of specialists. In these settings, effective coordination
functions, such as managers or other linking roles (Galbraith, 1977; Lawrence & Lorsch, 1967a)
are crucial to the success of multidisciplinary teams.
Coordination functions are often held by generalist professionals, who can use their
broader knowledge to make the appropriate linkages between different specialists (Puranam,
Raveendran, & Knudsen, 2012). However, as the complexity of work increases, such generalist
roles must possess greater technical expertise to effectively coordinate workflow (Langbert,
2005). As a consequence, generalists start to specialize when they gain more technical know-
how and experience (Gupta, 1984), which results in overlapping jurisdictions between
generalists and specialists in the workplace (Clark, 1996). Jurisdictions serve to identify work
boundaries among professionals, and one way these work boundaries are represented is through
task assignments—essentially, professionals assigned to perform certain work for a particular
24

client (Abbott, 2005). As jurisdictions overlap, work boundaries between generalists and
specialists are less clear, which can lead to coordination problems and conflict. Although such
problems are widely spread and growing in importance, the phenomenon of overlapping
expertise is understudied (MacDuffie, 2007)and thus, this research aims to understand the
following question: How do professionals with overlapping expertise negotiate their
jurisdictional domains in the daily practice of work?
I explore this overarching research question with two studies. In Study 1, I conducted an
ethnography at a children’s hospital, observing the task assignment practices of specialist and
generalist physicians. The generalist physicians are called hospitalists, a relatively new medical
profession responsible for coordinating and providing patient care solely in the hospital setting.
Study 2 is a survey of open-ended questions concerning physician assignment, which was
administered to 17 physicians (13 specialists and four hospitalists) at the children’s hospital.
Both studies provided rich insights into: (1) coordination challenges, or “contradictions,” in
collaborative work settings that lead to conflict over work boundaries; (2) factors that
professionals rely upon to identify work boundaries; and (3) the means by which professionals
manage work boundaries when conflicts arise. My data provide detailed descriptions of the
nature of jurisdictional negotiation among professionals who share expertise. Through my
analysis, I develop a model of boundary work for negotiating jurisdictions that is comprised of
three main components: contradictions, intrapersonal negotiation, and interpersonal negotiation.
In particular, the model shows how professionals identify and manage work boundaries in
response to coordination challenges that may arise in the assignment of work. This research
provides an actionable model for delineating overlapping roles in the workplace, which is of
25

particular importance to managers who wish to make informed decisions concerning the division
of labor between collaborating professionals.
THEORETICAL BACKGROUND
Negotiating Work Boundaries
Work boundaries simplify the environment in which professionals practice, providing a
means for effective coordination and collaboration (Ashforth et al., 2000; Majchrzak, Jarvenpaa,
& Hollingshead, 2007). However, sometimes work boundaries do not clearly define who is in
charge of particular tasks. For example, the degree to which work boundaries are permeable
varies between individuals and contexts, and this can pose a challenge in designating roles
(Kreiner, Hollensbe, & Sheep, 2009). Conversely, if work boundaries are broadly accepted and
upheld, they can become a norm of practice (Zerubavel, 1991).
Work boundaries can often change, especially amidst professionals who have the
autonomy to control the practice and criteria for evaluation of their work (Freidson, 2001).
Professionals can establish and alter work boundaries as they make claims to their jurisdictions,
or more specifically, domains of expertise and control. While there may be bureaucratic
processes that institutionalize professional jurisdictions and enable occupational closure over
time, the autonomy professionalism grants to individuals creates an interesting tension in the
workplace, as professionals have the ability to re-create their domains of expertise in light of
their current organizational context (Regoli, Culbertson, Crank, & Poole, 1988).
Correspondingly, professionals persistently negotiate jurisdictions by establishing and re-
establishing work boundaries. Scholars describe two primary modes of negotiation,
intrapersonal and interpersonal (L. L. Thompson et al., 2010). Negotiation may take place as an
individual professional contemplates trade-offs in claiming certain types of work for themselves
26

(Cohen, 2013a). For example, a professional may consider their personal workload or experience
level as they determine whether to claim certain types of work. This form of negotiation is
intrapersonal, meaning that behaviors and outcomes related to negotiation depend on the internal
perceptions and experiences of the negotiator. Alternatively, negotiation can take the form of a
head-to-head interaction between different professionals (Abbott, 1988). For example, a
professional may make claims to certain types of work based on dealings with others in their
work context. Such negotiations are interpersonal, in that behaviors and outcomes related to
negotiation depend on the presence of other professionals.
Thus, jurisdictional negotiation is an inherent component of the professional workplace,
both at an intrapersonal and interpersonal level. We now look to a mechanism for jurisdictional
negotiation that facilitates the division of labor between professionals in complex work
environments.
Boundary Work in the Division of Labor
Individuals engage in boundary work practices that solidify and give meaning to their
mental frameworks (Nippert-Eng, 1996). Often, boundary work involves identifying,
challenging, and upholding existing conceptions of social systems, thereby allowing categories
and classifications to exist (Schwartz, 1981). More specifically, in the organizational context,
professionals participate in boundary work as they define boundaries that exclude others,
ultimately setting themselves apart from different professionals in the organization (Andrew
Abbott, 1995). Several studies have found how work boundaries can shift as a new technology or
innovation is introduced into the work environment (Barley, 1986; Bechky, 2003; Boland Jr,
Lyytinen, & Yoo, 2007). Novelties such as the implementation of a new technology can have
27

both enabling and constraining effects on interactions and collaborations within the workplace
(Barrett et al., 2012).
While institutional-level factors, like formal education programs and licensure, are
fundamental mechanisms for establishing professional jurisdictions, the organizational
environment and the content of the actual work professionals perform on a daily basis are also
critical. Scholars explain how professionals’ claims to work can serve as rhetorical devices that
reinforce professional control (Zetka, 2003). For example, boundary work theorists explain that
professionals construct their jurisdictions in a way that is consistent with the advancement and
protection of their professional status (Allen, 2000; Gieryn, 1983). Further, jurisdictional claims
can act as an evaluative and integrative device when individuals encounter complexity and
uncertainty (Parsons, 1967), and are therefore a means for legitimating an appropriate response,
particularly in the division of labor.
In the work setting, jurisdictional claims involve an appraisal of the expertise required to
perform common tasks, and thus, task assignment is one way of representing professionals’
jurisdictional domains (Abbott, 2005). Through boundary work, professionals can construct task
domains and assert authority over different types of work relative to other professionals in the
organization (Allen, 2000). This phenomenon is called job crafting, in particular the process of
individuals creating work boundaries. Job crafting practices are largely shaped by intrinsic and
individual factors, such as personal motivations to perform certain types of work over others
(Wrzesniewski & Dutton, 2001). Job crafting occurs when there is a misfit between the
individual and their job that usually results from a misalignment between the work an employee
does and their abilities (Tims & Bakker, 2010). Such misalignment will result in a particular
28

form of job crafting called task crafting, which involves employees changing the form and
number of activities they engage in while at work (Berg, Dutton, & Wrzesniewski, 2013).
However, most prior work on job crafting has primarily examined the relationship of
personal job motivations to performance, and does not offer a detailed explanation of the content
and nature of professional work (Wrzesniewski, LoBuglio, Dutton, & Berg, 2013), which may
serve as antecedents of job crafting practices. The content and nature of work can fundamentally
shape the division of labor in complex and collaborative settings, where expertise often overlaps
between professionals, suggesting that more research is needed to understand how jurisdictional
overlap is managed.
Jurisdictional Overlap of Generalists and Specialists
Generalist professionals are usually in coordination roles, because generalists can claim
competence to oversee a variety of specialized work within an organization due to their broader
knowledge basis (Burkhardt, Erbsen, & Rüdiger-Stürchler, 2010; Lawrence & Lorsch, 1967b;
Puranam et al., 2012). Generalists can also assume linking roles that serve as a lateral means of
communication (Galbraith, 1977). Given these attributes, generalist professionals often function
as managers or executives, since they usually do not have the depth of expertise to perform
highly technical work, yet they possess the expanse of knowledge to holistically view the
organization (Freidson, 2001).
The generalist role evolves with more task complexity, because the greater the
uncertainty of the task, the greater the interdependencies and the need to process and share
information (Galbraith & Kazanjian, 1986). Arguably, one of the most influential roles a
generalist can have in an organization is to coordinate the work of other specialists, since without
effective coordination, specialist expertise cannot be successfully leveraged (Malone &
29

Crowston, 1994). In their coordination role, a large part of the generalist’s function is the
exchange of information between interdependent and specialized organizational units
(Nambisan, 2002). To be effective at coordination, generalists must possess a basic level of
knowledge corresponding to the various specialty units they coordinate, because they must
demonstrate enough expertise to establish their legitimacy (Lawrence and Lorsch, 1967b).
Further, “common ground” —or shared knowledge—exists to ease the burden of coordination
(Clark, 1996). Over time, therefore, generalists can begin to specialize as they gain more
experience and develop efficiencies in handling particular work in an organization (Gupta, 1984;
Kotter, 1982).
As generalists specialize, the practices of specialist colleagues can be affected.
Specialized professionals bring in a “custom identity” into their work setting, which is usually
formulated through their formal education and training, and eventually these identities are
compartmentalized in the organization into specialty departments (Martin, Currie, & Finn, 2009).
However, such identities are altered in the daily practice of work, as values and norms of a
profession are transferred from macro-level legal and public contexts to the micro-level
workplace (Wright, Zammuto, & Liesch, 2015). In the workplace, collaborations with generalists
possessing specialized expertise can be one such factor that alters the identities of specialists in
an organization.
With the pervasiveness of complexity in modern organizational environments, it is
increasingly commonplace for generalist professionals to function with more technical
knowledge (Langbert, 2005). El Sawy et al (2015) describe the example of the LEGO Group, a
company which took a novel approach to managing digitization. Rather than having one Chief
Digital Officer with a general knowledge of the dynamics of digitization, the LEGO Group
30

instead implemented numerous digital offers across different business units in the organization.
These digital officers possess more in-depth knowledge of the workings of their particular
business unit, all the while maintaining an understanding of the holistic digital environment, thus
shifting generalist expertise further into the domain of the specialist.
In a similar vein, Cappetta and Cillo (2008) study of brand managers in the fashion
industry shows how breadth and depth of knowledge are two key attributes of generalist
coordination roles. In order to effectively direct the flows of knowledge between different
specialty groups in an organization, as well as merge aesthetic objectives with economic ones,
brand managers demonstrate expertise in both product development and marketing.
Correspondingly, brand managers are specialized in relation to the particular integration task
they must work on, as well as the individuals they must deal with. Brand managers therefore
provide expertise which is then integrated into the existing functions of the organization, and can
further blur work boundaries.
While the literature develops abstract conceptions of what generalists do relative to
specialists and vice versa, we lack an in depth understanding of the way these professionals
collaboratively work together in practice. In particular: How do professionals with overlapping
expertise negotiate their jurisdictional domains in the daily practice of work? This research aims
to flesh out the understudied phenomenon of jurisdictional overlap by examining how generalist
and specialist professionals define their unique domains of expertise in day-to-day practice. In
studying this broader issue, three questions emerged:
Research Question 1: What are coordination challenges in collaborative professional
environments that lead to conflict over work boundaries?
31

Research Question 2: How do professionals with overlapping expertise identify work
boundaries?
Research Question 3: How do professionals with overlapping expertise resolve conflicts
over work boundaries?
METHODS
Research Setting
To explore the negotiation of overlapping jurisdictions, I focused on generalist and
specialist professionals in the medical field. This field is particularly appropriate, since in the last
few decades, it has seen a surge in complexity, and thus presents a fitting context to study the
challenges of increasing specialization. The volume of medical specialists exceeds the
generalized patient population, and is therefore causing an “unregulated” division of labor in
which specialist and generalist work starts to resemble each other (Caronna & Ong, 2011; Fryer,
1991). Previously, generalist and specialist physicians had starkly different functions, and so
their paths rarely crossed; however, more recently, there are numerous points of contact between
generalists and specialists that implies a greater potential for conflict (Fryer, 1991).
In this setting, we have also witnessed the emergence of a novel generalist profession, the
hospitalist, which originated from pressures on U.S. hospitals to increase quality and reduce the
cost of patient care. Hospitalists first emerged in the mid-1990s, and the role was primarily
created to bridge the gap between general practitioners in the outpatient setting (e.g. private
practice clinics) and specialists in the inpatient setting (hospitals), thereby improving the
continuity of patient care. However, a major aspect of such a “bridging” role is the overlap of the
hospitalists’ skillsets with that of other specialists in the hospital, particularly as they become
more effective in their coordination function. Thus, for this study, my field site was intentionally
32

selected as an extreme case by which to examine jurisdictional negotiation, for three main
reasons: (1) the hospital setting represents a highly complex, multidisciplinary workplace; (2) the
hospitalist profession is relatively new, and the role is still establishing its jurisdiction relative to
other professionals, thereby making issues around jurisdiction more explicit; and (3) hospitalists
and specialists possess frequently overlapping expertise, making conflict and negotiation
practices more salient.
I conducted an investigation at Metropolitan Children’s Hospital
2
(MCH), a U.S.
academic hospital. At MCH, the hospitalists work in the Hospital Medicine (HM) division,
which was officially established in 2009. Presently, there are about 50 active hospitalists at
MCH, which parallels the rapid growth of hospitalists nationwide. The HM division serves
nearly 15,000 children per year. Hospitalists coordinate patient care, make diagnosis and triage
decisions, and manage many other aspects of care once a patient is admitted. These functions
require hospitalists to be in the hospital twenty-four hours a day, seven days a week.
At MCH, work boundaries are established as hospitalists or specialists are assigned to
primary and consultant roles in patient care. In the inpatient setting, sometimes a hospitalist is
the primary attending
3
, making diagnosis and treatment decisions on a patient, while a specialist
is the consultant. Other times, a hospitalist is the consultant, and the specialist is the one in
charge as the primary attending. While hospitalists are given the informal authority to make an
initial assignment of a hospitalist or specialist as the primary attending, physicians can challenge
initial assignments as they see necessary.
I conducted two studies in this research. Study 1 reflects ethnographic work I performed
at MCH for the purpose of discovering my research questions and deriving theoretical concepts.

2
Pseudonyms were used for my field site and all informants.
3
“Attendings” are licensed, board-certified physicians, and are considered authorized practitioners of medicine.
33

Study 2 builds on the themes formulated in Study 1 to add richer support and refine the
conceptual model. The following details data collection and analyses for each study.
Study 1: Ethnography of Hospitalist and Specialist Physicians
I started my ethnographic work with the intention of better understanding physician
decision-making in the context of competing frames of efficiency and quality. However, as is the
nature of ethnographic research, a different theoretical question emerged during my observation
period. Over the course of nearly two years, I spent over 100 hours in the field observing
physicians in the HM division and in various specialty divisions at MCH. My observations
consisted of (i) weekly HM division “leadership meetings” (hour-and-a-half meetings comprised
of seven hospitalists and two division administrative members); (ii) three to five hour sessions
observing inpatient/outpatient rounds
4
among hospitalists and specialists; (iii) bi-monthly HM
division meetings; and (iv) annual two-day “HM division retreats” in which all hospitalists at
MCH were in attendance. Weekly leadership meetings took place in the HM division office’s
break room, and largely focused on operational challenges faced by the division. With regard to
the bi-monthly HM division meetings, the focus was on morbidity and mortality case
discussions, as well as administrative and practice updates. My role during observation was
“student observer,” and I was introduced as such to other medical professionals and patients,
without further explanation unless asked. While rounding, I was treated like a medical student
shadowing the attending physician. I chose these settings, mainly because they demonstrate
different contexts in which the physicians were to make decisions about themselves and other

4
Rounds are bedside visits by a physician (or group of physicians and/or other healthcare professionals) to make
treatment evaluations, assess the current course, and document the progress of a patient. Rounds can also be
teaching-focused meetings in which physicians educate residents about various medical problems. Both forms of
rounding were a part of this ethnographic study.
34

stakeholders, as well as have interactions with each other in the practice of their work.
Observations of inpatient rounds have taken place in the following specialty units:
Cardiovascular Acute (CV Acute), Pulmonology, Adolescent Medicine, Allergy-Immunology,
Neurology, Hematology-Oncology, Metabolic Diseases, and General Pediatrics (outpatient
clinic).
My ethnographic data is comprised of non-participant observation. In developing
grounded theory, I used the data as an inductive, interpretive case study to address my emerging
research questions (Glaser & Strauss, 1967; Strauss & Corbin, 1994). During the process of
collecting observational data, I employed the Gioia method (Gioia et al., 2012), which involves
multiple iterations of reading and coding notes, referring back to literature and existing theory,
refining coded concepts, and eventually building theory by repeating this process. The following
details these steps.
Step 1: Identifying empirical concepts.
While observing the hospitalists, as described above, I compiled copious notes of the
observations I made during my visits to the field site—whether or not they had relevance to
jurisdictional boundaries or any prior research interests. Upon completing each observational
session at the field site, I immediately transcribed my notes and proceeded to code them with any
concepts that seemed to emerge from my observations, so as to develop “first order” concepts.
These notes were comprised largely of detailed observations of the interactions among
hospitalists, as well as between hospitalists and other specialists.
“Open coding,” which generated these first order concepts, proceeded throughout the
course of my analysis, and included such example codes as “unit-based staffing model,” “focus
on patient condition,” “new role creation,” and “breakdown in work boundaries.” In order to help
35

me determine the importance of these codes, I created tables that documented the number of
times each of these concepts appeared in my setting and in which situations. However, since this
means of ranking is not sufficient in deriving a theoretical framework, I proceeded to Step 2.
Step 2: Deriving emergent themes.
As I observed patterns in the coded concepts, I went back to the literature to understand
what sort of theoretical implications may be evident—repeatedly asking myself “what is this a
case of?” (Ragin, 1994). Following each consultation of the literature, I would return to the field
to see how coded concepts and theoretical implications surfaced in different settings the
physicians encountered. I would then code the new notes and recode the older ones, and through
a process called “axial coding,” I derived higher order themes. These axial codes were the result
of the constant comparison of empirical concepts to each other and to the emergent themes in
this phase of analysis (Glaser and Strauss, 1967).
Step 3: Developing a theoretical framework.
After identifying higher order themes, I returned to the literature again to further refine
the themes and concepts, so that I could generate “aggregate dimensions” representing the
dynamic relationships between second order concepts. All coding is documented in ATLAS.ti
software, which aided in the organization and retrieval of data. Through multiple iterations of
open and axial coding, five types of boundary identification factors emerged, in addition to three
types of boundary management practices. A variety of examples of how hospitalists and
specialists experienced conflict over work boundaries also emerged from my field observations.
During the course of open and axial coding, I made comparisons of codes across different
contexts (e.g. leadership meetings versus rounds) and individuals (e.g. more experienced versus
less experienced physicians) to check the consistency.
36

After multiple rounds of re-coding, eventually no new themes emerged, which gave me
assurance that I had reached the point of theoretical saturation (Glaser and Strauss, 1967). In
order to ensure the consistency and dependability of my analysis, I frequently returned to my
field notes to analyze them again with the objective of making sense of my findings. I also
returned to the field site numerous other times after I approached theoretical saturation to ensure
that I was not overlooking or misinterpreting the key themes.
Study 2: Survey of Hospitalist and Specialist Physicians
Study 1 provided tremendous insight into the perspectives of the hospitalists; however,
the specialists’ perspective was comparatively lacking. Thus, Study 2 built on the findings of the
ethnographic data to hone the perspectives of both specialists and hospitalists. Based on the
emerging theory from my observational data, I developed a series of open-ended questions that
were to be the basis of an online survey I administered through Qualtrics software to 17
physicians (13 specialists and four hospitalists). The specialist respondents represented six
specialties at MCH: Cardiology, Neurology, Pulmonology, Hematology-Oncology,
Gastroenterology, and Endocrinology. Survey participants: (1) worked at MCH for at least two
years; (2) were full-time; and (3) could be assigned primary and consultant roles. When
combined with other methods, an open-ended survey approach assesses generalizability (Detert
& Edmondson, 2011). Additionally, open-ended survey methods allow the researcher to explore
different aspects of observational data, as well as increase the likelihood for more transparent
participant responses, since participants can better maintain anonymity (Raffaelli & Glynn,
2014).
The survey began with basic questions about the physicians’ background, including the
nature of their current role, level of experience, and prior training/education. In order to further
37

investigate boundary negotiation processes and challenges, the survey also contained several
open-ended questions about how physicians are assigned, challenges related to assignment, and
possible suggestions for enhancing the assignment process. All survey responses were uploaded
into Atlas.ti software, and I proceeded to follow Steps 2 and 3, as outlined in Study 1. Step 1 was
not necessary at this stage of the research, since Study 1 already defined emergent themes that
would be the focus of this investigation.
FINDINGS
Throughout my time in the field and in analyzing the survey results, I found that as the
physicians faced contradictions in their work, they relied on intrapersonal and interpersonal
negotiation to establish their jurisdictions. Intrapersonal negotiation occurred when individual
physicians identified work boundaries through various factors, and assessed whether these
factors aligned with their own perception of task assignment. Interpersonal negotiation occurred
as physicians managed their work boundaries co-constructively, both within their unit and across
units in the hospital. My observational and survey data provided the premise of a model that
predicts the boundary work process for jurisdictional negotiation once tasks are initially assigned
until assignment is complete, which is illustrated in Figure 3.
First, tasks are initially assigned
5
to either a generalist or specialist professional. The
assigned individual engages in intrapersonal negotiation as they evaluate each of the five
boundary identification factors to determine whether or not the task is in their jurisdiction. Based
on this evaluation, if there are no contradictions, the professional follows the pathway on the top

5
The process of initial assignment may be idiosyncratic to each organization. At MCH, hospitalists were given the
informal authority to suggest the initial assignment (primary attending versus consultant) of a hospitalist or
specialist, but those assigned may challenge the decision. Based on what is known of generalists serving a
coordination function, it is reasonable to conclude that such authority to recommend initial assignments most often
rests with the generalist.
38

of Figure 1, and they begin to work on the task. However, a contradiction in work boundaries
may present itself during the course of the professional’s work, which may again instigate
intrapersonal negotiation. From there, the professional may determine that they remain in charge,
or approach their colleagues to resolve any work boundary issues through boundary management
practices comprising interpersonal negotiation. Professionals may oscillate between interpersonal
and intrapersonal negotiation to continually assess boundary identification factors and work
through assignment conflicts. As the negotiation process unfolds, either the professional initially
assigned will remain in charge, or they may determine that another professional should in fact be
in charge.
If after the initial assignment, however, the assigned professional does not believe the
work should be in their domain based on intrapersonal negotiation, they will follow the bottom
pathway in Figure 1. In this case, the assigned professional will communicate their rationale for
challenging their assignment based on the boundary identification factors. The assigned
professional may remain in charge if they are convinced otherwise through interpersonal
negotiation, or another professional will be designated as more appropriate and reassignment of
work will occur. If reassignment occurs from a generalist to a specialist, or vice versa, this may
set a precedent for future assignments, thereby clarifying and potentially altering jurisdictional
domains between generalists and specialists.
To detail each component of this model, the remainder of this section will progress as
follows. First, I will describe common coordination challenges, or contradictions, identified in
my field observations and survey data. Next, I will describe intrapersonal negotiation, in
39

FIGURE 3 - Boundary Work of Jurisdictional Negotiation

40

particular as professionals identify work boundaries in their context. Professionals identified
work boundaries based primarily on five factors: knowledge and expertise requirements,
coordination requirements, degree of socialization, resource demands, and organizational norms.
While intrapersonal negotiation serves to inform interpersonal negotiation, the reverse is also
true, since individual professionals may need to adjust perceptions of how they identify work
boundaries after discussion with their colleagues. To conclude this section, I will explain the
interpersonal negotiation of jurisdictions, which occurs through boundary management practices
that hospitalists and specialists used when they encountered conflict in work boundaries.
Research Question 1: Contradictions
Although formal coordination mechanisms can exist in organizations to manage
collaborative work, such practices may vary based on work conditions and context (Crowston,
1997). As I analyzed my data, I came across several challenges to the successful coordination of
work, which I term contradictions. Contradictions lead to conflict over work boundaries, since
they obfuscate the relationship between professional jurisdiction and the practice of work. Table
2 summarizes these contradictions, which include: exceptions, interdependence issues,
information ambiguity, misaligned professional identity, and paradoxical demands.
Exceptions
One way contradictions occur is through exceptions, specifically the presence of novel
situations or problems (Perrow, 1967). Exceptions induce uncertainty and require a non-routine
response. At MCH, exceptions occurred when international patients were admitted into the
hospital:
Hospitalist Su talked about the International Child Health Program. “We need someone
in charge. [The specialists] don’t know what they’re doing,” she said, “All these patients
should come through us first because we end up managing them anyway.” Hospitalist Su
41

talked about how the doctors transferring these patients ask who the attending will be,
and each time, there is a “pass-off” of responsibility when hospitalists ask who will be the
patients’ attending and it’s frequently “not me, not me.” Hospitalist Su said, “There’s a
lot of things [the specialists] just don’t know about inpatient consultation.” Hospitalist
Crane added, “We just need to know ahead of time and coordinate.” (Leadership
Meeting, Nov 26 2012)
6

Hospitalist Su explains that there is not a typical protocol in handling the unique international
patient population, and that specialist colleagues are less willing absorb such patients into their
domain of practice. International patients typically require a different set of administrative needs
during their treatment at the hospital, and often, the process of being reimbursed for services is
more complicated. Pulmonologist Stockton confirms his confusion in stating that there is
“difficulty in planning and arranging for the timely discharge of international patients.”
Exceptions induce confusion, and professionals struggle with assimilating such work into their
domains, especially if payoffs do not outweigh the time needed to understand how to deal with
these novelties.
Interdependence Issues
Interdependence between professionals in a work system can also lead to conflict over
work boundaries. Interdependence refers to the intensity of interactions between different
professionals required to effectively complete a task (J. D. Thompson, 1967). Issues with
carrying out interdependent work usually stem from breakdowns in communication between
professionals or a lack of timely action (Lawrence & Lorsch, 1967a). When work boundaries
overlap, interdependence issues can become more commonplace, as professionals may debate
who should actually carry out a task, delaying completion of work or performing duplicative
work. These issues may result in errors and confusion over who is or should be in charge.

6
Parenthetical reference contains observation context and date of observation.
42

Hospitalist Crane mentioned that the number of calls for insulin corrections is too high.
Answering these calls every time they come up (several times throughout the day for the
same patient), is not safe and challenges productivity. Whenever a call happens,
Hospitalist Crane said he needs to interrupt his workflow to put an order in because
patients can’t eat until the insulin dosage is corrected. “Maybe we should bring this up
with [the endocrinologists],” Hospitalist Crane went on to suggest, “they should take the
calls themselves, not us.” (Leadership Meeting, Feb 11 2013)

Insulin corrections involve matching diabetic patients’ insulin dosage to the food they consume,
and thus, corrections must occur in a timely way throughout the day. Insulin corrections can be
performed by both hospitalist and endocrinologist physicians. However, calling on the hospitalist
to do these corrections when a patient may be primarily under the care of an endocrinologist
disrupts the hospitalist’s workflow. Interdependence issues in this example cause Hospitalist
Crane to rethink who actually should be in charge of insulin corrections, when both hospitalists
and endocrinologists have the expertise to handle this work.
Information Ambiguity
Other times, contradictions emerge from ambiguity in available information. Ambiguity
can result from high levels of uncertainty, as well as the possibility of misinterpretation in
collecting and processing information in order to perform certain work (Daft & Lengel, 1986). In
the following quotation, Endocrinologist Bando discusses the factors that determine physician
assignment:
“I think [physician assignment] depends on complexity of situations. For instance, a
patient with a brain tumor may then develop diabetes insipidus with concurrent
chemotherapy. In terms of current proposed treatment, this patient would become a
primary assignment to endocrinology if the diabetes insipidus is the limiting factor for
discharge. However, the complexity of the patient, particularly in determining other
aspects of oncologic care would make me uncomfortable to be a primary assignment.
Another instance would be a Type 1 diabetes patient with appendicitis. This may be a
good patient to go under a hospitalist system who can coordinate care with surgical
specialists and endocrine specialists.”

43

Endocrinologist Bando’s discomfort in handling a patient with multiple diagnoses spanning
different specialists’ domains reflects the ambiguity involved in treating such patients with
complex conditions. In other words, multiple diagnoses yield greater ambiguity in processing
information about treatment, making a certain physician’s authority over a complex patient
unclear.
Misaligned Professional Identity
Another main source of contradiction is misalignment between one’s professional
identity and work conditions. Professional identity is defined as “an individual’s self-definition
as a member of a profession and is associated with the enactment of a professional role”
(Chreim, Williams, & Hinings, 2007; Ibarra, 1999). At MCH, misalignments in professional
identity occurred pronouncedly as physicians considered the assignment of a hospitalist’s role
relative to the specialist’s role. As Gastroenterologist Jade described:
“It would be helpful to first understand what the current process for assignment is and
what the goals for assignment are. Is the goal to maximize patient satisfaction? Patient
care? Teaching for residents? Facilitate consistency in care for complex patients? My
current dissatisfaction arises from not knowing what the goals and expectations are and
thus not being able to adapt. These also do not seem to be consistent among all the
hospitalist faculty that I rotate with.”

Because the objectives of assignment were unclear, Gastroenterologist Jade was unsure how to
reconcile self-perceptions of her role with the inconsistent practices of her colleagues. In another
example:
Other problems with the pulmonologists were discussed. “Too often they give conflicting
impressions to patients of what [hospitalists] do. They don’t collaborate with us well.
And the fellows tell us what to do—which may be warranted because they know more
about their specialty, but the way they approach things is not good,” Hospitalist Kacik
said. Hospitalist Myers spoke up, “Maybe we should just be a consulting service for the
pulmonologists. We shouldn’t put our license at risk if they tell us to input orders [in the
EMR] on patients but ultimately are the ones making the decisions. They know how to
put in orders!” (HM Division Retreat, Nov 02 2012)
44

TABLE 2 - Contradictions Leading to Work Boundary Conflicts

Contradiction Description Examples
Exceptions Non-routine, novel or unusual
problems / situations that are
associated with a task (Perrow,
1967)
Hospitalist Su talked about the International Child Health
Program. “We need someone in charge. [The specialists] don’t
know what they’re doing,” she said, “All these patients should
come through us first because we end up managing them
anyway.” Hospitalist Su talked about how the doctors
transferring these patients ask who the attending will be, and
each time, there is a “pass-off” of responsibility when hospitalists
ask who will be the patients’ attending and it’s frequently “not
me, not me.” Hospitalist Su said, “There’s a lot of things [the
specialists] just don’t know about inpatient consultation.”
Hospitalist Crane added, “We just need to know ahead of time
and coordinate.” (Leadership Meeting, Nov 26 2012)
Interdependence
issues
Breakdown in the interactions and
interconnected practices of
professionals completing a task
(Thompson, 1967)
Hospitalist Crane mentioned that the number of calls for insulin
corrections is too high. Answering these calls every time they
come up (several times throughout the day for the same patient),
is not safe and challenges productivity. Whenever a call happens,
Hospitalist Crane said he needs to interrupt his workflow to put
an order in because patients can’t eat until the insulin dosage is
corrected. “Maybe we should bring this up with [the
endocrinologists],” Hospitalist Crane went on to suggest, “they
should take the calls themselves, not us.” (Leadership Meeting,
Feb 11 2013)
Information
ambiguity
Absence of information in
completing a task, or
misinterpretation of existing
information (Daft & Lengel, 1986)
“I think [physician assignment] depends on complexity of
situations. For instance, a patient with a brain tumor may then
develop diabetes insipidus with concurrent chemotherapy. In
terms of current proposed treatment, this patient would become a
primary assignment to endocrinology if the diabetes insipidus is
the limiting factor for discharge. However, the complexity of the
patient, particularly in determining other aspects of oncologic
care would make me uncomfortable to be a primary assignment.
Another instance would be a Type 1 diabetes patient with
appendicitis. This may be a good patient to go under a hospitalist
system who can coordinate care with surgical specialists and
endocrine specialists.” (Endocrinologist Bando)
Misaligned
professional identity
Aspects of a task which do not align
with one's self-perception of their
professional role (Ibarra, 1999)
“It would be helpful to first understand what the current process
for assignment is and what the goals for assignment are. Is the
goal to maximize patient satisfaction? Patient care? Teaching for
residents? Facilitate consistency in care for complex patients?
My current dissatisfaction arises from not knowing what the
goals and expectations are and thus not being able to adapt.
These also do not seem to be consistent among all the hospitalist
faculty that I rotate with.” (Gastroenterologist Jade)
Paradoxical demands Conflicting demands that must be
managed in order to effectively
complete a task (e.g. balancing
quality and cost)
“The priorities seen by the hospitalist and specialist can be
different. For instance, there may be pressure for better patient
[throughput] but from a specialist standpoint, the most important
issue is not having the patient admitted again.” (Pulmonologist
Roberts)
45

For Hospitalist Kacik, her professional identity does not coincide with pulmonologist fellows
telling her what to do, and the other pulmonologists not fostering an environment in which the
hospitalists’ input is respected. Such interactions with pulmonologist colleagues create tension in
hospitalists’ perception of their role, thereby inducing conflict over work boundaries.
Paradoxical Demands
Last, contradictions emerged as physicians dealt with balancing conflicting demands in
their work context. For physicians, the issue of managing both the quality and efficiency of care
may be at odds. For example, the desire to keep healthcare costs low may ultimately sacrifice the
quality of care provided to patients. Paradoxical demands typically surfaced when professionals
considered the tradeoff between organizational- and client-level concerns. As Pulmonologist
Roberts describes:
“The priorities seen by the hospitalist and specialist can be different. For instance, there
may be pressure for better patient [throughput] but from a specialist standpoint, the most
important issue is not having the patient admitted again.”

Throughput is the number of patients that move through the hospital system over a period of
time—counting the time it takes to assign a patient to a bed up to when the patient is discharged
from the hospital. Improved throughput is a sign of greater efficiency, and Pulmonologist
Roberts communicates the tension here that from her perspective, she would prefer as long a
length of stay as necessary to minimize the risk of a patient’s readmission for the same problem.
Throughput should be a concern for all physicians, since improving that measure could possibly
indicate shorter patient wait times and the ability to treat more patients as those no longer
requiring care are discharged more expediently. However, balancing conflicting demands makes
it more challenging to determine who should be in charge when it comes time to determine
whether a patient should be discharged from the hospital, for example.
46

Table 1 summarizes contradictions that lead to work boundary conflicts. When these
contradictions occur, either in the initial assignment of work or during the time work is carried
out, intrapersonal and interpersonal negotiation of jurisdictions take place to reconcile conflict.
The rest of my findings detail aspects of intrapersonal and interpersonal negotiation, which are
summarized in Table 3.
Research Question 2: Boundary Identification Factors
Intrapersonal negotiation reflects the cognitive activity of individual professionals as they
think about the factors that would distinguish one type of work boundary from another.
Hospitalists and specialists consistently relied on five factors in identifying their jurisdictions:
knowledge and expertise requirements, coordination requirements, degree of socialization,
resource demands, and organizational norms.
Knowledge and Expertise Requirements
Knowledge and expertise requirements are considered in work that necessitates a
distinctive skillset and experience level. Many functions of the professional rely on years of
education and training in a certain area, thereby serving as automatic demarcations in their future
practice. Unique skillsets and knowledge areas often serve as a way to parcel work. For example,
a Hospitalist Morst determined whether to be primary or consultant for a patient depending on
the experience necessary to complete a clearly defined task:
“Well [being primary or not] varies case-by-case. Almost always Cardiology kids have
hospitalists as the primary and the [Cardiology] doctors as consultants, except when
surgery is involved," Hospitalist Morst told me as we walked to the next patient's room.
(Inpatient Rounds in Overflow Unit, Mar 11 2013)

47

While Hospitalist Morst highlighted that variation exists on a case-by-case basis, the skills
required to conduct heart surgery, as the example above mentions, inherently makes the
specialist the primary physician on a case.
When Pulmonologist Roberts was asked how she distinguishes between her specialist
role and the hospitalist’s role, she emphasized the selectivity of the specialist’s know-how and
the focus on a particular organ system:
“Hospitalists provide the general pediatric care and the specialists provide the specific
knowledge regarding the specific organ system that they specialize in, either in a
consultative role or in direct collaboration if the organ system problem is rare and/or
complicated.”

Similar to Pulmonologist Roberts, Hospitalist Morst also explained that the unusualness
of a patient’s ailment can delineate who’s in charge. Hospitalist Morst made a point about
patients with atypical and complex diagnoses:
"...there may be some really atypical cases where the specialist should be the primary, but
otherwise, hospitalist should be primary and specialist should be consultant." (Inpatient
Rounds in Overflow Unit, Mar 11 2013)

Coordination Requirements
Professionals also assess the variety of different problems associated with effectively
handling a task. Specifically, a particular task may involve a diversity of issues that require
different types of expertise and multiple specialists. As more specialists are involved,
coordination requirements increase (Lawrence and Lorsch, 1967a).
For example, during inpatient rounds, Hospitalist Vai described Leonardo’s case, one of
the patients for whom she was primary physician, and highlighted why Leonardo’s condition is
under her jurisdiction:

48

TABLE 3 - Boundary Identification Factors and Boundary Management Practices

Elements of Negotiation Description Examples
Intrapersonal negotiation: Boundary identification factors
Knowledge and expertise
requirements
Type of knowledge, skillsets, and
experience required to handle a
particular task
“The hospitalist takes the view of the patient as a
whole person, including all body systems as well as
social and psychological concerns. [...] The specialist is
responsible for [...] lending their expertise in their
particular body system with attention to how their
medical conditions of interest may impact the patients'
other medical issues. In many cases the hospitalist is
consulted by a service to answer a particular clinical
question--this is common with consults by surgery and
surgical specialties. In such cases, the hospitalist's role
will more closely resemble the specialist's role..."
(Hospitalist Grossman)
Coordination requirements Diversity of knowledge and
expertise required to handle a task;
the greater the diversity, the more
types of specialists involved
"The hospitalists are helpful in coordinating patient
care when they manage complicated patients who need
[multiple specialists]. They manage general pediatrics
patients well, however, they do not have the training,
experience or support to manage [specific] specialty
complex patients." (Gastroenterologist Hoffman)
Degree of socialization Extent to which the professional
has assimilated into their work
environment, which may be a
product of tenure, frequency /
quality of interactions with
colleagues, etc.
“The combination of care provided by both a primary
inpatient care team [hospitalist] and consultant
[specialty] service can be very mutually beneficial to
the care of the patient. However, the hospitalist has to
recognize their own limitations when it comes to
specialty care and the specialist needs to trust the
hospitalists’ judgement regarding general
care/management of the patient.” (Cardiologist
Jackson)
Resource demands The amount and variety of
resources required to complete
work (e.g. workload
considerations, ancillary support)
The role of hospitalists includes "keeping the
continuity of care during the hospitalization, ordering
appropriate tests and getting the results, and addressing
social needs by mobilizing appropriate resources…"
(Pulmonologist Stockton)
Organizational norms The protocols that may exist to
guide how certain tasks should be
conducted and may either be
mandated from the top-down or
informally agreed upon between
professionals
Physician assignment depends on the following: "1.
Primary reason for admission; 2. If reason for
admission is related to an underlying known diagnosis
or suggests a possible diagnosis within the domain of a
particular specialty; 3. Co-morbidities and the
magnitude of influence (e.g. acuity) or complexity of
care related to the co-morbidities; 4. MCH policy or
culture related to the above factors..." (Hospitalist
Raavi)
49

TABLE 3 - Continued

Elements of Negotiation Description Examples
Interpersonal negotiation: Boundary management practices
Modifying the in-group A unit re-organizes itself to address
recurring contradictions in task
assignments typically, it is the less-
established group or the one
responsible for coordinating
workflow, that will engage this
process (i.e. in this context, the
hospitalist)
"I think liver transplant and intestinal rehab should be
moved to different hospitalist services as both are
complicated patients and rounding times tend to
overlap." (Gastroenterologist Tanalia)
Clarifying guidelines Rules and guidelines are sought to
clarify work boundaries in order to
create stability and predictability in
workflow, as well as to minimize
future jurisdictional contests
"Some issues we face when working with consultants
are disagreement on management decisions,
disagreement on length of treatment or certain
therapies, or feeling that the management decisions are
one-sided or being dictated by that specialist. […] As it
exists, we have certain diagnoses whose assignment
has been agreed upon between specialists and
hospitalists. Expanding that list and discussing
appropriate assignment seems like it would be the best
way to assure effective assignment at MCH."
(Hospitalist Yakira)
Managing the out-group Work boundaries are managed
through communication with those
outside of one's professional group,
largely to voice concerns with
assignments and address violations
of jurisdictions, as well as affirm
professional identify amongst
colleagues outside of the group
"If there are pre-determined guidelines in terms of how
physician assignments are done, this would be a benefit
to how clinical care would work efficiently...
[Establishing guidelines would be] a difficult task to
accomplish, as it would be a difficult task to achieve a
sort of universal consensus. Perhaps the best method is
to have guidelines be guidelines, but to have open
communication if there are questions or concerns
regarding physician assignment."
(Endocrinologist Bando)

Leonardo’s heart issues cause him to have breathing problems when he is not at rest;
since the heart does not pump blood efficiently, and when activity increases, it is forced
to beat faster but blood starts to back up into lungs when the heart cannot keep up,
causing trouble breathing. To determine if and when to schedule surgery, Hospitalist Vai
stated several things needed to be done, including consult with Surgery and also with
Genetics (since Leonardo’s parents are first cousins, and the potential for complications
may be elevated). In addition, Leonardo’s family came from Mexico, it appears just for
surgery. So there is a need to consult with a social worker to file a consent form in
addition to waiting on responses from Genetics, Surgery and Audiology (about his ear
condition). (Inpatient Rounds in Cardiovascular Acute Unit, Feb 04 2013)

Here, Hospitalist Vai pointed out significant factors affecting Leonardo’s health, and as his
primary physician, she explained one of her main functions is to coordinate his care among the
50

other specialists as well as with his social worker. Leonardo’s condition had not yet escalated (in
terms of requiring surgery), and so Hospitalist Vai still acts as primary until her discussions
about Leonardo’s case with the other consulting specialists indicate otherwise.
Specialists also reflect on coordination requirements as reason to invoke the hospitalist
role. For example, Cardiologist Groff explained that “the hospitalist is the glue that holds the
team together,” indicating that the hospitalist functions as a linking role on multispecialty teams.
Endocrinologist Justman agreed that “the specialists are able to provide care to patients with a
particular medical issue, and hospitalists coordinate care in more complex patients [those
requiring multiple specialists].”
Neurologist Walsh elaborates on her thought process for delineating her jurisdiction from
that of hospitalists, in particular as it relates to the variety of conditions afflicting a patient:
“Most children with solely neurological issues should be on a Neurology/primary service.
For example, new onset seizure management in an otherwise medically stable child or
new onset weakness (e.g. myelitis). Most children with ongoing neurological problems
but significant intercurrent illnesses requiring admission should go to the hospitalist, with
Neurology comanaging or consulting. For example, a child with cerebral palsy and
epilepsy who develops pneumonia, a urinary tract infection, or other acute illness, with
seizure exacerbation related to acute illness, should go to the hospitalist.”

According to Neurologist Walsh, a patient diagnosed with cerebral palsy and epilepsy (both
neurological disorders), and also presenting with other acute illnesses involving different body
systems and potentially different specialists, would belong in the hospitalist’s domain. The
rationale is that hospitalists’ broader expertise enables them to coordinate care and manage
treatment of complex patients.
Degree of Socialization
The degree of socialization hinges on such factors as the number of years a professional
has worked in a particular organization relative to others and the level of trust developed
51

between colleagues as they interact on a regular basis. The degree of socialization should not
counter one’s professional identity, meaning that socialization into an environment should not be
at the expense of diminishing one’s view of their role and personal career motivations. For
example, when I asked Hospitalist Morst, a young physician with about three years of experience
at MCH, whether she preferred to be primary attending or consultant, she replied “I prefer
primary. It’s what everyone wants—to be the one making the decisions” (Inpatient Rounds in
Overflow Unit, Mar 11 2013). Hospitalist Morst’s response resonates with a feature embedded in
professions, namely autonomy (Adler et al., 2008; Audet, Doty, Shamasdin, & Schoenbaum,
2005). Of course professionals can vary in the degree of autonomy that they associate with their
identity. As a young physician with nearly three years of experience at MCH, Hospitalist
Morst’s response is one of a strong association with autonomy, possibly indicating her
motivation to gain as much experience as possible early in her career.
Other times, trust in a colleague’s expertise can diminish one’s pursuit of autonomy and
authority over work. Cardiologist Jackson affirmed that trust is an important component of
determining if certain patients can and should be in a hospitalist or specialist physician’s
jurisdiction:
“The combination of care provided by both a primary inpatient care team [hospitalist]
and consultant [specialty] service can be very mutually beneficial to the care of the
patient. However, the hospitalist has to recognize their own limitations when it comes to
specialty care and the specialist needs to trust the hospitalists’ judgement regarding
general care/management of the patient.”

Gastroenterologist Hoffman shared a similar view: “If I know the hospitalist and trust his/her
judgement, I feel comfortable and have no issues [with their primary role].” Trust can be
developed over time, as physicians interact and build rapport or as physicians gain a reputation
52

of being good at what they do. As trust increases, work boundaries can shift outward toward
another’s domain.
However, trust may be lacking in practice. In this example, Hospitalist Casabian, who
was completing a fellowship in the HM department, encountered an issue with an Infectious
Disease (ID) unit physician, as described by Hospitalist Jennson:
Hospitalist Myers commented, “Yeah they can’t be consultant and dictating plans. We
need to talk to Adolescent Care about this too.” Hospitalist Jennson said, “Yep, we just
all need to talk about it first, that’s all.” Hospitalist Jennson then went back to talk more
about the ID consult situation from the prior weekend, stating that Hospitalist Casabian
would bring the ID consulting physician evidence and she would have a “non-academic”
answer in response to her counter-suggestions. “If ID is consultant, then we can’t be
doing it like this,” Hospitalist Jennson said. (Leadership Meeting, Mar 18 2013)

The violation of the consultant role by the ID specialist spurs a discussion among hospitalists
that inevitably will affect future primary-consultant decisions involving work boundaries with
the ID unit. It may be the case that Hospitalist Casabian, who was working directly with the
primary attending Hospitalist Jennson on a patient in ID, is less socialized at MCH, because he is
still a fellow and has not yet developed the rapport to assert authority when a specialist is in a
consultant role. Hospitalist Jennson explained that she would not want to take on the role (and
liability) of being primary physician on a patient, if the ID specialist is going to end up dictating
plans without the authority to do so. Correspondingly, rapport and trust in colleagues factors into
the primary and consultant assignments.
Resource Demands
Resource demands mainly refer to workload conditions, organizational costs, and the
need for auxiliary support. Physicians manage their workload by assessing census and caps,
meaning the number of patients admitted into a unit at a given time and the number of patients
53

assigned to a certain physician per shift, respectively. When there is an imbalance in workload,
this may cause physicians to refuse patients:
This patient, John, is an 18-year-old boy who has a myriad of complications and complex
conditions that span multiple specialties. He is currently in the Adolescent Care unit, but
because of the nature of all his complications, the 5 East team thought it best to move him
under HM care. “Adolescent Care has three patients, and we’re maxed out. It only makes
sense to keep him in Adolescent Care,” Hospitalist Casabian said. (Inpatient Rounds in
Pulmonology Unit, Feb 18 2013)

Hospitalist Casabian made the point that the Adolescent Care unit’s request to have the
hospitalist be primary physician on a patient, John, was not a proper assessment because of the
unequal distribution of workload between the two groups. Hospitalist Casabian points out the
patient safety concerns that can result from imbalanced workloads. Workload considerations can
cause physicians to either absorb more work into their domain or not.
Concerns over workload persistently surfaced, mainly because this affected the quality of
care provided. The designation of primary attending puts the responsibility of the patient onto a
particular doctor, and therefore, if workload is too high, this poses safety issues in administering
care:
The concerns came up during the HM division’s annual retreat, as well, when the
discussion shifted to ENT and plastic surgery specialists, who seemed to continually and
without much consideration, request that hospitalists act as primary on every post-surgery
patient. Several hospitalists articulated that “the plastic surgeon’s demands were
unrealistic,” and they came to an agreement that it is “impractical” to have hospitalists
follow every surgery patient, especially from the standpoint of safety. (HM Division
Retreat, Nov 02 2012)

In this case, workload considerations affect work boundaries in the future, as hospitalists
determine that it is unrealistic to be in charge of every post-surgery patient, and therefore, these
patients should remain in the surgeon’s domain going forward.
54

Nevertheless, specialists associate the hospitalist’s role with organizational efficiency.
Cardiologist Jackson stated that “hospitalists are helpful. They manage inpatient care for all
types of patients and streamline throughput in the hospital.” In addition, hospitalists are
associated with managing resources, Pulmonologist Stockton explained:
“Hospitalist play a very important part in patient satisfaction by addressing their concerns
quickly as they arise and directing to appropriate resources as they are available […]
They address social needs by mobilizing appropriate resources […] and keep the
continuity of care during the [patient’s] hospitalization.”

In line with Cardiologist Jackson’s description, work that requires greater resource use and
management is more closely aligned with the hospitalist’s role, and resultantly, work boundaries
are constructed around the demand for resources.
Organizational Norms
Professionals uphold norms of practice that largely stem from regulated and certified
associations (Suddaby & Greenwood, 2001). Norms of practice may be adapted in the
organizational setting to account for context-specific factors (Criscuolo, Dahlander, Grohsjean,
& Salter, 2016). Such adaptations can lead to taken-for-granted practices, premised on
organizational protocol or policy that do not necessarily demonstrate a requisite rationale. For
example, different staffing models exist in organizations housed by professionals, and these
models serve to carve out norms around the division of labor. In the hospital setting, staffing
models can either be unit-based or specialty-based. At MCH, physicians are mostly staffed using
a unit-based model, meaning that each unit in the hospital contains patients with diagnoses
spanning a variety of specialties. This is in comparison to the specialty-based model, which
groups patients in each unit by diagnoses belonging to a certain specialty, and thus would mean
hospitalists assigned to patients in certain units would treat a more limited set of diagnoses.
55

During one of her rounds, Hospitalist Morst compared specialty-based staffing models to
unit-based models; essentially that the former uses certain unquestioned attributes of the patient
case to automatically designate a specialist as primary. Hospitalist Morst explained that in the
specialty-based staffing model, it is almost invariably clear which physician is to be primary and
which is to be consult. However, at MCH, under the unit-based staffing model, there is
variability in this decision, and as Hospitalist Morst states, this means “the primary-consultant
lines are blurred.” With the unit-based models, it is not solely the patient’s diagnosis that dictates
staffing, but also an assessment of the unit’s composition (i.e. number of patients) and physical
location. Even though there is an underlying norm guiding work distribution in the unit-based
model, the “lines are blurred” in terms of who gets staffed as primary or consult, as Hospitalist
Morst points out, because a unit’s patient composition can change dramatically moment-to-
moment.
A few divisions at MCH actually have staffing models that deviated from the norm. One
such specialty area is Liver and Intestinal Transplant. During a leadership meeting in which
Hospitalist Su described how she would divide up primary and consultant roles, “there was no
point in including [Liver and Intestinal Transplant] because they have their own system and
management” (Mar 18 2013). Thus, organizational norms of practice, such as staffing models,
can serve to delineate work boundaries and influence the way jurisdictions are negotiated in the
work setting.
Research Question 3: Boundary Management Practices
The five factors of boundary identification enable professionals to clarify assignment and
recognize potential areas of contradiction. These factors also serve to facilitate communication as
professionals interact with each other to manage work boundaries. Correspondingly,
56

intrapersonal negotiation undergirds interpersonal negotiation. Interpersonal negotiation is
represented by three boundary management practices: modifying the in-group, clarifying
guidelines, and managing the out-group.
Modifying the In-Group
As professionals in a work group discuss recurring contradictions, they modify their
group’s structure to either adapt to or remedy contradictions. Hospitalists engaged this process,
since they possess the informal authority to direct workflow and resolve physician assignment
conflicts. Both of my studies did not identify prominent occasions of specialist groups modifying
their structure; this boundary management practice was most salient among hospitalists. Two
conditions may be the reason for mainly observing hospitalists, rather than specialists, using this
boundary management practice. First, hospitalists are still a relatively new medical profession,
and so the organization of their role can be more malleable compared to their entrenched
specialist colleagues. Second, MCH gives hospitalists the informal authority to direct workflow,
and so they are primarily tasked with resolving any issues with physician assignment.
Modifying the in-group was observed when hospitalists discussed creating a position to
fill silos in workflow:
Hospitalist Su pointed out that chronic care patient cases which last more than 1 week
don’t get followed up on. She suggested that there be a chronic care team with a “mid-
level care person” who does the case management all along the way. “There are families
that come here and say ‘we’ve been here 6 months and have no idea who our doctor is.’
That’s unacceptable,” Hospitalist Su said. According to Hospitalist Su, the mid-level
person will always be the clear point person for patients with a longer length of stay.
(Leadership Meeting, Mar 18 2013)

Hospitalists generally have training and experience to care for long-term patients. Hospitalist
Su’s suggestion corresponded with knowledge and experience requirements that would put long-
term patients in the hospitalist’s jurisdiction and also addressed a silo in assignment practices.
57

Modifying the in-group also took the form of reinstating the function of a professional
group in an organization. Since hospitalists typically monitor a patient’s condition, write orders,
and work overnight, their tasks seemingly overlap with the work of residents, as Hospitalist
Crane pointed out at the division retreat:
Hospitalist Crane says that the hospitalists’ role writing orders is too conflated with what
residents used to do, and that has caused warped perceptions of the division, as well as a
lack of respect. He said that the HM division needs to rid of that feeling of subordination
because the reality is that the pediatrician who runs his own office has to do all the things
hospitalists have to do in the hospital. "That doesn’t mean we’re like residents,"
Hospitalist Crane concluded. (HM Division Retreat, Nov 02 2012)

Even when hospitalists are primary on a patient case, specialists ask them to do undesirable and
mundane tasks, such as finalizing documentation and paper work. Clarifying the group’s identity
in response to these types of work issues has the potential to alter boundaries such that
hospitalists may resist mundane tasks in the future.
To further delineate jurisdictions and clarify work boundaries, Hospitalists made
efforts to define and create value in their role relative to specialists:
“We need to embrace the educational service role. We can’t just do something because
the other units are doing it,” Hospitalist Myers said. Hospitalist Jennson said that she was
reminded of her experience at a hospital in DC, in which she was on for two weeks at a
time and frequently had 17 or so kids at once, but it was a great experience from an
educational point of view because it was “your team” and there was more opportunity to
teach. (Leadership Meeting, Mar 18 2012)

If hospitalists create a strong presence as educators in the hospital, it would enable them to take
charge of resident teams that would be working on a variety of patient cases as part of their
training. In doing so, the hospitalists defined an aspect of their jurisdiction—the educator role—
which would give them more authority over different types of work, and also be another point of
distinction between themselves and specialists.
58

Clarifying Guidance
Both hospitalists and specialists sought out ways to clarify rules and guidelines of
practice that would make their jurisdictions more stable and predictable. For the hospitalists,
standardization would not only make their jurisdictions more stable, but establish them in the
first place. Standardization would also aim to provide for timelier, appropriate assignments so as
to not sacrifice patient safety. For example, issues arose when surgical specialists asked
hospitalists to follow patients post-surgery, which could substantially increase the patient load on
hospitalists:
Hospitalist Su said that they just need to make simple and clear guidelines: if they had a
transplant, went home and then came back, they are HM patients; if they are transfers or
new patients, the patients belong to the house staff. Hospitalist Zipvale added that the
“Hospitalists shouldn’t be used to make the [residents] look better.” Hospitalist Myers
summarized, “Ok well if we want to have simple rules, that transfers and new patients go
to house staff and the rest go to us that’s fine. But we won’t be able to have a cap on
patients since we’re accepting all pre-existing cases.” Hospitalist Crane agreed and said
that the division would then be narrowing their patient population to more acute/intense
cases if this was the accepted rule. “Ok so if they’ve [the patient’s] been home, we take
them. It’s as simple as that,” Hospitalist Su said. Hospitalist Zipvale said that it will be a
different story once the Respiratory Care Unit opens in a year and a half. (Leadership
Meeting, Nov 12 2012)

Arriving at some sort of consensus of rules and clear guidelines is a practical
device the physicians used to make primary-consultant distinctions more objective and
streamlined. The HM division designates a different hospitalist each day to be the
“Doctor of the Day” (DoD), the physician responsible for parceling out work using a
system of rules based to some degree on the patient’s principal diagnosis. The DoD has
informal authority to make jurisdictional claims that are enacted through a semi-
standardized decision-making process, guided in part by the contents of a DoD binder.
However, many hospitalists stated that the “algorithms” (i.e. decision trees) contained
59

within the binder are extremely convoluted, and much of the time, the DoD uses his or
her discretion to make a decision that deviates from the primary-consultant rules in the
binder, as the following scenario demonstrated:
Hospitalist Casabian listened to what the DoD was saying over the phone; then he
commented: “Well there’s no absolute guidance so discretion is up to the DoD. And
frankly it’s a matter of politics. We don’t want Adolescent [the Adolescent Care unit] to
think they can’t handle this kid.” Hospitalist Casabian finished up by confirming that the
boy was to stay assigned to an Adolescent Care specialist, and if there were any questions
or a need for a consult, hospitalists are around to help. (Inpatient Rounds in Pulmonology
Unit, Feb 18 2013)

Although the rules could at times be discretionary, the importance of clarifying guidance
is illustrated in the following example. Endocrinologist Bando explains a particular issue in
which having clear rules about who is in charge would reduce miscommunications that occur in
situations where both endocrinologists and hospitalists are qualified to treat a patient:
“One issue may be determining medical care in the situation of care that can be done by
both hospitalist and endocrine physicians. A particular issue I had revolved around
recommendations about steroid tapers. Often times, this aspect of care is done by
multiple physicians, not only endocrinologists. However, this is one aspect of care when
an [endocrinologist] physician may make a recommendation, whereas the primary
attending and family agree upon a different recommendation. This makes the role of the
‘consulting’ physician awkward, particularly if their recommendations were not followed
[…] If there are pre-determined guidelines in terms of how physician assignments are
done, this would be a benefit to how clinical care would work efficiently […] The
‘guidelines’ are a difficult task to accomplish, as it would be a hard to achieve a sort of
universal consensus. Perhaps the best method is to have guidelines be guidelines, but to
have open communication if there are questions or concerns regarding physician
assignment.”

In complex organizational settings, the ability implement simple rules that delineate
jurisdictional domains, yet are still open to continual negotiation with the changing work
environment, is a key factor in managing work boundary conflicts in day-to-day practice.
Managing the Out-Group
60

In order to negotiate overlapping boundaries, managing the out-group (i.e. those outside
of one’s work group) includes strategies such as devising rounding models and interacting with
colleagues in a collegial manner. Hospitalists thrive on communication, and regularly voice their
rationale or concerns with primary-consultant designations to specialists. Often after discussing
violations to the primary-consultant work boundaries amongst themselves, hospitalists approach
specialists with a solution to resolve breakdowns and prevent future challenges to professional
jurisdiction.
As an example, Hospitalist Jennson described a work boundary conflict and solutions to
the problem:
Hospitalist Jennson also commented about the Infectious Disease unit, leaning more
toward consulting patient service—that there needs to be some definite improvement here
if this will be the case. She gave an example from a bad experience this weekend, in
which she took on the role of primary for a patient who had a case of septic arthritis. She
and Hospitalist Casabian came up with a plan to treat the patient, and over the weekend
Infectious Disease Specialist Homan [consultant from ID] “changed the plan 180
degrees,” and then Infectious Disease specialist Markus changed it back to what she and
Hospitalist Casabian initially came up with. Infectious Disease Specialist Markus ended
up going to tell the family himself about the change in plans without communicating that
to Hospitalist Jennson and Hospitalist Casabian. “Poor [Hospitalist Casabian] was caught
in the middle, and so were the other residents. But it was a good example of potential
problems going forward,” Hospitalist Jennson said. She went on: “We just need
consensus on some level from the division and clearly defined roles. To start off,
consultants need to talk to us first before they go to the family and talk treatment plans.
We need to tell ID that, and make sure they know.” (Leadership Meeting, Mar 18 2013)

The solution to communicate rules and address issues directly with other specialists are common
practices in managing the out-group to negotiate work boundaries.
Discussion with out-group colleagues resolved jurisdictional contests, as Pulmonologist
Roberts described a specific issue related to putting in orders on a patient at the time of
discharge:
61

“Discharge was the primary concern [with physician assignment], where from a
pulmonary standpoint, there is persistent concerns but sometimes it feels like the
hospitalist feels that ‘we [hospitalists] aren't doing anything’. We have discussed and in
the end, especially related to ventilator patients, they will defer to our recommendation.”

The key to resolution was a discussion between the pulmonologists and hospitalists who came to
an agreement about when the hospitalist should defer to pulmonologists for discharge orders—
especially for a certain group of patients.
Gastroenterologist Jade also emphasized the importance of ongoing communication
between hospitalist and specialist groups:
“Firstly, I have to say, over all my experience with the hospitalist service has been very
pleasant over the past four years. There have been changes in approach by many of the
hospitalist staff in their patient care services. Secondly, I think providing answers
[regarding the practice of physician assignment] with a yearly reminder with a foot note
to changes adopted and their rationale will go a long way in cooperative working between
divisions.”

Since work boundaries are often unclear, Gastroenterologist Jade reiterated that it is important to
define and re-evaluate guidance on assignment. Many times, by simply communicating issues
and working together toward a resolution, disagreements over whether the hospitalist or
specialist is making the decisions can be worked through. However, sometimes involving other
specialties is what yields a resolution, as Cardiologist Jackson stated:
“None of these disagreements [over who’s primarily making the decisions] have led to a
bad outcome for a patient. Most times these disagreements are discussed in person and
can be worked out. Occasionally it may take a third party’s input such as [cardiothoracic]
surgery to come to an agreement.”
DISCUSSION
Focusing on jurisdictional negotiation among professionals with overlapping expertise,
this research describes: (1) coordination challenges in collaborative work settings that lead to
conflict over work boundaries; (2) factors that professionals rely upon to identify work
62

boundaries through intrapersonal negotiation; and (3) the means by which professionals manage
conflict over work boundaries through interpersonal negotiation. Through ethnographic and
survey approaches, I provide detail on the nature of jurisdictional negotiation among
professionals with overlapping expertise. After analyzing my data and linking literatures on work
design, negotiation, and the professions, I provide a more comprehensive theory of the boundary
work process for jurisdictional negotiation in the workplace.
Implications
A Model of Boundary Work
Using grounded theory to analyze my observational and survey data, I developed a
conceptual model which depicts how professionals engage in boundary work as they negotiate
their jurisdictions. Few studies (e.g. Kreiner et al., 2009) have empirically unpacked boundary
work concepts to understand what such processes look like in practice. The model I develop
shows how intrapersonal and interpersonal negotiation relate to boundary work as professionals
identify and manage their jurisdictional domains in relation to one another.
Along these lines, my findings illustrate how professionals derive and defend
jurisdictions in light of two conditions. First, this study shows how a new role, like the
hospitalist, establishes their jurisdictions in an organization among already-established
professionals. Hospitalists used boundary management practices such as modifying the in-group
to differentiate themselves and show value relative to other specialists. Further, hospitalists
clarified rules with the intention of standardizing practice to not only streamline workflow, but
also construct work boundaries that would minimize stigmas about their role (e.g. hospitalists
characterized as “super residents”). And last, hospitalists managed the out-group by initiating
constant communication with specialist colleagues to affirm their identity and legitimate their
63

value. Through these mechanisms of interpersonal negotiation, novel professional roles establish
their jurisdictions relative to other professionals in the workplace.
Second, this study explains how professionals with overlapping domains of expertise, in
particular generalists and specialists, negotiate jurisdictions in their daily work. Work boundaries
are constructed to represent jurisdictions, and I show how task assignments render jurisdictions
in the workplace. The boundary work model of jurisdictional negotiation can apply to multiple
scenarios in which professional expertise overlaps—between generalists and specialists, and also
among specialists in the same specialty. For example, an endocrinologist assigned to a patient
with multiple endocrine neoplasia, a rare disease, may feel uncomfortable treating such a patient
if they have fewer years of experience (knowledge and expertise requirements), are new to the
particular hospital and care team dynamics (degree of socialization), or already have too many
patients under their care (resource demands). For any of these reasons, the endocrinologist
initially assigned may wish to pass the patient on to another endocrinologist in their division.
The boundary work model of jurisdictional negotiation can also apply to industries
outside of healthcare. In the engineering field, for example, the systems engineer functions as a
generalist coordinating the work of other specialist engineers. To see how the model derived in
Figure 1 fits into the engineering context, let us imagine the scenario in which the systems
engineer is working on a project to design a radar, along with other assigned software, hardware,
and testing/evaluations specialist engineers. Through intrapersonal negotiation, each of the
engineers may decide they need to challenge their initial assignment on this project, and if they
determine there is a problem in assignment, they engage in interpersonal negotiation. As work is
carried out to design and build the radar, and contradictions arise, intrapersonal and interpersonal
negotiation of work boundaries can occur. Depending on the outcome of these negotiations, the
64

systems engineer may be assigned a certain task over one of the specialist engineers and vice
versa. As these assignment substitutions are made, jurisdictions are established over time, in as
much as task assignments can set a precedent for carving work boundaries going forward.
In unpacking the components of intrapersonal and interpersonal negotiation, an aim of
this work is to make the division of labor more effective and streamlined in professional work
environments. The boundary work model formulated in this study offers several practical
applications. First, contradictions and intrapersonal/interpersonal negotiation represent
components of boundary work that are transferrable to many professional work environments,
especially those that are collaborative and house professionals with overlapping skills. Second,
the boundary work model is functional, in that it can be used as a training tool for professionals
in collaborative settings. Professionals can learn to be more attuned to the various boundary
identification factors and contradictions that can emerge in their setting, as well as possible ways
to work together using boundary management practices to resolve conflict over jurisdictions.
Simply discussing this framework for jurisdictional negotiation reinforces a culture that
welcomes communication and resolution of work boundary conflicts. In creating that culture,
organizations can have systems in place which may facilitate interpersonal negotiation through
the boundary management practices derived in this research.
This research also has important consequences for fields like healthcare, in which change
and complexity are persistent. New roles emerge constantly, and it is imperative to understand
the nature of work boundary decisions that foster integration of these roles as professionals
collaborate in the workplace. The boundary work model explains how new professions are
established through the assignment of work, potentially causing shifts in the work boundaries of
other professions.
65

Contribution to Existing Theories
My study also aims to bring insights to existing theoretical streams of work. First, this
research informs the scholarship on generalists and specialists. Freidson (2001, p. 121) asked the
provocative question—how does professionalism, which derives its value largely in
specialization, contend with the growing prominence of managerial generalism? The present
study offers an empirical understanding of this question, namely at the workplace level. While
the literature offers abstract characterizations of generalists and specialists, this study shows the
persistent negotiations and shifts in work boundaries of such overlapping roles in a collaborative
environment. Although there has been prior work which describes possible points of contention
between generalists and specialists (Martin et al., 2009), to my knowledge, there are no studies
that show how generalist and specialist professionals distinguish themselves from each other and
navigate conflicts over jurisdictional domains in their daily work. This study aims to shed light
on such processes.
Second, this study provides greater understanding into how professionalization, or the
process of a profession establishing itself, unfolds in daily work life. There is a dearth of
literature that addresses the microprocesses that underlie professionalization occurring at the
workplace level (Bucher, Chreim, Langley, & Reay, 2016; Reay, Golden-Biddle, & Germann,
2006). This study attempts to show how professional roles are established and altered in practice,
and thus, can have significant implications for professions that experience rapid and substantive
transformations due to growing complexity. Intrapersonal and interpersonal negotiation may not
only serve to identify and resolve contests around work boundaries, but also offer strategies for
new professions to integrate into organizations among established professionals. For example,
Reay, Golden-Biddle and GermAnn (2006) study the way a new profession, the nurse
66

practitioner, institutionalizes their role amidst existing roles, based on the degree of the role’s
embeddedness in a given context. My research adds to this study by uncovering the ongoing
negotiation practices in the workplace that occur between newer professional roles and longer-
standing ones.
Presumably then, novel professions engage in intrapersonal and interpersonal negotiation
to establish their function relative to other professionals in an organization. Employing such
tactics can enable new professions to define and justify their role in contexts that have yet to
fully embrace their function. Practicing intrapersonal and interpersonal negotiation in the
workplace can serve to institutionalize a profession in an organization over time, potentially
advancing the process of professionalization at the macro-level. For example, as hospitalists
obtain board certification, their experiences in the workplace may shape residency training
programs and institutional-level criteria for assessing hospitalist performance. Thus, this study
shows foundational aspects of establishing and negotiating jurisdictions in work settings, which
can eventually influence the public and legal forums of professionalization.
Directions for Future Research
While the mechanisms detailed in this research are grounded in the interactions between
professionals sharing jurisdictions in an extremely complex work setting, there exist some
limitations that provide opportunities for future research. First, since this study focuses on a
single hospital, one might examine interactions between specialists and other types of generalists
(e.g. brand managers, systems engineers) across a variety of settings. Empirical investigations
performed in alternative settings could provide comparative evidence that may corroborate or
extend the boundary work model derived herein. Additionally, other settings might reveal the
67

generalizability of this research, since MCH represents an extreme case of workplace
complexity.
To develop a boundary work model of jurisdictional negotiation, I relied on individual
accounts, and these accounts may only provide a limited understanding of variations in
perceptions about jurisdictions. Future research can more systematically explore variation in
perceptions that may be related to such factors as political and demographic differences between
specialist-generalist groups within an organization. For instance, future studies may focus on
how demographic characteristics, such as organizational tenure of professionals, may influence
professionals’ identification and management of work boundaries. Tenure has been shown to
influence the way individuals construct boundaries around their identities (Kreiner, Hollensbe
and Sheep, 2009). In the professional workplace, tenure can shape attitudes about work relative
to colleagues. Said differently, the shorter the organizational tenure, the more professionals
contemplate their role in relation to others (Katz & Kahn, 1978; Miller & Jablin, 1991). It may
be the case that professionals with shorter tenure lack familiarity with their organizational
context, and therefore, may more frequently engage in intrapersonal negotiation to assess work
boundaries compared to professionals with longer tenure. Although professionals with shorter
tenure in an organization tend to seek out information as they clarify their role (Bauer, Bodner,
Erdogan, Truxillo, & Tucker, 2007; R. Kraut, Burke, Riedl, & Resnick, 2010), those who are
newer to an organization are less comfortable speaking up whenever they think there is a
problem (Rudman, Borgida, & Robertson, 1995). Consequently, professionals with longer tenure
may be more inclined to engage in interpersonal negotiation whenever issues arise, and therefore
use boundary management practices to resolve contradictions in work boundaries. From these
arguments, scholars can explore whether conflicts over work boundaries are resolved more
68

effectively by professionals with longer or shorter tenure, and also the conditions which promote
confluence of work boundary perceptions among professionals.
As it is generally true for inductive qualitative research, the intent of this study is not to
be directly transferrable to any setting. While the chosen setting has many commonalities with
other collaborative work environments dealing with high levels of complexity, there are unique
attributes of this context which are not necessarily all-encompassing representations of other
work environments. Nevertheless, the deep insights generated by ethnographic research sheds
light on the nature of jurisdictional negotiation that occurs in the workplace. Such insights
provide opportunities for novel research investigating generalist and specialist work boundaries
in a variety of settings. The boundary work model in this study illustrates how persistent
negotiation facilitates the division of labor in collaborative settings. In doing so, this research
provides a springboard for future explorations of how professions are established, altered, and
maintained in everyday work interactions.
69

CHAPTER 3 - The Nature of Work and Professionals’
Jurisdictional Claims (Phase 2 Study)
70

ABSTRACT
The division of labor between professionals is largely driven by their claims to expertise, yet how
professionals’ claims associate with the nature of the work they actually practice is not well
understood. In today’s multidisciplinary organizations, the overlapping expertise of professionals,
particularly generalists and specialists, further complicates the association between jurisdictional
claims and the attributes of daily work. Using the healthcare context, this study investigates how
generalist and specialist professionals claim their jurisdictions in relation to the nature of tasks
they regularly perform. Four task dimensions were examined, based on prior scholarly work and
my observations in the field: fundamentality, breadth, client contact, and resource use. To carry
out this research, I administered a survey to generalist and specialist physicians in eight divisions
at a children’s hospital (36 percent response rate). The generalists physicians are called
hospitalists, and they or the specialist can be assigned patients, based on the patient’s principal
diagnosis. The survey presented the most common pediatric diagnoses for the various specialties,
and asked respondents whether they believe each diagnosis belongs in the specialist’s domain or
in the hospitalist’s domain. Descriptive statistics provided insight into demographic differences
between generalist and specialist respondents (e.g. level of experience, organizational tenure), as
well as how much agreement, or lack thereof, respondents demonstrated concerning demarcations
of their domains of expertise. Results show that specialists, rather than generalists, were preferred
to be in charge of tasks with high knowledge fundamentality, extensive breadth of knowledge, and
high client contact. No significant results were found for the resource use category. In identifying
dimensions of tasks that are most associated with a specialist’s or generalist’s jurisdiction, this
study aims to theorize who should be in charge of commonly-performed tasks when professional
expertise overlaps.
71

INTRODUCTION
Organizations are increasingly becoming multidisciplinary environments, comprised of
many different types of specialists, as well as generalists who coordinate workflow (Brock,
2006; Greenwood, Suddaby, & Hinings, 2002). With the variety of professionals who may
possess overlapping expertise, multidisciplinary organizations experience challenges in
effectively carrying out the division of labor (Ben-Menahem, von Krogh, Erden, & Schneider,
2015; Cronin & Weingart, 2007). In the context of professional service organizations, the
division of labor represents the distribution of control among professionals, and in particular, the
settlement of jurisdictions (Abbott, 1988). Jurisdictions are settled by external forces (e.g.
policies, regulations) and internal forces (e.g. professionals’ desire for autonomy) (Abbott,
1988). However, regardless of influence, jurisdictional settlements in the workplace are rendered
by professionals who persistently negotiate task boundaries (Boyce, 2008; Freidson, 1976). As
professionals negotiate, they make knowledge claims to particular work, which serve to establish
their jurisdictions and resolve disputes (Hirschkorn, 2006).
The practice of negotiating jurisdictions involves applying abstract knowledge to
particular types of work (Abbott, 1995). Therefore, the nature of work can have a crucial
influence on professionals’ knowledge claims, and consequently, jurisdictional settlements in the
division of labor. Several scholars have studied the microprocesses of professionals’
jurisdictional claims. For example, Barrett et al (2012) explored how the implementation of a
new technology reconfigures work boundaries between professionals. In addition, Currie et al
(2012) described the impact of social position and status on the carving of jurisdictions between
new nursing roles and existing specialist physician roles. Further, Allen (2000) studied the
micropolitical practices of nurses as they establish their jurisdictions in the organization.
72

However, amidst such impactful studies, there is a marked absence of research investigating the
relationship between the nature of work and professionals’ jurisdictional claims.
Thus, this research aims to examine how professionals with overlapping expertise claim
their jurisdictions, according to the nature of the work they regularly perform. In particular, I use
the healthcare context to investigate the relationship between the nature of tasks and the division
of labor among professionals with overlapping expertise, namely generalists and specialists.
Through an inductive approach using field observation and prior organizational theory, I develop
hypotheses about the particular technical features of tasks that lead to specialist versus generalist
task assignment. I test my hypotheses using a survey I administered to generalist and specialist
physicians in the Department of Pediatrics at Metropolitan Children’s Hospital
7
. In this context,
the generalist physicians are called hospitalists. The survey asked the physicians whether a
hospitalist or a specialist should primarily have jurisdiction over patient care for common
pediatric diagnoses in each of the seven specialties surveyed. I subsequently asked eight focus
groups of hospitalists to categorize each of the 176 diagnoses in the survey into four categories
that correspond to task dimensions of professional work: fundamentality, breadth, client contact,
and resource use. Through my analysis of the survey results, I show novel, and in some cases
counterintuitive, findings: generalists were less likely than specialists to be associated with tasks
categorized as high fundamentality, high breadth, and high client contact. Significant results
were not found for tasks categorized in terms of resource use.
An important step in performing this analysis was to measure agreement about claims to
expertise among generalists and specialists. Agreement, or lack thereof, would quantify variation
in jurisdictional claims within the sample, and further validate the necessity of investigating the

7
Pseudonyms have been used in this analysis.
73

problem identified herein. As a result, in examining jurisdictional claims in complex work
environments, this study also introduces a way to assess professionals’ claims to their work, as
well as identify potential areas of contestation about jurisdictions, advancing our understanding
of what professionalization looks like in the workplace. Further, this study provides descriptive
results that instigate more investigation into the characteristics of the professional (e.g.
experience-level, organizational tenure) that might relate to the enactment of jurisdictions in
practice. Ultimately, the aim of this study is to unpack professionals’ associations between their
jurisdictions and the attributes of their daily work, and in doing so, generate greater clarity on the
professional division of labor in multidisciplinary settings.
THEORETICAL BACKGROUND
Generalist and Specialist Professionals in the Workplace
For many professions, establishing control over specialized work is a persistent process
whereby professionals must continually negotiate their jurisdictions (Greenwood & Lachman,
1996). Organizational structure can affect the way professionals establish their roles and practice
work (Adler & Kwon, 2013; Dahlander & O’Mahony, 2011; Vough, 2012). The workplace can
have an important influence on professionalization and the regulation of work, such that
institutional-level rules and standards are translated into daily practice (Cooper & Robson, 2006).
An understanding of the organization as a site for professionalization is necessary, particularly in
complex environments, where uncertainty and idiosyncratic conditions are enduring
characteristics. Organizations deal with greater complexity by adopting a multidisciplinary
orientation, in that they house more specialist roles, all the while increasing generalist roles to
coordinate differentiated activity (Brock, 2006; Greenwood et al., 2002). In these contexts, the
overlap of generalist and specialist expertise complicates coordination practices (Waring &
74

Currie, 2009), since overlapping jurisdictions are intensely contested in an effort to preserve
one’s professional status and advance the process of professionalization (Abbott, 1988).
Overlapping jurisdictions could have advantages, such as enabling professionals to speak
a “common language” (Argyres, 1999; Bechky, 2003) and subsequently reduce complexity
(Nonaka & Takeuchi, 1995). However, an unintended consequence of overlapping domains is
the potential difficulty in identifying the work boundaries of different individuals who possess
enough knowledge to perform similar tasks. To mitigate the challenges of overlapping
jurisdictions, organizations may attempt to routinize professional practice such that either
generalists or specialists can claim authority over certain tasks (Crowston, 1997). Nevertheless,
professionals regularly deal with idiosyncratic, multifaceted, and uncertain issues corresponding
to the clients they serve (Champy, 2009). As a result, traditional bureaucratic forms (e.g. Weber,
1947) become increasingly ineffective with more complexity, and systematizing professional
work can have disastrous outcomes. Despite the problems posed by bureaucratic structures in
professional organizations, coordination is more necessary as the jurisdictions of previously
separated professionals converge in more collaborative work environments (Thornton, Jones, &
Kury, 2005). Coordination, therefore, plays an important role in enacting the field-level logics of
jurisdictional claims that are upheld and altered in the everyday practice of professional work
(Smets, Morris, & Greenwood, 2012).
While organizations facing complex environments require effective coordination in the
division of labor to succeed, little is known about how jurisdictional claims are managed
between professionals working in the same context. Carlile & Rebentisch (2003) describe how
knowledge of task dependencies and of divergent viewpoints is required to effectively manage
boundaries between specialties. Other studies describe the use of “boundary objects,” such as
75

new technologies, in delineating specialized work (e.g. Majchrzak, More, & Faraj, 2012).
However, the literature provides scant explanation of how the actual work performed by
professionals is interpreted with regard to one’s jurisdiction. In order to understand how
specialist and generalist work is defined in practice, we must first delve into prior scholarship
that examines the division of labor between professionals and its relation to the core knowledge
requirements.
Professional Division of Labor and Knowledge
While there are many perspectives on the essence of professions, this study takes the
view that construction of the knowledge basis for an occupation is central to a profession’s
existence (Abbott, 1991; Freidson, 2001; Halliday, 1987). Along these lines, the specialized
knowledge of professionals contributes significantly to high production and performance in
organizations. Organizations value specialized knowledge based on two primary assumptions,
foundational to the knowledge-based view of the firm (Grant, 1996). The first extends from
Simon’s principle of bounded rationality, in that individuals have a limited capacity to obtain,
store, and process information; therefore, organizational efficiency hinges on the creation of
specialized roles which possess expertise over a select knowledge area. Relatedly, the second
assumption is that all human productivity depends on knowledge, and therefore, the appropriate
application of specialized knowledge areas. Both of these assumptions require effective division
of labor to delineate the knowledge and expertise of different professionals in an organization.
Since professionals are autonomous agents largely responsible for their day-to-day
practice, the division of labor is synonymous with the division of knowledge (Sharma, 1997).
The division of knowledge is a product of differentiated group knowledge, whereby each
member contributes unique know-how (Lewis & Herndon, 2011). Although shared knowledge
76

can be helpful in assessing group members’ specialized expertise (Lewis and Herndon, 2011),
such knowledge may create notable instances of overlapping capabilities, especially as observed
in the case of generalist and specialist work. Further still, the idiosyncratic nature and
unpredictable demands of complex work environments also creates notable challenges in
distinguishing one professional role from another. Consequently, we must make more effort to
understand the way knowledge is organized in such settings where the potential for jurisdictional
overlap is high, because this has tremendous bearing on coordination strategies of organizations
dealing with complexity (Iansiti, 1995).
Although knowledge-based theories describe various mechanisms to coordinate
specialized work (e.g. the use of rules, routines, formalized communication), it is unclear how
the division of labor between professionals with overlapping expertise occurs in complex work
environments. Mismatches between the work boundaries and professional roles are not
uncommon (Carlile, 2004), and this issue is complicated by the variety of potentially intersecting
jurisdictions in complex settings. However, certain attributes of the work itself may serve to
clarify whose jurisdiction is invoked, and we explore this notion next.
Nature of Tasks in Identifying Expertise
The division of labor is a core organizational design problem, namely that a task must be
divided into subtasks which must be assigned to proper individuals (Galbraith, 1977).
Professional organizations, in which discretionary power resides at the lower levels, may deal
with additional challenges in streamlining the division of labor because control over work is
often not centrally mandated (Galbraith, 1977). In exploring this core organizational design
problem, contingency theory may explain the fit between the type of work professionals perform
and the role most suited to handle such work to optimize performance. Contingency theorists
77

(e.g. Thompson, 1967) have traditionally focused on the macro-environmental variables that
relate to behavior at the organizational level, and have not make predictions about the internal
dynamics of an organization at the meso-level (Levitt et al., 1999; Sinha & Van de Ven, 2005).
Fit between professional expertise and work may be difficult to achieve, especially when
jurisdictions between different types of professionals can intersect.
The boundaries between generalist and specialist professionals serve as an important
interface to understand the organization of knowledge and division of labor in the workplace.
One may be able to abstractly define generalist and specialist jurisdictions on absolute terms, but
to empirically comprehend the nuances of their areas of expertise, we must study generalist and
specialist professionals on relative terms (Freidson, 2001). Specifically, what is it about work– in
particular, the nature of tasks– that makes it more generalized or specialized? In asking such a
question, this study takes a task-based approach, similar to recent scholars (Garicano & Wu,
2012; McIver, Lengnick-Hall, Lengnick-Hall, & Ramachandran, 2013), to understand how
generalist and specialist roles in the organization effectively match expertise with the
requirements of a task.
I therefore examine the issue of the division of labor in the workplace, particularly as
rendered by professionals who organize knowledge and subsequently make claims to their work
in relation to the attributes of individual tasks. I focus on four task dimensions which emerged
from over three years of field observation at a large metropolitan children’s hospital, as well as
from examination of the management and medical literature: fundamentality, breadth, client
contact, and resource use. Each of these task dimensions channels the perspectives of classical
organizational design theorists (e.g. Daft and Lengel, Lawrence and Lorsch, Perrow), while
building on more recent theories on knowledge domains and the professions.
78

Fundamentality. When assessing a task, organizational members consider the types of
information available to them, largely in terms of two broad attributes: uncertainty and
equivocality (Daft and Lengel, 1986). Uncertainty refers to the absence of information, while
equivocality refers to the possible misinterpretations of information. When equivocality and
uncertainty are high, ambiguity is also high and managers are needed to bring more clarity in
such situations (Daft and Lengel, 1986). However, when equivocality is low and uncertainty is
low, these situations are well-defined and require no real intervention.
Translating these elements of information processing to the knowledge-base of
professional work, we turn to Stinchcombe (2001) work on knowledge stratification, which
describes how knowledge organizes social groups. Knowledge is stratified according to whether
it is more or less fundamental. Fundamentality can be defined in terms of two related
components: epistemological clarity and invariability in procedures. First, epistemology refers to
the means for knowing the nature of something– what an entity is and how it came into
existence. It explains how “cognitive subjects come to know the truth about a given phenomenon
in reality” (Bodenreider, Smith, & Burgun, 2004). Since epistemology explains how knowledge
can be incorporated into practice, it can be a term used to describe the scope of knowledge
pertaining to an entity. In the work environment, a task with high epistemological clarity means
that both the type of problem and its source can be understood and measured. Relatedly, the
second feature of fundamentality is that procedures in handling a task are largely invariable. This
means that knowledge is applied to tasks through procedures that have been tried and tested,
yielding greater certainty in the content and the context of application. Therefore, practicing
fundamental knowledge involves little variability, and often incorporates elemental abilities
associated with one’s formal education and training.
79

In connection to Daft and Lengel’s view, greater epistemological certainty and
invariability in the knowledge associated with tasks translates into less ambiguity. Tasks
requiring more fundamental knowledge can be thought of as work that comprises a core domain–
meaning work that is constructed around that which an occupation originally mobilized. Such
tasks are likely to be associated with a dominant profession, since they most clearly align with
the normative perceptions about the profession generally upheld by society (Zetka, 2003).
FIGURE 4 - Knowledge Fundamentality and the Use of a Specialist versus Generalist

Figure 4a illustrates the relationship between knowledge fundamentality and the use of a
specialist versus generalist. As a simplification, the diagram shows a single fundamental
component of knowledge associated with a certain specialist X, while peripheral knowledge,
associated with the generalist, surrounds the fundamental component and can serve as areas
linking disparate knowledge bases. Peripheral knowledge may be invoked if ambiguity is
introduced to a task that was once more certain. To better explain the difference between
fundamental and peripheral knowledge, Figure 4b provides an example from the medical field. A
patient may be diagnosed with chronic liver failure. This condition involves an organ (the liver)
80

that is in a specialist’s field (gastroenterologist), and it is also well understood through evidence-
based medicine. Therefore, chronic liver failure can be considered as fundamental to the
gastroenterologist’s expertise. However, let us consider the case in which a patient is admitted to
a hospital for chronic liver failure, but during the course of treatment, the patient also develops
symptoms that are unrelated to the principal diagnosis (e.g. fever, emesis, seizures, chest pain).
In the event of such ambiguity, the peripheral knowledge of a generalist may be invoked to
understand what is causing changes in the patient’s condition.
To elaborate upon the association between fundamental knowledge and specialized
professions, we now turn to the literature on knowledge brokering. Knowledge brokering is the
translation of knowledge in the daily practice of professional work (Currie & White, 2012), and
involves professionals’ shared understandings of each other’s work (Brown, Collins, & Duguid,
1989). The work of professionals may be comprised of ‘component knowledge,’ which is
rendered through the linkage of context to specific professional practice (Currie and White,
2012). This type of knowledge is often considered the core of specialized work, and is
relentlessly defended (Abbott, 1988). Generalists do not have such a core basis of their practice,
and instead more closely align with what is referred to as ‘architectural knowledge,’ which is
knowledge that superficially overlays specialist knowledge and resultantly can engage multiple
types of component knowledge in a cursory manner (Currie and White, 2012). So on the one
hand, generalists do not have a defined knowledge basis; in other words, there is no clear
knowledge premise for generalist work, since their practice comprises a holistic, interdisciplinary
system of knowledge (Von Bertalanffy, 1972). On the other hand, specialist knowledge
comprises foundational ‘docking points’ that serve as the basis for more intricate knowledge
development (Postrel, 2002). Correspondingly, I propose the following:
81

Hypothesis 1. Tasks with high fundamentality are associated with the specialist’s domain,
rather than the generalist’s domain.
Breadth. Classical organizational design theories identify two key features of the scope of work:
variety and interdependence (Langfred & Moye, 2004). Task variety refers to the number of
exceptions, or novel situations and problems, encountered in performing a task (Perrow, 1967).
Tasks with high variety have many exceptions, so cannot be easily standardized or routinized.
One result of an increasing number of exceptions may be the need to invoke different types of
specialists’ knowledge and skills to creatively handle a novel situation. Since high task variety
settings require more flexibility, bureaucratic and rules-based structures are not as effective. The
other feature of work, interdependence, refers to tasks that require coordination with others
(Thompson, 1967). Interdependence may be sequential or reciprocal between workers, and as
interdependencies along either of these dimensions increase, so does the need for a more
centralized coordination role to oversee such dependencies (Stewart & Barrick, 2000). For highly
interdependent tasks, individual autonomy can be detrimental, leading to silos and disjointed
efforts (Langfred and Moye, 2004).
Professionals in the work environment apply their knowledge to handle varying degrees
of variety and interdependence. Tasks that involve more variety and more interdependence
conceivably depend on a greater breadth of knowledge for successful coordination of efforts.
Greater knowledge breadth involves a broader assortment of information and skills, and given
the limits of human cognition, often incorporates only superficial aspects of subject areas
relevant to the issue at hand. Generalists are known to deal with tasks that require a greater
breadth of knowledge, while specialists are associated with tasks embodying greater depth of
knowledge (Cohen, 2013b; Grant, 1996). Such a distinction is supported by ethnographic
82

research which reveals that, in comparison to specialists, generalists use a greater variety of tools
to complete a task and may deal with large quantities of information that span a variety of
disciplines (Treem, 2012).
Figure 5a captures the relationship between knowledge breadth and specialist versus
generalist expertise. The diagram shows fundamental knowledge in two different dominant
domains, one related to specialist X and the other to specialist Y. The generalist in this case
serves as the linkage between these two disparate knowledge bases, and he or she coordinates the
work of specialist X and specialist Y to ensure successful completion of the focal task. Figure 5b
illustrates a simplified example of diabetes mellitus. An endocrinologist is associated with the
diabetes mellitus diagnosis. However, due to complications of the condition, a patient may also
suffer from coronary artery disease, which is associated with the work of a cardiologist. To
coordinate the work of the endocrinologist and cardiologist– essentially to ensure unified
decision-making and effective communication– a generalist will be responsible in managing the
care of such a patient.
Developing this example further, generalists acting as integrators possess an extensive
breadth of knowledge, and so they are able to function as linking mechanisms that coordinate
work between different groups of specialists (Nadler & Tushman, 1999). Because a network of
different specialists may be involved in tasks requiring greater breadth of knowledge, generalists
function as linking roles that open channels of communication between specialists, provide more
holistic services to clients, and route clients to different specialist services as necessary (Whetten
& Aldrich, 1979). Thus, generalists intrinsically deal with a greater variety of issues, and
therefore can link specialists through their intermediary function, as well as manage task-task
dependencies between specialists (Crowston, 1997). I therefore hypothesize the following:
83

FIGURE 5 - Knowledge Breadth and the Use of a Specialist versus Generalist

Hypothesis 2. Tasks with high breadth are associated with the generalist’s domain,
rather than the specialist’s domain.
Client contact. Chase and Tansik (1983) describe the extent of ‘customer contact,’ or the degree
to which a customer is in direct contact with a particular service facility relative to the total time
needed to service the customer, as a contingency variable that affects organizational
performance. Organizations with high levels of customer contact were found to be less
productive, since the presence of customers can disrupt routines and the flow of work, as well as
put exaggerated demands on professionals that would not otherwise occur if customer contact
were more limited (Danet, 1981). As a result, decisions are largely programmable for work
involving low customer contact, because these settings are typically well-structured and less
turbulent compared to high customer contact settings (Chase & Tansik, 1983).
84

Since this paper addresses professional work, I use the term ‘client’ (rather than
‘customer’) as someone who specifically engages the services of a professional, and I also make
necessary adaptations of this concept to the professional context. I define client contact more
specifically as the intensity of professional services a client requires, such as the professional’s
time, effort, or expertise to handle unpredictable events related to a certain task. High client
contact would be associated with more unstructured and customized professional services (e.g.
tending to an acute event in a patient’s condition), while low client contact is associated with
routinized and efficient professional services (e.g. monitoring a patient’s condition). Thus,
linking to Chase and Tansik’s findings, high client contact would require greater effectiveness
through customized professional services, and low contact would require greater efficiency
through routinized professional services. In comparing professionals, specialists with a deeper
knowledge basis are more apt to provide customized services to increase effectiveness for high
client contact tasks, while generalists with a broader knowledge basis are apt to provide more
routinized services to increase efficiency for low client contact tasks.
A professional group’s client relationships mark one of the important distinguishing
features between specialties. Since it is the client who is seeking the services of the professional,
the client greatly influences the requirement of more specialized work to support their needs
(Freidson, 1959). This is a central aspect of the functionalist view of professions, in that
professional work hinges on the client-expert relationship (Parsons, 1968; Wilson, 1959). Audit
firms, for example, are known to hone and compartmentalize expertise in order to address the
particular needs of their clients (Boone, Brocheler, & Carroll, 2000).
The reason is that it is economical for professional work to have a substantial client base
to scale knowledge, skills, and know-how (Benston, 1985). Scholars claim that “the division of
85

labor is limited by the extent of the market” (Smith, 1965), meaning that if there is demand for a
particular service, this demand substantiates more specialized work to handle greater client
needs. More specialized services are used for work demonstrating greater client demands
because knowledge is costly, in the sense that it is expensive to integrate different types of
expertise required to address particular needs. Thus, clients with high demands for professional
service tend to act as a device for both differentiating work and settling jurisdictions (Abbott,
1988). Hence, I suggest the following:
Hypothesis 3. Tasks with high client contact are associated with the specialist’s domain,
rather than the generalist’s domain.
Resource use. Resources refer to the inputs required to effectively complete a task, which for the
purposes of this study, may include both human and physical capital. One of the most important
functions of a manager centers on allocating resources among different groups in an organization
(Kraut, Pedigo, McKenna, & Dunnette, 1989), because managerial decisions about task
partitioning involve efficient utilization of specialized resources (Von Hippel, 1990). Since
generalists are associated with efficiency and organizational-level concerns, resource use and
allocation is one of their main functions as coordinators.
Generalists are most suited for linking roles, and in this capacity, serve to oversee
resource dependencies and effective use (Crowston, 1997; Galbraith, 1977). This means that
generalists link specialists such that they facilitate procurement of resources necessary to
complete a task and simultaneously minimize professional conflict (Wholey & Huonker, 1993).
However, resource management practices often pose a threat to professional autonomy in that the
allocation of resources can either facilitate or hinder practice (Wynia, Cummins, VanGeest, &
Wilson, 2000). To minimize professional conflict, generalists are supposed to be impartial to any
86

particular subgroup, and should be the most suited for managing resources in the completion of
complex, collaborative tasks. Thus, I hypothesize the following:
Hypothesis 4. Tasks with high resource use are associated with the generalist’s domain,
rather than the specialist’s domain.
DATA AND METHODS
Empirical Context
This research focuses on generalist and specialist physicians at Metropolitan Children’s
Hospital (MCH). The generalists I study are called hospitalists and they belong to the Hospital
Medicine (HM) division at MCH. MCH hired its first hospitalist in 2000, and subsequently, the
HM division was established in 2009. Today, MCH has about 50 active hospitalists. The
hospitalists at MCH are general pediatricians acting as coordinators of patient care whom solely
work in the hospital setting and do not have outside clinical practices. While hospitalists have the
informal authority to assign work to either themselves or a specialist, a specialist has the
authority to also contest those assignments. Thus, as in most professional environments,
hospitalists and specialists have the discretion to define their own daily practice, and so they co-
construct their work boundaries on a regular basis with virtually no top-down control.
Additionally, since hospitalists possess a certain degree of knowledge relevant to each
specialty, they may perform the tasks of the specialists they coordinate. This means that
hospitalists serve as “primary” or “consultant” physicians in a variety of specialty divisions in
the Department of Pediatrics at MCH. For a patient admitted into a particular division’s unit in
the hospital, the hospitalist may be assigned as primary physician (or the foremost decision-
maker in treating a patient), while the specialist may be a consultant physician (only provide
decision support to the hospitalist on a required or optional basis); alternatively, the specialist
87

may be assigned as the primary physician for a particular patient, while the hospitalist may serve
a consultant role. Thus, hospitalist and specialist physicians’ primary or consultant assignments
conceivably delineate jurisdictional domains in daily practice. It should be noted that assignment
determinations are made by hospitalist and subspecialist physicians, and so are at the discretion
of the professionals performing the work.
Data Collection
Using Qualtrics software, I administered an online survey to hospitalists and to specialist
physicians in seven specialty divisions– Cardiology, Endocrinology, Gastroenterology,
Hematology/Oncology (Hem-Onc), Neurology, Pulmonology, and Rheumatology– within the
Department of Pediatrics at MCH. I worked closely with each Division Head physician, as well
as the Research Director of the HM division, in developing the survey tool. The sample includes
66 full-time, contract, and fellow physicians in each of the divisions at MCH (response rate of
approximately 36 percent). The goal of the survey is to understand the jurisdictional claims of
generalists and specialists as indicated by the assignment of physicians to primary and consultant
roles for various top diagnoses of the seven specialties. The top diagnoses comprise of a list of
the most frequently-admitted patient cases by specialty, and were vetted by the respective
Division Heads. Each specialty’s top diagnoses list is essentially viewed as a list of the different
types of tasks that are most commonly performed in the daily work of these professionals.
In analyzing the division of labor concerning patient diagnoses, this study focuses on
jurisdictional claims associated with the presence of certain attributes embedded in tasks. I
analyzed the data using the physician-diagnosis as the unit of analysis, meaning that I looked at
each distinct physician response in categorizing top diagnoses in either the specialist’s or
hospitalist’s domain (n = 1,589).
88

TABLE 4 - Demographic Characteristics Comparing Specialists and Hospitalists

Respondent Characteristics
My first objective was to compare generalist and specialist characteristics. Table 4 shows
the demographic differences between hospitalist and specialist respondents. There was no
significant difference in the position held (full-time attending physician versus fellow physician)
between the specialists and hospitalists. However, results show that on average, compared to
specialists, hospitalists tended to be younger, have less years of experience (i.e. fewer years post-
residency, fewer years at MCH, fewer years practicing in their current domain), and have more
weeks per year at MCH serving in the inpatient clinical setting. These differences make sense,
given the more recent emergence of the hospitalist profession, since generally, this newer role
has fewer physicians who have been in the profession for longer periods of time. Further, the
variation in demographic characteristics between respondents necessitates the inclusion of the
control variables used in my analysis.
Next, I examined variation in claims to expertise among hospitalists and specialists by
measuring agreement concerning selection of domains for each of the diagnoses included in the
Specialists
(N= 46)
Hospitalists
(N= 20) P
Age
≤ 30 yrs 2% 5% < 0.01
a
31 to 40 yrs 44% 84%
41 to 50 yrs 19% 0%
51 to 60 yrs 9% 5%
> 60 yrs 19% 0%
Years post-residency, avg 14.6 6.2 0.01
b
Years in domain, avg 13.7 6.0 0.01
b
Years at CCH, avg 11.0 5.2 0.04
b
Position
Attendings 84% 89% 0.35
a
Fellows 16% 11%
Inpatient weeks, avg 9.0 15.0 < 0.01
b
a
Wilcoxon-Mann-Whitney test
b
one way ANOV A
89

survey. Agreement was measured using the reliability coefficient, Pi, which is part of Fleiss'
(1971) derivation of the inter-rater reliability measure, kappa:
𝑃 𝑖 =
1
𝑛 (𝑛 −1)
∑ 𝑛 𝑖𝑗
(𝑛 𝑖𝑗
− 1),
𝑘 𝑗 =1

where n is the number of raters, i is the subject unit being rated (i.e. diagnosis), and j represents
categories selected by raters (i.e. hospitalist primary or specialist primary). Advantages of using
Pi include the following: (1) Pi may be used with small or varying sample sizes; (2) agreement is
weighted by the number of raters, meaning that diagnoses with fewer physician raters will have
an adjusted Pi value; and (3) Pi reflects agreement when raters have prior knowledge and chance
ratings are unlikely. The value of Pi ranges from zero (perfect disagreement) to one (perfect
agreement). To make clearer comparisons of agreement across diagnoses and specialties, a
threshold of at least 0.6 was used to identify diagnoses with sufficient agreement (or
“consensus”) among physician respondents, which is consistent with other applications of inter-
rater agreement measures (see Krippendorff, 2012).
Figure 6 illustrates the variation in agreement about jurisdictional claims amongst
hospitalists and specialists for all seven specialties. For each specialty, the percentages of
diagnoses that had a Pi of at least 0.6 is shown, as well as whether a specialist or a hospitalist
was selected for those diagnoses; “no consensus” diagnoses are those which Pi was less than the
0.6 threshold. It is noteworthy that a substantial amount of diagnoses for each division
demonstrate disagreement, or no consensus, among physicians concerning whether a hospitalist
or specialist should be assigned. Over 50 percent of the diagnoses demonstrated no consensus (Pi
< 0.6) in three of the seven specialties (Endocrinology, Gastroenterology, and Neurology).
Additionally, three of the seven specialties demonstrated consensus on diagnoses in which a
90

specialist was selected to be in charge, rather than a hospitalist. The variation in agreement on
jurisdictional claims (specialist or hospitalist) for different specialties’ top diagnoses further
validates the need to explore the issues described herein. In particular, what is it about the
diagnoses that may be related to the variation in jurisdictional claims between physicians?
FIGURE 6 - Percentage of Diagnoses by Specialty Demonstrating Consensus (Pi ≥ 0.6)
Among Specialists and Hospitalists

VARIABLES AND ANALYSIS
Dependent Variable
Using the physician-diagnosis as the unit of analysis, I analyze the outcome variable of
whether the generalist’s domain (rather than the specialist’s domain) is selected for a particular
diagnosis. The independent and control variables are discussed below. Since the outcome
variable is binary (1 if generalist, 0 if specialist), I used logit regression models clustered on
specialty division and physician ID to examine the likelihood of selecting a specialist as it relates
to each of the four task dimensions identified. Clustering provided for robust standard errors, and
accounted for the nonindependence of observations.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Rheumatology
Hematology-Oncology
Pulmonology
Cardiology
Endocrinology
Gastroenterology
Neurology
Specialist domain Hospitalist domain no consensus (Pi < 0.6)
91

Independent Variables
The main effects correspond to the task dimensions described earlier: fundamentality,
breadth, client contact, and resource use. In order to operationalize the nature of tasks as it relates
to diagnosis categories, I asked eight focus groups of two to three hospitalists each to classify the
176 top diagnoses from the survey into four categories: high fundamentality, complex chronic
condition, intensive monitoring required, and high resource use. These categories were
inductively derived during an observational period at MCH (e.g. patient rounds, physician
meetings), as well as consultation with the medical and management literature. A description of
each of these task categories is provided in the following discussion of the independent variables.
I developed these descriptions with the help of several physicians at my field site in order to
ensure relevance to the professionals in my context, and by referencing the medical literature. All
of the main effect variables are binary (1 represents presence of the task dimension for a
particular diagnosis, 0 represents absence of the task dimension for a particular diagnosis).
Fundamentality high. The focus groups categorized a diagnosis as having high fundamentality
if the diagnosis has a:
“Well-defined expected course, complications, treatment and monitoring needs
that are in a certain [physician’s] domain of knowledge, skills and comfort-level.”

Such a diagnosis is less ambiguous, and is usually associated with what is known as the
“Standard of Care” in the medical field, which is comprised of treatment guidelines that specifies
appropriate patient care based on scientific evidence or collaboration among relevant medical
professionals. The Standard of Care outlines patient treatment for a particular condition, such
that medical errors and possible malpractice issues could be avoided. Thus, these diagnoses
demonstrate high fundamentality because they have a clearly outlined course of treatment and
often coincide with legal protections of physician practice.
92

Breadth high. I also asked the focus groups whether each diagnosis listed was a complex chronic
condition (CCC), which is defined in the pediatrics literature as follows:
"Child has a diverse set of conditions and multisystem disease; may be
technology dependent; has frequent inpatient admissions; and requires multiple
medications, multiple specialists, and optimal care coordination across
inpatient/outpatient settings.” (Feudtner et al, 2014; Simon et al, 2010)

A CCC diagnosis serves as a proxy for tasks that are greater in breadth because such a diagnosis
would likely require a holistic understanding of several affected organ systems and the ability to
integrate multiple specialists’ expertise in the care of the patient.
Client contact high. The focus groups also categorized a diagnosis as requiring intensive
monitoring if the following criteria were met:
“Patient has a condition that requires frequent intervention and has a propensity
for acute deterioration, and who is likely to require a physician who is/can be
rapidly available.”

Such diagnoses require greater client contact for multiple reasons. This type of patient is in a
very unstable state, in which unexpected deterioration can rapidly take place, and therefore
instinctive decision-making under conditions of uncertainty places greater demands on the
physician in charge. Also, a patient with this type of diagnosis requires much time from their
assigned physician, who may frequently intervene during the course of treatment.
Resource use high. Lastly, focus groups were asked to classify diagnoses as having a high
resource use requirement, according to the following definition:
“Diagnosis/workup often requires use of multiple ancillary services and support
(e.g. physical/occupational/speech therapy, social work, discharge planning),
possible frequent admissions, and longer length of stay.”

The use of multiple types of professionals in different fields, as well as the higher costs incurred
from potentially longer patient stays and frequent readmissions, makes these types of diagnoses
more resource intensive.
93

Controls
I included several covariates to account for the demographic characteristics of the survey
respondents. I controlled for basic characteristics, such as age, gender (male), and whether the
respondent was a hospitalist or a specialist (hospitalist). Age was measured using a nominal scale
(e.g. 1 is under 30 years, 2 is 31 to 40 years, 3 is 41 to 50 years, 4 is 51 to 60 years, and 5 is more
than 61 years).
I also controlled for the number of years the physician had been practicing in their current
domain, post-residency training (years in current domain). This variable was included because it
is possible that the more years of experience the physician has as a hospitalist or as a specialist,
the more likely they are to assume work in their own domain due to greater comfort. It is often
the case that physicians use their individual sovereignty (Freidson, 1986), as well as their social
and cultural capital (Starr, 1982), to define their daily work; both sovereignty and socio-cultural
capital are likely to be established over time in one’s professional practice.
Further, I included a variable for the amount of time the physician respondent actually
spends working in the clinical setting through measurement of the number of weeks per year of
the physician’s inpatient service (practical experience). This practical experience controls for the
possibility that the professional possesses more or less familiarity with the day-to-day activities
of their work environment. Greater practical experience would imply that the professional has
more iterations performing daily tasks, and therefore better intuition about jurisdictional
domains. Such intuition may not just be a product of repeated performance of tasks, but also
socialization and more consistent interaction with colleagues. I also controlled for the size of
each division (size of division), or the number of physicians practicing in each. This was to
account for the possible effect of power differences, namely control over work, since larger
94

groups may perceive authority over more tasks. Group size can also influence the degree of
cohesiveness between members, because it is more difficult to unify perceptions when more
people are involved (House & Miner, 1969).
Last, I controlled for the impact the physician’s current institution could have on their
practice. I measured this by dividing the physician’s tenure in years at MCH by the number of
years since the physician had finished their residency (organizational impact factor). Thus, the
higher the organizational impact factor, the more likely the professional is to be influenced by
their present organizational environment through the indoctrination of certain policies and
guidelines. As a result, professionals with a higher organizational impact factor would tend to
make jurisdictional claims that are more influenced by their degree of socialization and taken-
for-granted procedures of their current institution.
I analyzed pairwise correlations of the control variables described above (Table 2).
Variables with high correlations were adjusted for multicollinearity: (1) physicians’ age was
adjusted by their number of years in their current domain; and (2) physicians’ number of years in
their current domain was mean-centered.
RESULTS
Table 5 shows the descriptive statistics and correlations for the main effects: high
fundamentality, high breadth, high client contact, and high resource use. The table displays
weakly positive correlations between each of the task dimensions, except for the extensive
breadth and high resource use variables, which demonstrate a moderately positive correlation (r
= 0.664). The stronger relationship between extensive breadth and high resource use would be
expected, since highly complex tasks that potentially involve multiple specialists and integration
of a diverse set of activities would also likely require the implementation of many resources to
95

complete such tasks. As a result, resource use high was residualized from extensive breadth tasks
to account for potential collinearity.
Table 6 captures the results for the logit models predicting selection of the generalist’s
domain for a particular diagnosis. Model 1 shows only the control variables. Model 2 displays
the controls and fundamentality high variable. Model 3 includes the controls and the breadth
high variable. Model 4 adds the client contact high variable to the control variables. Model 5
shows the resource use high variable, along with controls. Model 6 leaves out all the control
variables and shows only the 4 main effect variables. Model 7 is the full model.
Examining the full model, a few of the control variables demonstrated significant results.
The age of the physician respondents and the organizational impact factor had a positive and
statistically significant relationship with the selection of a hospitalist. In other words, older
physicians were more likely to select a hospitalist in charge rather than a specialist, as were the
physicians who spent more years at MCH compared to any other institution. Interestingly, years
in domain had a significant negative relationship with physician assignment. In particular, those
physicians who had spent more years working in their domain had a greater likelihood of
selecting a specialist’s domain over the generalist.
With regard to the main effects, the full model shows that diagnoses demonstrating high
fundamentality had a negative and highly significant effect on selection of a hospitalist
physician, rather than a specialist (p < .001). This result supports Hypothesis 1: tasks with high
fundamentality are likely to correspond to the specialist’s domain, rather than the generalist’s
96

TABLE 5 - Descriptive Statistics and Correlations

Variable Mean s.d. 1 2 3 4 5 6 7 8 9 10
1 Hospitalist 0.427 0.495 1
2 Years in current domain 9.446 9.850 -0.3092* 1
3 Practical experience 11.703 7.065 0.3701* -0.3257* 1
4 Organizational impact factor 0.952 0.596 0.1392* -0.3040* 0.4002* 1
5 Male 0.444 0.497 0.0392 0.2444* -0.1500* 0.1199* 1
6 Age 2.546 1.019 -0.4034* 0.8944* -0.2654* -0.2195* 0.1658* 1
7 Fundamentality high 0.581 0.494 -0.011 -0.008 0.0556* 0.0675* -0.031 -0.028 1
8 Breadth extensive 0.459 0.499 -0.023 -0.010 0.0979* -0.047 -0.037 0.020 0.2072* 1
9 Client demands high 0.492 0.500 0.040 0.033 0.0583* 0.0790* 0.0673* -0.005 0.3106* 0.1307* 1
10 Resource use high 0.420 0.494 -0.008 -0.026 0.0995* -0.001 -0.0599* 0.011 0.1358* 0.6642* 0.1805* 1
N = 1589; * p < 0.05
97

domain. In addition, Model 7 counterintuitively indicates that, conditioned on other task
dimensions, diagnoses categorized as having greater breadth were less likely to be associated
with the hospitalist’s domain compared to the specialist’s domain (p < .001), which contradicts
Hypothesis 2. The full model also shows when client contact is high for certain diagnoses, a
hospitalist is less likely selected to be in charge, compared to the specialist (p < .001). This
supports Hypothesis 3. Finally, the results in Model 7 do not present significant findings
concerning diagnoses with high resource use. Thus, Hypothesis 4, which proposed that
generalists were more likely to be selected for tasks with high resource use, had no significant
results.
To better visualize the effects of each of the task dimensions on selection of a hospitalist
rather than a specialist, Figure 7 shows how the outcome variable varies with changes in the
parameters of interest. Here we see the average predicted probabilities of selecting a generalist’s
domain, conditioned (or marginalized) on the task dimensions specified in Model 7 (CI 95%).
This plot compares the predicted effects of the four task dimensions. Tasks characterized as high
resource use and high client demands had the highest predicted effect on selection of a generalist,
while tasks with high breadth and high fundamentality had the lowest predicted effect on
selection of a generalist.
DISCUSSION
Different levels of agreement concerning jurisdictional claims for various types of tasks points to
a more thorough investigation of the results in Table 6, in order to understand variation in
jurisdictional claims and the potential for contestation. I first suggested that tasks which can be
codified and routinized– or have high knowledge fundamentality– are primarily associated with
the specialist’s jurisdiction. Prior scholarly work provides contradictory support on whether tasks
98

TABLE 6 - Results of Logit Regressions for Assigning Hospitalist Physician by Diagnosis

Variables M1 M2 M3 M4 M5 M6 M7
CONTROLS
Hospitalist -0.602 -0.829 -0.584 -0.869 -0.622 -0.851
(0.125) (0.077) (0.123) (0.071) (0.120) (0.096)
Years in current domain
a
-0.0233*** -0.0235*** -0.0262*** -0.0226*** -0.0233*** -0.0261***
(0.000) (0.000) (0.000) (0.001) (0.000) (0.000)
Practical experience -0.0277** -0.0215* -0.0220* -0.0212* -0.0274** -0.0171
(0.004) (0.028) (0.027) (0.030) (0.004) (0.106)
Organizational impact factor
b
0.612*** 0.621*** 0.533*** 0.671*** 0.615*** 0.557***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Male 0.143 0.146 0.170 0.180 0.139 0.171
(0.216) (0.214) (0.153) (0.124) (0.229) (0.159)
Age 0.273* 0.250 0.313* 0.294* 0.275* 0.293*
(0.043) (0.070) (0.023) (0.032) (0.041) (0.040)
Size of division 0.0190 0.00997 0.0141 0.0214 0.0178 -0.00459
(0.579) (0.817) (0.639) (0.638) (0.613) (0.906)
MAIN EFFECTS
Fundamentality high -0.833*** -0.574*** -0.575***
(0.000) (0.000) (0.000)
Breadth high -1.023*** -0.869*** -0.580***
(0.000) (0.000) (0.000)
Client contact high -0.679*** -0.380** -0.818***
(0.000) (0.002) (0.000)
Resource use high
c
-0.165 -0.220 -0.150
(0.262) (0.149) (0.336)
constant -0.604 -0.0290 -0.0376 -0.387 -0.583 -0.531 0.718
(0.235) (0.964) (0.934) (0.561) (0.263) (0.719) (0.221)
N 1589 1589 1589 1589 1589 1589 1589
Log-likelihood -1054.4 -1029.5 -1011.0 -1037.8 -1053.8 -1026.3 -985.0
Note: p-values in parentheses
* p<0.05; ** p<0.01; *** p<0.001
a
mean-centered;
b
organizational impact factor = number of years working at CCH / number of years post-residency;
c
Resource use high residualized from extensive breadth tasks
99

FIGURE 7 - Marginal Effects of Task Dimensions on Probability of Selecting a
Hospitalist’s Domain for a Given Diagnosis

Note: bars represent 95% confidence interval range; predicted mean of fixed portion only.

with high fundamentality belong in a specialist’s domain or a generalist’s domain. One side of
the argument claims that such tasks can be easily transferrable to a generalist even if associated
with the work of a specialist (Garicano & Wu, 2012); the other side suggests that tasks with more
knowledge fundamentality reinforce the jurisdictional boundaries of specialists (Zetka,
2003). The results of this study coincide with the latter argument and support Hypothesis 1.
From the specialist’s perspective, tasks with high fundamentality provide an element of certainty
in an unpredictable workplace, and are therefore guarded by professionals who draw strong
linkages to these foundational aspects of their education and training. From the generalist’s
perspective, high fundamentality tasks have lower coordination costs because they can be more
readily communicated and transferred, making the coordination role less necessary (Garicano
100

and Wu, 2012). Said differently, if tasks largely depend on fundamental knowledge, generalists
are unlikely to associate themselves with such tasks since their coordination function is less
necessary.
The results contradicted Hypothesis 2, since the data suggests that the specialist’s domain
is associated with tasks that require greater breadth of knowledge and likely involve multiple
other specialists. Such a finding appears counterintuitive since generalists are associated with the
capacity to coordinate multiple specialists working on the same task, and they are largely able to
do so because of their expanse of knowledge. After discussing this result with several physicians
at my field site, an alternative explanation may be that one particular aspect of a task may be
preeminent over others. Professionals may associate a preeminent aspect of a task with a single
professional in an effort to streamline the division of labor and quell complexity when multiple
specialists are involved. For example, at MCH, a patient admitted for blockage of their lung
airways resulting from cystic fibrosis often has many different organ systems involved in their
illness, not just their lungs. However, a pulmonologist would be the primary physician (rather
than a hospitalist or another specialist) for a patient with cystic fibrosis because the patient’s
ailing lungs are the main reason other organs are affected as well. Therefore, it makes sense for a
specialist to be primarily held responsible for a particular task that requires an extensive breadth
of knowledge, when one aspect of the task may be the driving cause of all other resulting issues.
While on the surface, current theory would argue that tasks involving a greater breadth of
knowledge would more likely be in the generalist’s domain, these results offer an extension to
theory: tasks with a greater breadth of knowledge that also have an underlying problem
associated with the expertise of a particular specialist would likely require that specialist to be in
charge.
101

This study also provides evidence for the notion that work involving high client contact is
related to the specialist’s domain, rather than to the generalist’s domain. Such findings parallel
the economic literature that asserts specialist work is necessitated by greater market demand.
Organizations depend on the specialized knowledge of professionals to service the custom needs
of clients (Mills, 1990; Sharma, 1997). Clients get value from the application and delivery of
professionals’ specialized knowledge as their demands become more non-routine. This is the
reason why non-routine work is higher-priced, as we see in the example of in-house lawyers who
charge more for providing tailored services to clients with particular demands (Dawson, 2012).
In other words, more specialized work aggregates around where there is more demand for
customized services, since it is economically efficient to do so.
Finally, while there are no significant results for jurisdictional claims on high resource
use tasks, the literature can provide some guidance as to why such tasks may be difficult to
attribute certain expertise. Though prior studies make the assertion that generalists are better at
managing organizational concerns and coordinating with efficiency in mind, resource
dependence scholars identify the role of power in reshaping organizational structure, resulting
from reciprocal dependence on others for resources (Drees & Heugens, 2013). Administrators
are faced with the challenge of making sure a continuous supply of resources are provided to the
more powerful groups both within and outside of the organization, as well as ensuring that these
groups are satisfied (Aldrich and Pfeffer, 1976). Thus, when considering resources, attention to
the politics of decision-making inside the organization is important, particularly in contexts in
which organizations seek to manage or adapt to their environments (Pfeffer, 1972). The question
of the extent of power held by each of the specialties within an organization needs to be
addressed to further examine the role of resource use on the division of labor.
102

Therefore, one of the limitations of this study is that I did not address the degree of
relative power the hospitalist and specialist groups hold within the hospital, and their resultant
access to or control over resources. Another limitation of this work is that this is a single-site
study. Future research should look at other hospitals, and fields apart from medicine, to further
understand the relationship between task dimensions and jurisdictional claims. In addition, this
study did not give sufficient attention to the influence of professionals’ demographic
characteristics on the selection of jurisdictions. While control variables were included in the
analysis, the significant results demonstrated with such variables as the organizational impact
factor and years in current domain, warrant more investigation. Thus, future work could more
thoroughly examine the influence of professional characteristics on claims to jurisdictions.
Lastly, this study does not investigate how jurisdictions are actually enacted as various tasks
arise in the workplace– who is really assigned to patients with diagnosis X: a hospitalist or a
specialist? While there is value in understanding claims to jurisdictions, scholars should look at
what the division of labor between professionals with blurred domains truly looks like, keeping
the nature of tasks and professionals jurisdictional claims in mind. Other work may also explore
the influence of the actual (versus claims to) division of labor on performance outcomes, such as
the quality and cost of patient care in the healthcare context.
CONCLUSION
The aim of this study is to understand how the nature of tasks relates to professionals’
claims to expertise and the division of knowledge between professionals in complex,
collaborative organizational settings. I identify 4 dimensions that are salient attributes of
common tasks professionals perform: fundamentality, breadth, client demands, and resource use.
I argue that these task dimensions correspond to claims professionals make to their jurisdictions,
103

and therefore the division of labor in the workplace. To test relationship, I collected survey data
at a children’s hospital to analyze how the generalist (hospitalist) and specialist physicians define
the boundaries of their practice for tasks that may have overlapping areas of expertise between
these professionals. Results indicate that specialists are associated with tasks that require high
knowledge fundamentality, extensive breadth of knowledge, and high client demands. This
research informs our understanding of the way knowledge is organized in terms of professionals’
claims to expertise when jurisdictions blur between generalists and specialists, as well as the way
professional jurisdictions are enacted at the workplace level.

104

CHAPTER 4 - Conclusion
105

CONTRIBUTIONS AND NEXT STEPS
Theoretical and Practical Contributions
Multidisciplinary organizations house distinct yet interlinked professional expertise, in
response to the rise of more complex knowledge (Leathard, 2000: 7). With the growing
prevalence of multidisciplinary work environments, scholars argue that we are entering an “age
of hyperspecialization” in which the knowledge applied to carry out tasks is becoming more
granular to provide higher quality goods and services at lower costs (Johns et al, 2011).
However, work boundaries between specialized professionals do not perfectly compartmentalize
domains of expertise. Although prior research largely assumes that work boundaries remain
stable, it is often the case that boundaries are in fact permeable (Kreiner, Hollensbe and Sheep,
2009) and may allow for one professional to operate in the domain of another (Leathard, 2000).
My dissertation takes the perspective that interprofessional work boundaries are permeable and
frequently overlap, which lends to notable challenges in the professional division of labor. In my
dissertation, I conducted two studies which aim to unpack the nature of the division of labor
between professionals whose expertise overlaps, particularly generalists and specialists. In doing
so, this research makes several theoretical and practical contributions.
First, in the Phase 1 study, I construct a conceptual model that elaborates on boundary
work theory, in particular showing how professionals negotiate their jurisdictions through the
assignment of tasks. While theorists have developed significant insights related to the notion of
boundary work, few studies exists on how boundary work is applied in practice (Kreiner,
Hollensbe and Sheep, 2009). Since multidisciplinary organizations are growing in number, my
dissertation elucidates the way different types of professionals carry out boundary work to deal
106

with issues they encounter on a regular basis in the division of labor. By clarifying this process, I
aim to show how organizations can be more effective in the coordination of professional work.
Second, since this research hones the boundary work of jurisdictional negotiation, my
dissertation contributes to our understanding of the professional division of labor, specifically
looking at the roles of generalists and specialists in collaborative environments. By providing
ethnographic evidence in Phase 1, as well as a more systematic examination of professionals’
perceptions of their work relative to each other in Phase 2, I offer insights beyond our current
abstract understandings of generalist roles relative to specialist roles. I show how these
professionals interact and perceive their respective functions in their everyday collaborative
practice. Accordingly, this research shows how professionals extract the most critical factors in
identifying their work boundaries for common tasks and the practices professionals employ to
manage work boundary conflicts. Such findings contribute to scholarly work on the professional
division of labor, which does not comprehensively capture generalist roles relative to specialist
roles at the workplace level.
Third, by deriving factors that delineate professionals with overlapping expertise, this
research enhances our understanding of professionalization, or the process of a profession
establishing itself over time. Professionalization can occur at the macro-level public and legal
contexts, as well as the micro-level workplace context (Abbott, 1988). However, jurisdictions
established at the macro-level may differ substantially as they are translated in practice in the
workplace (Cooper & Robson, 2006). By examining how a relatively new role, the hospitalist,
integrates in a work setting of already-established professionals, my dissertation shows how
work boundaries can be subject to change in the daily practice of work, and therefore influence
the professionalization process at the workplace level. I add to the findings of scholars who have
107

studied the implementation of new professional roles in the workplace (e.g. Reay, Golden-Biddle
and Germann, 2006), by examining the daily practices of professionals of conceivably similar
status and overlapping domains of expertise. The practices I describe have the potential to
change work boundaries and reformulate perceptions about one’s professional role, ultimately
reshaping jurisdictional domains as they are enacted in the workplace. Thus, my dissertation
provides greater insight into the micro-level process of professionalization by giving an in-depth
account of the way professionals enact their jurisdictions by carving work boundaries and
assigning tasks.
With regard to practical contributions, my dissertation speaks broadly to
multidisciplinary organizations and more specifically to the healthcare field. The boundary work
model derived in Phase 1 explains the process of task assignment and ongoing negotiation in the
division of labor, and is transferrable to a variety of interprofessional settings. The boundary
work model is also actionable, in that it may serve as a training tool for more effective
collaboration between professionals, showing various factors that can make overlapping roles
readily discernable. Professionals may be trained to understand the means by which conflicts
over task assignment can be resolved through the boundary management practices described in
Chapter 2 (modifying the in-group, clarifying guidance, and managing the out-group). Further,
the boundary management practices provide a tangible means by which organizations can create
a culture of open communication as issues around the assignment of tasks arise. To facilitate a
culture of employing boundary management practices, organizations can develop mechanisms
catered to their particular context and available resources. As organizations create this culture of
communication about work boundaries, the potential to enhance professionals’ understanding of
108

their roles relative to each other, as well as promote team work and cooperation, may markedly
increase.
The findings herein might also be embedded in medical education and training programs.
In particular the workplace factors in identifying work boundaries (Phase 1) and the nature of
commonly treated diagnoses (Phase 2) can provide guidance on how medical professionals
should think about their role in relation to their colleagues and as they practice in
interdisciplinary care teams. In doing so my dissertation provides practical guidance in making
interprofessional practices more effective. Additionally, the findings of my dissertation have
important implications for comanagement practices in healthcare. Comanagement, or the sharing
of responsibility in patient care, is growing in popularity among hospitalists and specialists
(Society of Hospital Medicine, 2008). What this means is that there will likely be a growing
overlap in expertise between hospitalist and specialist physicians over the coming years, and my
dissertation identifies prominent factors that would delineate work boundaries between these
physicians. In doing so, this research aims to streamline the division of labor between hospitalist
and specialist physicians such that timelier patient assignments are made, roles are better
clarified, and communication is established to manage work boundaries more systematically.
Next Steps: Phase 3 Study
The Phase 1 and 2 studies led me to a third phase of study in which I investigate
performance outcomes resulting from the assignment of a generalist or specialist. In Phase 3, I
explore how physicians’ perceptions of whether a hospitalist or specialist should be in charge of
common patient diagnoses, as captured in the Phase 2 survey, aligns with the physicians actually
assigned. The goal is to see whether patient outcomes are affected when perceptions of who
should be in charge actually align with or deviate from the physician actually assigned to patients
109

at MCH. In a broader sense, the question I ask in Phase 3 is as follows: What are the
performance outcomes (i.e. quality and cost of patient care) when the actual assignment of a
generalist or specialist professional either aligns with or deviates from perceptions of who
should be in charge? Below, I will briefly highlight some aspects of the burgeoning study.
Theoretical Background
While scholars argue that there is an optimal level of bureaucracy imposed by an
organization over the professional-client relationship (Engel, 1969), others explain that
standardization of practice could be disastrous since professionals regularly deal with distinctive
and complex issues in serving their clients (Champy, 2009). In other words, traditional
bureaucratic forms (e.g. Weber, 1947), mainly visible in formalization and standardization of
practice, may be detrimental for complex professional work. Management theorists have long-
acknowledged the tradeoff between the efficiency that bureaucracy provides and the flexibility
inherent in adhocracy (Thompson, 1967), which highlights the challenge professionals face in
performing both routine and nonroutine work on a daily basis. In managing this tradeoff,
professional work is becoming more collaborative (Adler, Kwon and Heckscher, 2008), and
these collaborative professional settings are simultaneously experiencing growing pressures to
improve organizational performance (Adler & Heckscher, 2006).
Individual discretion and autonomy to carry out daily work is inherent to professional
practice, and much of the healthcare field adheres to the tenets of autonomy despite the tendency
toward collaborative communities of practice (Berwick, 2005). Professionals constantly navigate
the tension between their interest in personal autonomy and their professional commitment to
clients and colleagues (Mazmanian, Orlikowski, & Yates, 2013), and it is unclear the effect
discretionary decision making on a case-by-case basis can have on performance outcomes. Thus,
110

this Phase 3 study will explore the influence of professionals’ discretion on their performance.
Specifically, I will investigate patient outcomes that result when the actual assignment of a
hospitalist or specialist physician either aligns with or differs from their perceptions of who
should be in charge of common pediatric diagnoses.
Data Collection
8

I collected electronic medical record (EMR) data of patients admitted during January
2009 through August 2015. A medical record coder worked directly with me to derive the ICD-9
codes that match the top diagnoses listed in the Phase 2 survey, and after sorting through the
diagnoses that could be precisely coded, 77 of the 176 top diagnoses remained in the data set. Of
those diagnoses, only 57.8 percent demonstrated consensus from the Phase 2 survey in terms of
whether a hospitalist or specialist was to be assigned in charge of such patient cases; the
remaining percentage had no consensus (see Chapter 3 for calculations of the agreement score,
Pi, and determination of consensus). During the specified time period, 5,092 patients had a
principal diagnosis of one of those identified in the remaining list of top diagnoses.
The EMR data indicates the physician assigned to either a hospitalist or specialist for
patients with any of those identified principal diagnoses. Other variables in the data include
various patient demographic characteristics (e.g. age, gender, insurance type), the nature of the
patient’s condition (e.g. severity, mortality risk), and the outcome of the patient’s visit (e.g.
discharge to another facility, discharge home). Performance outcomes are captured along two
dimensions, namely the quality and cost of patient care. In the EMR data, quality measures
include 30-day readmission and medical errors (e.g. hospital acquired conditions). Cost measures

8
Since Phase 3 was in progress at the time of drafting my dissertation, the statistics reported for this study are
preliminary.
111

include the length of stay in the hospital and disaggregated hospital charges incurred over the
course of the patient’s visit.
Preliminary Findings
Table 7 summarizes a preliminary descriptive analysis of the data. Over all the patient
encounters during the specified time period (n = 5,092), approximately 30.3 percent were in the
hospitalists’ jurisdiction (i.e. assigned the physician of record) rather than the specialists’. The
statistics also show that patients new to MCH were assigned to hospitalists and specialists with
similar frequency. In addition, the data indicates that approximately 43.2 percent of patients were
assigned a physician that deviated from the consensus assignment selections in the Phase 2
survey (i.e. a specialist was actually assigned when a hospitalist was selected in the Phase 2
survey and vice versa). More specifically, actual assignments aligned with consensus perceptions
of assignment 55.8 percent of the time for hospitalists and 44.2 percent of the time for
specialists.
TABLE 7 - Descriptive Statistics for Hospitalist and Specialist Assignment at MCH
Primary Attending Assigned
Descriptive Statistics Hospitalist Specialist P
Total patient encounters 1544 3548
Percentage new patient encounters 73.4% 73.5%
Percentage of assignments that match consensus 55.8% 44.2%
Patient severity, avg 2.0 2.1 < 0.000
a

1.0 0.9
Patient mortality risk, avg 1.3 1.4 0.004
a

0.7 0.7
Length of stay, avg 5.1 7.2 < 0.000
b

8.9 20.2
Total hospital charges, avg $40,765 $75,474 < 0.000
b

$86,152 $285,274

Note: EMR data from MCH, representing Jan 2009-Aug 2015 period; n = 5,092 total patient encounters; standard deviations
reported in italics below averages for patient severity, mortality risk, length of stay, and total hospital charges
a
Wilcoxon-Mann-Whitney test;
b
t-test
112

In terms of patient characteristics, severity and mortality risk are each captured in the
EMR on a scale of 0 to 4, with 4 representing the highest severity level and mortality risk. On
average, both hospitalists and specialists were in charge of patients that had similar mortality risk
and severity scores. With regard to patient outcomes, the average length of stay was higher for
specialists compared to hospitalists, 7.2 days and 5.1 days respectively. The variation in the
length of stay for specialists was notably higher than for hospitalists. Further, on average, total
hospital charges (i.e. costs) were higher and had substantially more variation for specialists
compared to hospitalists.
To visualize how consensus assignments changed over time, Figure 8 shows the
percentage of patient cases per year in which the consensus assignment selections from the
survey matched actual assignment, comparing when a hospitalist or a specialist was actually
assigned as the primary attending.
FIGURE 8 - Percentage of Patient Encounters per Year with Assigned Physician Matching
Consensus Perceptions of Assignment at MCH

113

Note that the hospitalist program was officially established in September 2009 at MCH, and in
Figure 8, there is an initial decline but steady increase in the percentage of patients assigned to a
hospitalist when the physicians’ consensus perception was that a hospitalist should be in charge.
However, there is a slight decrease in the percentage of assigned hospitalists that aligned with
consensus perceptions that a hospitalist should be in charge in 2015, possibly because this
reflects only eight months of data, while the prior years reflect the full 12 months. A possible
reason for the increase in 2009 through 2014 is that the physicians had more iterations working
together and the Hospital Medicine division gradually established its presence, thereby making
physician assignments more consistent over time. On the other hand, Figure 8 demonstrates a
gradual decline in the percentage of patients assigned to a specialist when the physicians’
consensus perception was that a specialist should be in charge. One could speculate that the
specialists’ work boundaries were being shifted inward over time, and there may be more recent
attempts to reclaim work.
Figure 8 highlights the fact that even for those diagnoses with consensus agreement,
nearly half of patient encounters at MCH had a physician that was assigned which deviated from
agreed on perceptions of who should be in charge year over year. With such a noticeable
percentage of physician assignments that do not align with agreed upon determinations of who
should be in charge, several questions surface. Is professional discretion in the division of labor
being carried out properly? What effect do alignment and deviations from consensus assignment
have on patient outcomes? How does these outcomes compare to those patients with diagnoses
that did not have consensus about assignment in the first place? Such questions will be the
subject of more sophisticated statistical analyses using the EMR data collected in Phase 3.
114

FUTURE RESEARCH DIRECTIONS
My dissertation aims to be a springboard for future studies on the division of labor
between generalist and specialist professionals. While there are many possibilities, I will
summarize a few future research directions. Since this research was conducted in a single
hospital setting, future work should study these issues in different hospitals and professional
contexts to corroborate and further develop the findings herein. Additionally, scholars should
explore the role of the organization, and in particular executive management, in the professional
division of labor. Said differently, how should management engage with professionals in
facilitating the division of labor? How much professional autonomy should be maintained in this
process?
Another aim of future work may be to examine the role of specializing generalists
compared to generalizing specialists. As this dissertation discussed, prior scholarly work
describes that generalist roles are increasingly developing more technical expertise as they gain
greater experience and become more effective at their coordination role (Langbert, 2005; Gupta
1984). Generalist professionals in a variety of fields, including the fashion industry (Caronna &
Ong, 2011) and other industries experiencing digital transformation (El Sawy et al, 2015), are
specializing to handle the complexities of their work and manage a variety of specialist expertise
in the organization. Interestingly, the reverse is also taking place—in which specialists are
starting to generalize. For example, in the healthcare field, the greater volume of specialist
physicians compared to the patient demand for specialized services are causing an unregulated
division of labor in which specialists are starting to perform more generalized work (Fryer,
1991). As another example, specialists in the hospital setting, are themselves embracing the
hospitalist model and hiring specialists trained in a particular area, such as obstetrics, orthopedics
115

and behavioral health, as hospitalists (Darves, 2009). Ambler (2005) also describes the
importance of generalizing specialists in the software industry, as organizations seek to become
more agile and improve the quality of services provided as technologies continue to advance.
Thus, there is also a trend toward generalizing specialists.
Much remains unexplored concerning the emergence of generalizing specialists and
specializing generalists. For instance, the fact that a particular task can be performed by different
types of professionals can either reinforce each individual role in an interdependent system or
determine that a certain role may in fact be obsolete (Hasan, Ferguson, & Koning, 2015). Other
questions arise in exploring this important issue. In particular, what are the work boundary
tensions that specializing generalists face compared to generalizing specialists? How do
specializing generalists perform relative to generalizing specialists? What are the limitations that
specializing generalists encounter in the practice of their work, compared to generalizing
specialists?
To conclude, one of the hopes of this dissertation is to encourage management scholars to
pursue research on the professional division of labor. This type of scholarly work holds much
promise for fields like healthcare that depend on appropriate workflow in order to thrive. Among
the many detrimental consequences of an ineffective professional division of labor include a lack
of accountability among professionals, the potential for errors, relational tensions, and the
likelihood of incurring high costs due to ineffective coordination. While there are numerous
downsides, I believe interprofessional relationships and performance in an organization can
markedly improve as future research explores the mechanisms that underlie and enhance the
professional division of labor. I hope other scholars will join me in this endeavor.

116

117

REFERENCES

Abbott, A. (1988). The system of professions: An essay on the division of expert labor.
University of Chicago Press. Retrieved from http://psycnet.apa.org/psycinfo/1988-97883-
000
Abbott, A. (1991). The Order of Professionalization An Empirical Analysis. Work and
Occupations, 18(4), 355–384.
Abbott, A. (1995). Boundaries of Social Work or Social Work of Boundaries?: The Social
Service Review Lecture. The Social Service Review, 545–562.
Abbott, A. (2005). Linked ecologies: states and universities as environments for professions*.
Sociological Theory, 23(3), 245–274.
Adler, P. S., & Heckscher, C. (2006). Towards collaborative community. The Firm as a
Collaborative Community: Reconstructing Trust in the Knowledge Economy, 11–105.
Adler, P. S., & Kwon, S.-W. (2013). The mutation of professionalism as a contested diffusion
process: clinical guidelines as carriers of institutional change in medicine. Journal of
Management Studies, 50(5), 930–962.
Adler, P. S., Kwon, S.-W., & Heckscher, C. (2008). Perspective—professional work: The
emergence of collaborative community. Organization Science, 19(2), 359–376.
Allen, D. (2000). Doing occupational demarcation the “boundary-work” of nurse managers in a
district general hospital. Journal of Contemporary Ethnography, 29(3), 326–356.
Ambler, S. (2005). Quality in an agile world. Software Quality Professional, 7(4), 34.
Argyres, N. S. (1999). The impact of information technology on coordination: Evidence from the
B-2 “Stealth” bomber. Organization Science, 10(2), 162–180.
118

Ashcraft, K. (2012). The Glass Slipper:’Incorporating’Occupational Identity in Management
Studies. Academy of Management Review. Retrieved from
http://amr.aom.org/content/early/2012/07/31/amr.10.0219.short
Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000). All in a day’s work: Boundaries and micro
role transitions. Academy of Management Review, 25(3), 472–491.
Audet, A.-M. J., Doty, M. M., Shamasdin, J., & Schoenbaum, S. C. (2005). Measure, learn, and
improve: physicians’ involvement in quality improvement. Health Affairs, 24(3), 843–
853.
Barley, S. R. (1986). Technology as an occasion for structuring: Evidence from observations of
CT scanners and the social order of radiology departments. Administrative Science
Quarterly, 78–108.
Barrett, M., Oborn, E., Orlikowski, W. J., & Yates, J. (2012). Reconfiguring boundary relations:
Robotic innovations in pharmacy work. Organization Science, 23(5), 1448–1466.
Bauer, T. N., Bodner, T., Erdogan, B., Truxillo, D. M., & Tucker, J. S. (2007). Newcomer
adjustment during organizational socialization: a meta-analytic review of antecedents,
outcomes, and methods. Journal of Applied Psychology, 92(3), 707.
Bechky, B. A. (2003). Object Lessons: Workplace Artifacts as Representations of Occupational
Jurisdiction1. American Journal of Sociology, 109(3), 720–752.
Ben-Menahem, S., von Krogh, G., Erden, Z., & Schneider, A. (2015). Coordinating knowledge
creation in multidisciplinary teams: Evidence from early-stage drug discovery. Academy
of Management Journal, amj–2013.
Benston, G. J. (1985). The market for public accounting services: Demand, supply and
regulation. Journal of Accounting and Public Policy, 4(1), 33–79.
119

Beresford, L. (2011.). The Comanagement Conundrum From: The Hospitalist, April 2011.
Berg, J. M., Dutton, J. E., & Wrzesniewski, A. (2013). Job crafting and meaningful work.
Purpose and Meaning in the Workplace, 81–104.
Bessler, R.A. (2012). The Economics of Hospital Medicine. Society of Hospital Medicine
[Presentation].
Bodenreider, O., Smith, B., & Burgun, A. (2004). The ontology-epistemology divide: A case
study in medical terminology. In Formal ontology in information systems: proceedings of
the... International Conference. FOIS (Conference) (Vol. 2004, p. 185). NIH Public
Access. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4346778/
Boland Jr, R. J., Lyytinen, K., & Yoo, Y. (2007). Wakes of innovation in project networks: The
case of digital 3-D representations in architecture, engineering, and construction.
Organization Science, 18(4), 631–647.
Boone, C., Brdcheler, V., & Carroll, G. R. (2000). Custom service: Application and tests of
resource-partitioning theory among Dutch auditing firms from 1896 to 1992.
Organization Studies, 21(2), 355–381.
Boyce, R. A. (2008). Health workforce: innovation, substitution and reform’. Analysing Health
Policy: A Problem-Oriented Approach, 105–118.
Brock, D. M. (2006). The changing professional organization: A review of competing
archetypes. International Journal of Management Reviews, 8(3), 157–174.
Brock, D. M., & Powell, M. J. (2005). Radical strategic change in the global professional
network: the “Big Five” 1999-2001. Journal of Organizational Change Management,
18(5), 451–468.
120

Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning.
Educational Researcher, 18(1), 32–42.
Bucher, S. V., Chreim, S., Langley, A., & Reay, T. (2016). Contestation about Collaboration:
Discursive Boundary Work among Professions. Organization Studies,
0170840615622067.
Burkhardt, U., Erbsen, A., & Rüdiger-Stürchler, M. (2010). The hospitalist as coordinator: an
observational case study. Journal of Health Organization and Management, 24(1), 22–
44.
Burns, L. R., & Muller, R. W. (2008). Hospital-Physician Collaboration: Landscape of Economic
Integration and Impact on Clinical Integration. Milbank Quarterly, 86(3), 375–434.
Cappetta, R., & Cillo, P. (2008). Managing integrators where integration matters: insights from
symbolic industries. The International Journal of Human Resource Management, 19(12),
2235–2251.
Carlile, P. R. (2004). Transferring, translating, and transforming: An integrative framework for
managing knowledge across boundaries. Organization Science, 15(5), 555–568.
Carlile, P. R., & Rebentisch, E. S. (2003). Into the black box: The knowledge transformation
cycle. Management Science, 49(9), 1180–1195.
Caronna, C. A., & Ong, M. K. (2011). OrganizatiOn OF MEDICAL CARE. Jonas and Kovner’s
Health Care Delivery in the United States, 181.
Centers for Disease Control and Prevention, National Center for Health Statistics (February
2016). National Health Interview Survey Early Release Estimates [Data set].
Champy, F. (2009). 12. La culture professionnelle des architectes. Recherches, 152–162.
121

Chase, R. B., & Tansik, D. A. (1983). The customer contact model for organization design.
Management Science, 29(9), 1037–1050.
Cheng, H. Q. (2012). Comanagement: Who’s in Charge? Retrieved from
https://psnet.ahrq.gov/webmm/case/271/comanagement-whos-in-charge
Chien, A. T., & Rosenthal, M. B. (2013). Medicare’s physician value-based payment modifier—
will the tectonic shift create waves? New England Journal of Medicine, 369(22), 2076–
2078.
Chreim, S., Williams, B. B., & Hinings, C. B. (2007). Interlevel influences on the reconstruction
of professional role identity. Academy of Management Journal, 50(6), 1515–1539.
Clark, H. H. (1996). Communities, commonalities, and communication. Rethinking Linguistic
Relativity, 17, 324–355.
Coffman, J., & Rundall, T. G. (2005). The impact of hospitalists on the cost and quality of
inpatient care in the United States: a research synthesis. Medical Care Research and
Review, 62(4), 379–406.
Cohen, I. G. (2013a). Rationing legal services. Journal of Legal Analysis, 5(1), 221–307.
Cohen, I. G. (2013b). The globalization of health care: legal and ethical issues. Oxford
University Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=_TBpAgAAQBAJ&oi=fnd&pg=PP1&d
q=negotiation+contemplate+trade+off+cohen&ots=JQf3BiTorv&sig=cn9q6T0mEP3dIQ
HDMBOuHoocNaM
Cooper, D. J., & Robson, K. (2006). Accounting, professions and regulation: Locating the sites
of professionalization. Accounting, Organizations and Society, 31(4), 415–444.
122

Criscuolo, P., Dahlander, L., Grohsjean, T., & Salter, A. (2016). Evaluating novelty: the role of
panels in the selection of R&D projects. Academy of Management Journal, amj–2014.
Cronin, M. A., & Weingart, L. R. (2007). Representational gaps, information processing, and
conflict in functionally diverse teams. Academy of Management Review, 32(3), 761–773.
Crowston, K. (1997). A coordination theory approach to organizational process design.
Organization Science, 8(2), 157–175.
Currie, G., Lockett, A., Finn, R., Martin, G., & Waring, J. (2012). Institutional work to maintain
professional power: recreating the model of medical professionalism. Organization
Studies, 33(7), 937–962.
Currie, G., & White, L. (2012). Inter-professional barriers and knowledge brokering in an
organizational context: the case of healthcare. Organization Studies, 33(10), 1333–1361.
Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness
and structural design. Management Science, 32(5), 554–571.
Dahlander, L., & O’Mahony, S. (2011). Progressing to the center: Coordinating project work.
Organization Science, 22(4), 961–979.
Danet, B. (1981). Client-organization relationships. Handbook of Organizational Design, 2, 382–
428.
Darves, B. (2009). Specialists: The newest entrants in the hospitalist movement. The New
England Journal of Medicine Career Center. January.
Dawson, R. (2012). Developing knowledge-based client relationships. Routledge. Retrieved
from
https://books.google.com/books?hl=en&lr=&id=PA4m4r1ZAG0C&oi=fnd&pg=PR3&dq
123

=%22client%22+and+%22profession%22+and+%22knowledge%22&ots=M0rgek4P_X
&sig=3OrDOP18FnjPNyx-OOUHGS_IVr4
Deneckere, S., Euwema, M., Lodewijckx, C., Panella, M., Mutsvari, T., Sermeus, W., &
Vanhaecht, K. (2013). Better interprofessional teamwork, higher level of organized care,
and lower risk of burnout in acute health care teams using care pathways: a cluster
randomized controlled trial. Medical Care, 51(1), 99–107.
Detert, J. R., & Edmondson, A. C. (2011). Implicit voice theories: Taken-for-granted rules of
self-censorship at work. Academy of Management Journal, 54(3), 461–488.
Dougherty, D. (2001). Reimagining the differentiation and integration of work for sustained
product innovation. Organization Science, 12(5), 612–631.
Drees, J. M., & Heugens, P. P. (2013). Synthesizing and extending resource dependence theory a
meta-analysis. Journal of Management, 0149206312471391.
El Sawy, O.A., Kraemmergaard, P., Amsinck, H., Vinther, A.L., 2015. Building the Foundations
and Enterprise Capabilities for Digital Leadership: The LEGO Experience. 2015 Society
for Information Management Academic Paper Competition.
Engel, G. V. (1969). The effect of bureaucracy on the professional autonomy of the physician.
Journal of Health and Social Behavior, 30–41.
Fleiss, J. L. (1971). Measuring nominal scale agreement among many raters. Psychological
Bulletin, 76(5), 378.
Freidson, E. (1959). Specialties without roots: the utilization of new services. Human
Organization, 18(3), 112–116.
Freidson, E. (1976). The division of labor as social interaction. Social Problems, 304–313.
Freidson, E. (1986). Professional powers. Chicago: University of Chicago Press.
124

Freidson, E. (2001). Professionalism, the third logic: on the practice of knowledge. University of
Chicago press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=Tgn3sThG8QEC&oi=fnd&pg=PR7&dq
=freidson+2001+professionalism&ots=iNspIJBoMz&sig=RmzXYKWO_3IG8kXBqPIF
OsSkJk0
Fryer, G. E. (1991). The United States medical profession: an abnormal form of the division of
labour. Sociology of Health & Illness, 13(2), 213–248.
Galbraith, J. R., & Jay, G. (1977). Organization design. Addison-Wesley Reading, MA.
Galbraith, J. R., & Kazanjian, R. K. (1986). Organizing to implement strategies of diversity and
globalization: the role of matrix designs. Human Resource Management, 25(1), 37–54.
Garicano, L., & Wu, Y. (2012). Knowledge, communication, and organizational capabilities.
Organization Science, 23(5), 1382–1397.
Gieryn, T. F. (1983). Boundary-work and the demarcation of science from non-science: Strains
and interests in professional ideologies of scientists. American Sociological Review, 781–
795.
Gioia, D. A., Corley, K. G., & Hamilton, A. L. (2012). Seeking Qualitative Rigor in Inductive
Research: Notes on the Gioia Methodology. Organizational Research Methods.
Retrieved from
http://orm.sagepub.com/content/early/2012/07/20/1094428112452151.abstract
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. Aldine de Gruyter. Retrieved from
http://books.google.com/books?hl=en&lr=&id=rtiNK68Xt08C&oi=fnd&pg=PA1&dq=gl
aser+and+strauss&ots=UUyZUi_EVL&sig=4DSqTVq8o-MNg_-GGRUcVhviK50
125

Glasheen, J. J., Siegal, E. M., Epstein, K., Kutner, J., & Prochazka, A. V. (2008). Fulfilling the
promise of hospital medicine: tailoring internal medicine training to address hospitalists’
needs. Journal of General Internal Medicine, 23(7), 1110–1115.
Goodrich, K., Krumholz, H. M., Conway, P. H., Lindenauer, P., & Auerbach, A. D. (2012).
Hospitalist utilization and hospital performance on 6 publicly reported patient outcomes.
Journal of Hospital Medicine, 7(6), 482–488.
Grant, R. M. (1996). Toward a knowledge-based theory of the firm. Strategic Management
Journal, 17, 109–122.
Greenwood, R., & Lachman, R. (1996). Change as an underlying theme in professional service
organizations: an introduction. Organization Studies, 17(4), 563–572.
Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of
professional associations in the transformation of institutionalized fields. Academy of
Management Journal, 45(1), 58–80.
Gulati, R., Puranam, P., & Tushman, M. (2012). Meta-organization design: Rethinking design in
interorganizational and community contexts. Strategic Management Journal, 33(6), 571–
586.
Gupta, A. K. (1984). Contingency linkages between strategy and general manager
characteristics: A conceptual examination. Academy of Management Review, 9(3), 399–
412.
Halliday, T. C. (1987). Beyond monopoly: Lawyers, state crises, and professional empowerment.
University of Chicago Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=G1u5Yk-
126

pBtoC&oi=fnd&pg=PR9&dq=professionalization+halliday&ots=Z1T0o7gNfa&sig=ThIf
JGPXZoMR6m6g88dHntaFzIw
Hasan, S., Ferguson, J.-P., & Koning, R. (2015). The Lives and Deaths of Jobs: Technical
Interdependence and Survival in a Job Structure. Organization Science. Retrieved from
http://pubsonline.informs.org/doi/abs/10.1287/orsc.2015.1014
Hirschkorn, K. A. (2006). Exclusive versus everyday forms of professional knowledge:
legitimacy claims in conventional and alternative medicine. Sociology of Health &
Illness, 28(5), 533–557.
House, R. J., & Miner, J. B. (1969). Merging management and behavioral theory: The interaction
between span of control and group size. Administrative Science Quarterly, 451–464.
Hughes, E. C. (1963). Professions. Daedalus, 655–668.
Iansiti, M. (1995). Technology integration: Managing technological evolution in a complex
environment. Research Policy, 24(4), 521–542.
Ibarra, H. (1999). Provisional selves: Experimenting with image and identity in professional
adaptation. Administrative Science Quarterly, 44(4), 764–791.
Jencks, S. F., Williams, M. V., & Coleman, E. A. (2009). Rehospitalizations among patients in
the Medicare fee-for-service program. New England Journal of Medicine, 360(14), 1418–
1428.
Johns, T., Laubscher, R. J., & Malone, T. W. (2011). The age of hyper specialization. Harvard
Business Review, 89(7/8), 56–65.
Katz, D., & Kahn, R. L. (1978). The social psychology of organizations. Retrieved from
http://sites.idc.ac.il/dice/files/activity2.pdf
127

Kotter, J. P. (1982). General managers are not generalists. Organizational Dynamics, 10(4), 5–
19.
Kraut, A. I., Pedigo, P. R., McKenna, D. D., & Dunnette, M. D. (1989). The role of the manager:
What’s really important in different management jobs. The Academy of Management
Executive, 3(4), 286–293.
Kraut, R., Burke, M., Riedl, J., & Resnick, P. (2010). Dealing with newcomers. Evidencebased
Social Design Mining the Social Sciences to Build Online Communities, 1, 42.
Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. (2009). Balancing borders and bridges:
Negotiating the work-home interface via boundary work tactics. Academy of
Management Journal, 52(4), 704–730.
Krippendorff, K. (2012). Content analysis: An introduction to its methodology. Sage. Retrieved
from
https://books.google.com/books?hl=en&lr=&id=s_yqFXnGgjQC&oi=fnd&pg=PR1&dq=
krippendorff&ots=b1_TZTmmCV&sig=HbKEZREl6Lm0cDIqh6BHTkdM5iI
Langbert, M. (2005). The master’s degree in HRM: midwife to a new profession? Academy of
Management Learning & Education, 4(4), 434–450.
Langfred, C. W., & Moye, N. A. (2004). Effects of task autonomy on performance: an extended
model considering motivational, informational, and structural mechanisms. Journal of
Applied Psychology, 89(6), 934.
Lawrence, P. R., & Lorsch, J. W. (1967a). Differentiation and integration in complex
organizations. Administrative Science Quarterly, 1–47.
Lawrence, P. R., & Lorsch, J. W. (1967b). New management job: the integrator. Organizational
Effectiveness Center and School, 27.
128

Leathard, A. (2000). Health Care Provision: Past, present and into the 21st century. Nelson
Thornes. Retrieved from
https://books.google.com/books?hl=en&lr=&id=D2lVXWNQ7ZsC&oi=fnd&pg=PR7&d
q=(Leathard,+2000)&ots=0b3i5h1TiF&sig=GwgRXT4tUtYMRmtxYfvRq-aoW-I
Leicht, K. T., & Fennell, M. L. (1997). The changing organizational context of professional
work. Annual Review of Sociology, 215–231.
Leicht, K. T., Fennell, M. L., & Witkowski, K. M. (1995). The effects of hospital characteristics
and radical organizational change on the relative standing of health care professions.
Journal of Health and Social Behavior, 151–167.
Levitt, R. E., Thomsen, J., Christiansen, T. R., Kunz, J. C., Jin, Y., & Nass, C. (1999).
Simulating project work processes and organizations: Toward a micro-contingency
theory of organizational design. Management Science, 45(11), 1479–1495.
Lewis, K., & Herndon, B. (2011). Transactive memory systems: Current issues and future
research directions. Organization Science, 22(5), 1254–1265.
Lindenauer, P. K., Rothberg, M. B., Pekow, P. S., Kenwood, C., Benjamin, E. M., & Auerbach,
A. D. (2007). Outcomes of care by hospitalists, general internists, and family physicians.
New England Journal of Medicine, 357(25), 2589–2600.
MacDuffie, J. P. (2007). 12 HRM and Distributed Work: Managing People Across Distances.
The Academy of Management Annals, 1(1), 549–615.
Majchrzak, A., Jarvenpaa, S. L., & Hollingshead, A. B. (2007). Coordinating expertise among
emergent groups responding to disasters. Organization Science, 18(1), 147–161.
Majchrzak, A., More, P. H., & Faraj, S. (2012). Transcending knowledge differences in cross-
functional teams. Organization Science, 23(4), 951–970.
129

Malone, T. W., & Crowston, K. (1994). The interdisciplinary study of coordination. ACM
Computing Surveys (CSUR), 26(1), 87–119.
Martin, G. P., Currie, G., & Finn, R. (2009). Reconfiguring or reproducing intra-professional
boundaries? Specialist expertise, generalist knowledge and the “modernization”of the
medical workforce. Social Science & Medicine, 68(7), 1191–1198.
Mathews, A.W. (2015, September 21). Health-Care Providers, Insurers Supersize. The Wall
Street Journal. Retrieved from http://www.wsj.com/.
Max, W. (1947). The theory of social and economic organization. Translated by AM Henderson
and Talcott Parsons. Edited with an Introduction by Talcott Parsons.) The Free Press of
Glencoe Collier-Macmillan Limited. London.
Mayer, M. L., Skinner, A. C., & Freed, G. L. (2009). Interspecialty differences in the care of
children with chronic or serious acute conditions: a review of the literature. The Journal
of Pediatrics, 154(2), 164–168.
Mazmanian, M., Orlikowski, W. J., & Yates, J. (2013). The autonomy paradox: The implications
of mobile email devices for knowledge professionals. Organization Science, 24(5), 1337–
1357.
McDonough, J. E. (2011). Inside national health reform (Vol. 22). Univ of California Press.
Retrieved from https://books.google.com/books?hl=en&lr=&id=e1Z6D-
K9Rx8C&oi=fnd&pg=PP1&dq=mcdonough+inside+national+health+reform&ots=Tdh1
CIdHmb&sig=SuaBCfSbgq-2LAyjjJ18iUgoSdA
McGregor, M. (2010). A system dynamics approach to jurisdictional conflict between a major
and a minor healthcare profession. Systems Research and Behavioral Science, 27(6),
639–652.
130

McGuire, T. G. (2011). Demand for Health Insurance1. Handbook of Health Economics, 2, 317.
McIver, D., Lengnick-Hall, C. A., Lengnick-Hall, M. L., & Ramachandran, I. (2013).
Understanding work and knowledge management from a knowledge-in-practice
perspective. Academy of Management Review, 38(4), 597–620.
Meguid, C., Ryan, C. E., Edil, B. H., Schulick, R. D., Gajdos, C., Boniface, M., … McCarter, M.
(2015). Establishing a framework for building multidisciplinary programs. Journal of
Multidisciplinary Healthcare, 8, 519.
Meltzer, D., Manning, W. G., Morrison, J., Shah, M. N., Jin, L., Guth, T., & Levinson, W.
(2002). Effects of physician experience on costs and outcomes on an academic general
medicine service: results of a trial of hospitalists. Annals of Internal Medicine, 137(11),
866–874.
Miller, V. D., & Jablin, F. M. (1991). Information seeking during organizational entry:
Influences, tactics, and a model of the process. Academy of Management Review, 16(1),
92–120.
Mills, P. K. (1990). On the quality of services in encounters: An agency perspective. Journal of
Business Research, 20(1), 31–41.
Nadler, D. A., & Tushman, M. L. (1999). The organization of the future: Strategic imperatives
and core competencies for the 21st century. Organizational Dynamics, 28(1), 45–60.
Nambisan, S. (2002). Designing virtual customer environments for new product development:
Toward a theory. Academy of Management Review, 27(3), 392–413.
Nippert-Eng, C. (1996). Calendars and keys: The classification of “home” and “work.” In
Sociological Forum (Vol. 11, pp. 563–582). Springer. Retrieved from
http://link.springer.com/article/10.1007/BF02408393
131

Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company: How Japanese companies
create the dynamics of innovation. Oxford University Press, USA.
Pannick, S., Beveridge, I., Wachter, R. M., & Sevdalis, N. (2014). Improving the quality and
safety of care on the medical ward: a review and synthesis of the evidence base.
European Journal of Internal Medicine, 25(10), 874–887.
Parsons, T. (1967). Sociological theory and modern society. Free Press New York. Retrieved
from http://library.wur.nl/WebQuery/clc/437153
Parsons, T. (1968). Professions. I International Encyclopaedia of the Social Sciences, vol. 12.
New York: the Macmillan Company and The Free Press.
Perrow, C. (1967). A framework for the comparative analysis of organizations. American
Sociological Review, 194–208.
Pfeffer, J. (1972). Size and composition of corporate boards of directors: The organization and
its environment. Administrative Science Quarterly, 218–228.
Postrel, S. (2002). Islands of shared knowledge: Specialization and mutual understanding in
problem-solving teams. Organization Science, 13(3), 303–320.
Puranam, P., Raveendran, M., & Knudsen, T. (2012). Organization design: The epistemic
interdependence perspective. Academy of Management Review, 37(3), 419–440.
Raffaelli, R., & Glynn, M. A. (2014). Turnkey or tailored? Relational pluralism, institutional
complexity, and the organizational adoption of more or less customized practices.
Academy of Management Journal, 57(2), 541–562.
Rawson, D. (2002). Models of inter-professional work. Going Interprofessional: Working
Together for Health and Welfare, 38.
132

Reay, T., Golden-Biddle, K., & Germann, K. (2006). Legitimizing a new role: Small wins and
microprocesses of change. Academy of Management Journal, 49(5), 977–998.
Regoli, R. M., Culbertson, R. G., Crank, J. P., & Poole, E. D. (1988). Linkages between
professionalization and professionalism among police chiefs. Journal of Criminal Justice,
16(2), 89–98.
Rudman, L. A., Borgida, E., & Robertson, B. A. (1995). Suffering in Silence: Procedural Justice
Versus Gender Socialization issues in University Sexual Harassment Grievence
Procedures. Basic and Applied Social Psychology, 17(4), 519–541.
Schwartz, B. (1981). Vertical classification.
Sharma, A. (1997). Professional as agent: Knowledge asymmetry in agency exchange. Academy
of Management Review, 22(3), 758–798.
Siegal, E. M. (2008). Just because you can, doesn’t mean that you should: A call for the rational
application of hospitalist comanagement. Journal of Hospital Medicine, 3(5), 398–402.
Sinha, K. K., & Van de Ven, A. H. (2005). Designing work within and between organizations.
Organization Science, 16(4), 389–408.
Smets, M., Morris, T. I. M., & Greenwood, R. (2012). From practice to field: A multilevel model
of practice-driven institutional change. Academy of Management Journal, 55(4), 877–
904.
Smith, D. L. (1965). Sociology and the rise of corporate capitalism. Science & Society, 401–418.
Spitz‐Oener, A. (2006). Technical Change, Job Tasks, and Rising Educational Demands:
Looking outside the Wage Structure. Journal of Labor Economics, 24(2), 235–270.
http://doi.org/10.1086/499972
133

Starr, P. (1982). The social transformation of American medicine. Basic Books. Retrieved from
http://books.google.com/books?hl=en&lr=&id=FK4pBXGvQzoC&oi=fnd&pg=PR9&dq
=starr+1982+professional+social+cultural&ots=xXx3n29Z6N&sig=5qNY9IKrr6B_ogD
EtM1cEaOppLk
Stavert, R. R., & Lott, J. P. (2013). The bystander effect in medical care. The New England
Journal of Medicine, 368(1), 8.
Stewart, G. L., & Barrick, M. R. (2000). Team structure and performance: Assessing the
mediating role of intrateam process and the moderating role of task type. Academy of
Management Journal, 43(2), 135–148.
Stinchcombe, A. L. (2001). When formality works: Authority and abstraction in law and
organizations. University of Chicago Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=mP2HetnX4EEC&oi=fnd&pg=PR5&dq
=(Stinchcombe,+2001)&ots=wMHquSLzBv&sig=RP_RHLqSNvZ5WCF-
7aARAM0eL3Y
Strauss, A., & Corbin, J. (1994). Grounded theory methodology. Handbook of Qualitative
Research, 273–285.
Stucky, Erin R. and Cynthia L. Kuelbs. "Chapter 14: Medical Co-Management and
Consultation." Comprehensive Pediatric Hospital Medicine. Elsevier Health Sciences,
2007.
Suddaby, R., & Greenwood, R. (2001). Colonizing knowledge: Commodification as a dynamic
of jurisdictional expansion in professional service firms. Human Relations, 54(7), 933–
953.
134

Thompson, J. D. (1967). Organizations in action: Social science bases of administrative theory.
Transaction publishers. Retrieved from
https://books.google.com/books?hl=en&lr=&id=8aNwAAAAQBAJ&oi=fnd&pg=PR1&
dq=(Thompson,+1967)+interdependence&ots=s7Q62jMTbM&sig=GDuRF0JZCQFQwr
XS0zo3jrNdIBc
Thompson, L. L., Wang, J., & Gunia, B. C. (2010). Negotiation. Annual Review of Psychology,
61, 491–515.
Thornton, P. H., Jones, C., & Kury, K. (2005). Institutional logics and institutional change in
organizations: Transformation in accounting, architecture, and publishing. Research in
the Sociology of Organizations, 23, 125–170.
Tims, M., & Bakker, A. B. (2010). Job crafting: Towards a new model of individual job
redesign. SA Journal of Industrial Psychology, 36(2), 1–9.
Treem, J. W. (2012). Communicating expertise: Knowledge performances in professional-
service firms. Communication Monographs, 79(1), 23–47.
Vicini, J., & Stempel, J. (2013). US top court upholds healthcare law in Obama triumph”.
Reuters.
Von Bertalanffy, L. (1972). The history and status of general systems theory. Academy of
Management Journal, 15(4), 407–426.
Von Hippel, E. (1990). Task partitioning: An innovation process variable. Research Policy,
19(5), 407–418.
Vough, H. (2012). Not all identifications are created equal: Exploring employee accounts for
workgroup, organizational, and professional identification. Organization Science, 23(3),
778–800.
135

Wachter, R. M., Goldman, L., & Hollander, H. (2005). Hospital medicine. Lippincott Williams
& Wilkins. Retrieved from
https://books.google.com/books?hl=en&lr=&id=5Czdw9n759cC&oi=fnd&pg=PA19&dq
=wachter+goldman+2005&ots=DJRnDysxVF&sig=IK3KZcvv3JDaDkGizEtrVVRAykk
Waring, J., & Currie, G. (2009). Managing expert knowledge: organizational challenges and
managerial futures for the UK medical profession. Organization Studies, 30(7), 755–778.
Whetten, D. A., & Aldrich, H. (1979). Organization Setsize and Diversity People-Processing
Organizations and Their Environments. Administration & Society, 11(3), 251–281.
Wholey, D. R., & Huonker, J. W. (1993). Effects of generalism and niche overlap on network
linkages among youth service agencies. Academy of Management Journal, 36(2), 349–
371.
Wilson, B. R. (1959). The pentecostalist minister: role conflicts and status contradictions.
American Journal of Sociology, 494–504.
Wright, A., Zammuto, R., & Liesch, P. (2015). Maintaining the Values of a Profession:
Institutional Work and Moral Emotions in the Emergency Department. Academy of
Management Journal, amj–2013.
Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning employees as active
crafters of their work. Academy of Management Review, 26(2), 179–201.
Wrzesniewski, A., LoBuglio, N., Dutton, J. E., & Berg, J. M. (2013). Job crafting and cultivating
positive meaning and identity in work. Advances in Positive Organizational Psychology,
1(1), 281–302.
136

Wynia, M. K., Cummins, D. S., VanGeest, J. B., & Wilson, I. B. (2000). Physician manipulation
of reimbursement rules for patients: between a rock and a hard place. Jama, 283(14),
1858–1865.
Zerubavel, Y. (1991). The politics of interpretation: Tel Hai in Israel’s collective memory. AJS
Review, 16(1-2), 133–160.
Zetka, J. R. (2003). Surgeons and the Scope. Cornell University Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=-tfrJ-
jl_aMC&oi=fnd&pg=PA1&dq=PROFESSION+jr+zetka&ots=TKDxGNoeEm&sig=W1
1lGsZ-vFgJJnKCnpdKMc1Ey_M
Zilber, T. B. (2002). Institutionalization as an interplay between actions, meanings, and actors:
The case of a rape crisis center in Israel. Academy of Management Journal, 45(1), 234–
254.

Abstract (if available)

Abstract To manage complexity, a growing number of organizations take a multidisciplinary approach, housing a variety of specialist and generalist professionals. Generalists and specialists possess overlapping knowledge areas and expertise, which complicates the way jurisdictions are claimed in the daily practice of work. Said differently, work boundaries between generalists and specialists are permeable and often overlapping, which frequently results in problems with the division of labor. This dissertation focuses on the healthcare context to highlight the challenges and mechanisms that underlie the professional division of labor, specifically in the hospital setting. Through two studies, I show how generalist physicians, called hospitalists, and specialist physicians enact their respective jurisdictions in daily practice. The first study, uses ethnographic and survey data from a children’s hospital. In this study, I describe the coordination challenges that lead to conflict over work boundaries, the factors professionals use to identify their work boundaries, and the boundary management practices professionals employ to resolve conflicts over work boundaries. The second study uses survey data collected at a children’s hospital to understand how hospitalist and specialist physicians make claims to their jurisdictions in relation to the nature of the tasks (i.e. pediatric diagnoses) they commonly encounter. The first two studies provide the springboard for future research, including a third study currently underway. This study uses electronic medical record data from a children’s hospital to analyze the performance outcomes resulting from the assignment of a hospitalist in charge of patients with certain diagnoses when a specialist should have been in charge (and vice versa). Altogether, this research has several theoretical and practical implications that aim to provide a more comprehensive understanding of the professional division of labor.

Linked assets

University of Southern California Dissertations and Theses

Asset Metadata

Creator Krikorian, Mariam Lisa (author)

Core Title Who's the physician in charge? Generalist and specialist jurisdictions in professional practice

School Marshall School of Business

Degree Doctor of Philosophy

Degree Program Business Administration

Publication Date 06/16/2016

Defense Date 05/05/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag boundary work,division of labor,ethnography,expertise,generalist,hospitalists,jurisdictional domains,knowledge management,OAI-PMH Harvest,professions,specialist,survey,task design

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Fiss, Peer (committee chair), Adler, Paul (committee member), Romley, John (committee member)

Creator Email mariamlk14@gmail.com,mkrikori@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-252056

Unique identifier UC11280456

Identifier etd-KrikorianM-4437.pdf (filename),usctheses-c40-252056 (legacy record id)

Legacy Identifier etd-KrikorianM-4437.pdf

Dmrecord 252056

Document Type Dissertation

Format application/pdf (imt)

Rights Krikorian, Mariam Lisa

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA