Background: Patient navigation (PN) has been proposed to address disparities in cancer mortality. Appalachia is largely rural with a population characterized by a high poverty rate and limited access to [ Read More ]
June 2013 VOL 4, NO 3
Spanning Boundaries: System Dynamics, Externalities, and Patient Navigation
Christopher Louis, MHA, PhD(c); Jonathan Clark, MS, PhD; Katelyn Holmes, BS; Betsy Aumiller, MEd, DEd; Eugene J. Lengerich, VMD, MS
Background: Patient navigation (PN) has emerged as a possible strategy to bridge gaps, overcome barriers, reduce delays, and improve outcomes across the cancer care continuum. However, we lack an explicit theoretical lens through which to view the emergence of patient navigators and the resulting roles they play.
Purpose: To view PN through the lens of systems theory and the related theoretical themes of boundary spanning and social networks, and, drawing on these perspectives, to evaluate their usefulness for explaining observed patterns of behavior and to present a framework for understanding PN.
Methodology/Approach: Nonprobability discriminative snowball sampling was used to identify 29 key informants in 6 contiguous Appalachian states. Our final sample consisted of 3 groups of informants: health system informants, community-based organization informants, and university-affiliated informants. We used directed content analysis to examine the extent to which patient navigator roles reflect the use of boundary spanning and social networking.
Findings: Each group of informants discussed similar themes and characteristics with relative frequency. Patient navigators were found to be spanning boundaries and using their social networks to overcome negative externalities resulting from a variety of sources. We elaborate on these findings in our theoretical framework.
Practice Implications: Our findings suggest that cancer patient navigators should build and use their social networks to span boundaries and overcome the negative externalities in the system. Our framework suggests that the need for these capabilities among navigators derives from 4 primary sources—the nature of patient problems, external influences, organizational boundaries, and individual behavior—and how they interact. Addressing the negative externalities associated with these issues reflects an ability to understand and respond to both macro- and microlevel issues across the cancer care continuum. These findings provide insights both for managers attempting to address underlying health system issues and for patient navigators who desire to be more effective.
Coordinating care; overcoming patient, provider, and health system barriers; and reducing delays across the care continuum have been long-standing challenges for healthcare managers. Recent research suggests that patient navigation (PN) has emerged as a growing organizational and community-based strategy to combat these issues.1 Prior research has conceptualized PN programs primarily in terms of the activities patient navigators engage in and the functional roles they play to ensure smooth transitions across the continuum of care.1-3 It has been postulated that the value of PN programs may be tied to their ability to fit their approach to the characteristics and complexities of local systems.2,3 This perspective implies that the emergence of patient navigators is closely tied to the characteristics and complexity of the healthcare system itself. Despite this implication, we lack an explicit theoretical lens through which to view the emergence of patient navigators and the roles they play.
Simon has argued that the way organizations are structured and organized may result in “negative externalities” (eg, miscommunication, errors, barriers, delays) that arise from the inherent interdependencies that exist between the elements of an often complex system.4 While part of the administrative task is to minimize these externalities, organizations often fall short, and healthcare organizations are no exception. Indeed, there are substantial negative externalities that exist throughout healthcare processes.5,6 Accordingly, patients may benefit from entities that can not only make better sense of the patterns of interdependence across the care continuum, but also bridge gaps and overcome the barriers arising from the complexity of the healthcare system. Such capabilities have been the object of prior research on boundary spanning7-9 and social networking.10,11
In this study, we view PN through the lens of systems theory and outline the connection between systems theory, boundary spanning, and social networking within the context of cancer care. Further, we evaluate the extent to which these perspectives characterize PN programs. Based on this exploratory evidence, we propose a framework describing both the antecedents of PN programs and the functional roles played by patient navigators.
BACKGROUND and THEORETICAL FRAMEWORK
The concept of PN was initially introduced at Harlem Hospital Center in New York City in 1990 as an intervention to reduce inequities in cancer mortality.12,13 Following promising results, which suggested that navigated patients completed a breast biopsy and received a diagnosis in a significantly shorter time than did nonnavigated patients, a large and growing number of organizations have employed PN as a strategy to combat inequities and inefficiencies seen in various types of cancer and other diseases.6 For example, Wells and colleagues found that many healthcare organizations across the United States have instituted PN to minimize delays in breast and cervical cancer diagnosis and treatment, but the paradigms employed vary greatly from site to site.1 Meanwhile, national attention has also honed in on PN, and many public (eg, Centers for Medicare & Medicaid Services; Health Resources and Services Administration) and private (eg, American Cancer Society, Avon Foundation for Women, Susan G. Komen for the Cure) funding agencies and publicly traded companies (eg, Pfizer Inc.) have allocated resources toward implementing the PN intervention model across the United States.6 Moreover, the Patient Protection and Affordable Care Act has broadly identified PN as a useful strategy to help coordinate services and provider referrals, improve access to healthcare services through community and organization partnerships, assist patients in overcoming barriers to care, and coordinate patient financial support.14 PN networks have also emerged on the national and local level, serving as conduits for communicating best practices and other successful PN strategies to broader audiences.
The Role of System Interdependence
Given that barriers to care as well as poor patient outcomes have long served as the impetus for PN programs, we begin by attempting to explain the system drivers of these problems. While other factors may be important, we will focus on issues with particular relevance to the organization and delivery of healthcare services, doing so through the lens of systems theory. Systems theory—and its practical application in systems thinking15—characterizes the complexity of a system based on the extent to which it is “dealing simultaneously with a sizable number of factors which are interrelated into an organic whole.”16 In the US delivery system, such complexity may be seen in the differentiation of providers (eg, the number of specialties and subspecialties) and provider organizations (eg, physician groups, hospitals, long-term care), and the degree to which they are interdependent. Prior research and our own observations suggest that, in health delivery systems, such interdependence derives from at least 2 categories of factors: (1) the nature of the patient’s problem to be solved,17 and (2) external influences.4,18 With respect to the latter, external influences (eg, regulatory agencies, payment systems, institutional pressures) may complicate relationships across the care continuum. As an example, administered pricing systems (eg, the diagnosis-related group system) can create payment disparities across service types, impacting the relative profitability of those services and creating financial interdependence between service providers. For example, research suggests that obstetric services are generally reimbursed at a level that makes providing these services relatively unprofitable, while the provision of cardiovascular and orthopedic care is generally profitable.19 Accordingly, the sustainable provision of obstetric care may depend on coupling it with more profitable services, such as cardiovascular or orthopedics, in the same organization. Thus, interdependence between otherwise independent service providers (in this case, obstetricians and cardiologists) may derive from influences originating outside the health delivery system.
While such interdependence is real and relevant, it may seem peripheral to the importance of the patient’s problem itself. Patient conditions are often characterized by a high degree of uncertainty and necessitate a substantial degree of interaction between a diverse set of healthcare providers and other agents, such as physicians, hospitals, and therapists. For instance, cancer manifests itself in patients in varying degrees of severity, with accompanying implications for how treatment will proceed (eg, surgery, chemotherapy, radiation therapy) and which specialists and subspecialists will be involved. Moreover, patients with cancer often suffer from multiple comorbidities, requiring coordination across several different medical specialties. The characteristics of the patient’s underlying condition define the problems to be solved, and the task of solving these problems generates interdependence between care providers across the continuum of care.
What are the organizational implications of such underlying interdependence? Simon suggests that, within such interdependent subsystems, any organizational configuration (ie, division of decision-making responsibilities and labor) “creates externalities, which arise out of the inter- dependencies among the subsystems that are ignored.”4 While the object is to minimize these negative externalities, organizations often fall short. In healthcare, these negative externalities may include anything from inconsequential miscommunication, to diagnosis and treatment barriers, to medical errors (iatrogenesis) and even death.
Simon emphasizes that, because of the underlying interdependencies in the system, the location of organizational boundaries (eg, between tasks, people, units, and organizations) plays a defining role in the extent of the externalities observed in the system.4 In addition to organizational configurations, individual behavior can also serve to exacerbate the boundaries within the system.20 Even under the impossible scenario whereby the perfect organizational configuration exists, behavioral inconsistencies by otherwise interdependent providers may result in poor coordination of care and barriers to diagnosis and treatment. In short, boundaries placed between otherwise interdependent subsystems—whether by configuration or individual behavior—may compromise the effectiveness and efficiency of their coordinated activities.
It is no wonder that so much of the US healthcare debate revolves around the location of organizational boundaries, including the extent of specialization21,22 and the degree of system integration.23,24 These debates are fueled by the negative externalities in the system (eg, problems related to cost, quality, access). These externalities—in part derived from the misplacement of organizational bound- aries and in part by individual behavior—may ultimately provide an explanation for why PN programs exist (ie, PN as a boundary-spanning mechanism) and the capabilities we might expect to find in effective patient navigators.
Consistent with the Simon perspective on complex systems, empirical work suggests that, as system complexity and externalities increase, boundary-spanning behaviors become more prevalent.4,25 This idea is intuitive—where there are more boundaries, more boundary spanning will occur. Boundary spanning has been described as a bridging of intra- and interorganizational boundaries “by individuals who understand…both sides of the boundary, enabling them to search out information on one side and disseminate it on the other.”18 Such boundary-spanning behavior is reflected in a study by Parker and Lemak, who suggest that patient navigators play an important role in communicating information that is then used to help accomplish individual- and system-level goals.3 Indeed, such boundary-spanning behavior has been shown to improve organizational performance.8,9
Boundary-spanning roles typically incorporate the following 3 characteristics: (1) information selection, processing, translation, and transfer across organizational boundaries, (2) external representation and communication, and (3) internal representation and communication.7,18,26 The first characteristic—information selection, processing, translation, and transfer—refers to the full spectrum of cognitive and relational aspects involved in creating knowledge from information and the ability to communicate that knowledge across organizational boundaries.26 The second characteristic—external representation and communication—describes “all boundary roles that involve resource acquisition and disposal, political legitimacy and hegemony, and a residual category of social legitimacy and organizational image” among external agents,26 whereas the third characteristic—internal representation and communication—reflects similar boundary roles within one’s own organization.7,18,26
Research suggests that individuals in boundary-spanning roles, such as patient navigators, use their social networks as a conduit for communicating information across internal and external organizational boundaries, as described above.27 Organizations with agents who have networks that bridge structural holes (ie, separations between non- redundant contacts) have been found to learn faster and adapt more quickly to changing circumstances.11 Conversely, boundary-spanning activities have also been linked to overcoming structural holes within social networks.28 Social networks are, in themselves, subsystems and have been referred to as a sophisticated “…system of objects (eg, people, groups, organizations) joined by a variety of relationships.”29 Implicit in this definition is the idea that social networks may span all sides of organizational boundaries. The management of such relationships is essential in enabling the transfer of knowledge between agents30 and dealing with the complexities and interdependencies inherent in organizations.
Social networks have also been tied to individual- and group-level performance.31 Inkpen and Tsang found that establishing more intimate, stable, and long-lasting relationships among agents within a network facilitates knowledge transfer across boundaries, a key component of organizational success.32 Moreover, Parker and colleagues suggest that patient navigators are engaged in a range of tasks and behaviors with a variety of agents in their social networks.2 For example, patient navigators were routinely found to be communicating and integrating information across organizational boundaries between internal and external agents.2
Behavior within social networks typically relates to the following 4 characteristics: (1) network size, or number of contacts in a network, (2) range or diversity of contacts, (3) scope of contacts, such as contacts in a single industry or social group, and (4) stability of contacts, which reflects the strength of ties, or the “closeness of social ties,” between agents in a network.2,10,33 To this list of network characteristics we add agent behaviors, which relate to “how” agents use their relationships within social networks.2
In the sections that follow, we examine the extent to which the characteristics of boundary spanning and social networks outlined above are observed in PN programs. In doing so, we ask: Do patient navigators serve as integrators, using their social networks to span boundaries and overcome the negative externalities of a complex health system? Or, do patient navigators “simply” represent mechanisms for managing care, ensuring proper utilization of services, and encouraging patient compliance? Although extremely useful, such a role suggests a narrower set of capabilities for PN. In contrast, the theory we have presented suggests a wider-ranging role for PN that specifically necessitates building and capitalizing on social networks that span provider and organizational boundaries to ensure effective and efficient care for patients with cancer.
We examined the emergence of the themes presented in our theoretical framework in interviews about PN among healthcare providers in Appalachia. Appalachia comprises 420 counties in 13 states, ranging from southern New York to northern Mississippi. This primarily rural, mountainous region is characterized as having high cancer mortality rates.34,35 Moreover, access to healthcare services in Appalachia is hindered by a complex, fragmented infrastructure,36 inadequate cancer-control strategies,37 and a shortage of healthcare professionals.38 For example, there are only 2 National Cancer Institute–designated comprehensive cancer centers in Appalachia, and many hospitals are small community hospitals that contract with independent nonstaff physicians. In addition, prior research has also suggested that Appalachian residents are unique: Puckett found that people living in Appalachia are reliant on self, family, and friends, and have trusted relationships with people they know rather than with authority figures and highly trained professionals.39 The challenges faced by healthcare providers in Appalachia have spurred the need for a strategy that both effectively manages cancer care processes and understands the needs of local communities.
We used nonprobability discriminative snowball sampling to identify key informants. In all, we interviewed 29 key informants who reside in 6 contiguous Appalachian states (Kentucky, New York, Ohio, Pennsylvania, Virginia, and West Virginia), and each informant represented 1 of 3 groups: (1) health system informants (n=12), (2) community-based organization informants (n=10), or (3) university-affiliated informants (n=7). Table 1 summarizes key informants by group. Informants within each group represented organizations with and without PN programs in place. Informants from organizations with PN programs were either patient navigators themselves or individuals who regularly interacted with the navigator. Informants at sites without PN programs were interviewed because their institution was contemplating a PN program or because they were knowledgeable about PN activity in their local health system. These individuals were included in our sample to provide an objective perspective different from that of those actively performing PN.
Three study investigators conducted audiotaped, semistructured telephone interviews between April and October 2010. Interviews began with a grand tour question broadly addressing PN programs in the informant’s community. Subsequent questions focused on various themes, including the patient navigator’s background and experience, job roles performed by patient navigators, barriers for patient navigators, strategies employed to overcome barriers, community outreach, communication, and champions and gatekeepers of PN programs. Interviews were conducted in an open-ended manner and were not specifically focused on the themes presented in this article; specifically, the terms boundary spanning and social network (and their related themes) were not explicitly addressed by interviewers. The average length of the interviews was approximately 60 minutes. Interview protocols are available from the authors on request.
We used directed content analysis to analyze the interview data for boundary-spanning and social-network themes. This method of qualitative data analysis “starts with a theory or relevant research findings as guidance for initial codes.”40 The analysis process consisted of several steps to ensure reliability and validity. First, we created a predetermined coding scheme based on the 8 key characteristics
of boundary spanning and social networks (Table 2).2,7,10,18,26,29,33 Next, 2 investigators independently used the predetermined coding scheme to code 5 transcripts. In subsequent discussions, they reconciled ambiguous passages or where there was disagreement on codes applied to the interview text and clarified the coding scheme as needed. Then, the same 2 investigators independently coded each of the remaining 24 transcripts. Transcripts were coded using NVivo 9 (QSR International), a computer software package that supports qualitative and mixed methods research. This approach produced an acceptable level of interrater reliability, with Cohen’s k = .82.41 Each investigator prepared a short memorandum summarizing his or her findings across the themes.42 In total, more than 1900 coded units were applied to the text of the 29 transcripts. Finally, we performed t-tests to identify statistical significance across informant groups for each characteristic.
Table 3 summarizes the extent to which our interviews across the 3 groups of informants reflected the relevant conceptual characteristics and also identifies statistically significant differences between informant groups. For example, column 9 shows the frequency with which university-affiliated informants discussed information selection, processing, translation, and transfer; external representation and communication; and range of contacts and how the frequency significantly differs from informants from community-based organizations. However, the magnitude of this difference was not substantial. Even in the most statistically significant scenario (eg, information selection, processing, translation, and transfer) where P <.01, university-affiliated informants referenced this characteristic only 2 more times for every 10 pages of interview text than did informants in community-based organizations. While these differences might reflect differences in professional emphasis, we are confident that they do not represent systematic divergence in the observed roles of patient navigators. In general, Table 3 shows that, when 1 informant group focused on a given characteristic, the other 2 groups focused on the same characteristic with approximately the same frequency. Thus, we present our core findings according to the most frequently discussed characteristics across all 3 groups, rather than distinguishing between informant groups.
Table 4 presents the overall frequency (from most frequent to least frequent) with which conceptual characteristics were reflected in our informant interviews. From this table, we observe that the top 2 characteristics reflect both boundary spanning (information selection, processing, translation, and transfer) and social networking (agent behaviors), suggesting a close relationship between the 2 in practice. Moreover, while some characteristics appear more frequently than others, in total the overall frequency with which boundary-spanning and social-networking characteristics were reflected in our interviews was substantial, representing more than 24 mentions for every 10 pages, or nearly 2.5 mentions per page of transcript. We believe this provides preliminary evidence for the relevance of our framework in that the characteristics of boundary spanning and social networking appear to be highly prevalent in cancer PN. To illustrate further, we provide below more descriptive detail from our interviews for the top 4 characteristics: (1) agent behaviors, (2) information selection, processing, translation, and transfer, (3) scope and stability of contacts, and (4) external representation and communication. We also include findings from other characteristics as relevant.
1. Agent Behaviors (Social Networking)
During the interviews, informants discussed agent behaviors more frequently than any other characteristic (4.49 coded units per 10 pages of interview text). Patient navigators were commonly found broadening and using their social networks to overcome barriers to cancer services (ie, transportation, health insurance), to bridge gaps in the healthcare system (ie, scheduling appointments for cancer services at different organizations, coordinating care), and, as one informant stated, “…[to] help navigate patients through the fragmentation” (Informant 19). Informants described various approaches navigators use to perform this role among internal and external agents. The following quote illustrates how one patient navigator initiated and fostered a strong relationship with an external agent:
“Well, I will tell you this: that the most important and the best thing that I ever did when I got this job [as a patient navigator] is I went over to the Social Security office… and I did a presentation to them and told them what I did.…As a result of that, two of the workers over at the Social Security office, we formed a kind of a bond, and we got a system going so that all I had to do is, when I get a [patient] in my office, I pick up the telephone…they do the interview over the phone…and, bam, it is done and we have done it and gotten patients their [Medicaid] cards within two weeks.” (Informant 26)
Patient navigators were also found to have learned detailed information about patient circumstances and operational processes so that they may facilitate patients attending scheduled appointments. For example, one informant noted how she works with patients as follows:
“…you also run into coordination-of-care issues like helping the patient remember doctor appointments, or schedule doctor appointments, or schedule a scan. …Sometimes, I will make the appointment for them, or if they are those [types of] patients who have trouble remembering, I will just call them the day before and just say, ‘Don’t forget you have your test or your doctor’s appointment with whomever tomorrow at such and such time.’” (Informant 11)
Another informant described how the patient navigator at her organization leverages her long-standing relationships with internal agents: “So if [patients] give [the patient navigator] permission, then she will assist in either ordering additional testing, getting them set up to see a surgeon, or…whatever the patient needs” (Informant 16). Similarly, another informant echoed these sentiments by saying, “…if someone comes in and says they don’t have a PCP for some reason…the navigator will work with them to find someone who is accepting patients that the report can be sent to” (Informant 15). Furthermore, the patient navigator’s connections to alternative healthcare providers (ie, physicians, home health agencies, hospice, etc) provide patients with direct links to resources that they may not have known about. One informant discussed the importance of connecting patients with palliative care providers:
“…we have pulled together a circle-of-care team…it is increasing the awareness to palliative care and can be for anybody who is going through cancer treatment and isn’t feeling good. And so Dr. B and I have had this discussion and I have started referring people to her…and that has been really successful.” (Informant 27)
2. Information Selection, Processing, Translation, and Transfer (Boundary Spanning)
Interviews revealed how patient navigators often facilitated the role of information selection, processing, translation, and transfer (4.13 coded units per 10 pages of transcripts). As reflected in the quotes about agent behaviors above, informants frequently described the patient navigator’s role in gathering, synthesizing, and disseminating cancer-related information, which was used to improve the quality of care delivered to patients with cancer. For example, one informant stated,
“…I maintain membership within the Oncology Nursing Society, and I am the one that then brings that information back here and trains the nurses, also lining up oncology in-services that benefit our IV clinic or nurses on our floors. Most of the time I find those [in-services] myself, but I also pick up on leads from them on what they feel like they need help with. So I kind of become their consultative person.” (Informant 12)
Similarly, another patient navigator discussed how she joined a statewide networking group to gain access to new information and resources. In turn, she compared statewide initiatives with her organization’s practices in seeking to identify opportunities for improvement:
“…Pfizer drug company had sponsored a navigator workshop, it was one of the first in the state.…I went and learned about a statewide initiative….We then got everyone into a networking group…they wanted to share resources…to find out well, what are you doing? Maybe I should be doing something better [at my organization]….” (Informant 7)
Informants also identified patient navigators as impor- tant links between patients with cancer and their care providers. During a time when patients are inundated with new, cancer-specific information, patient navigators not only help patients with cancer formulate key questions to ask but also teach them how to interpret the answers. For example, one informant noted,
“…I tell everybody that [patient navigators] are the missing link. They are that link that the patients themselves can actually ask, ‘You know, I didn’t understand what the nurse said, I didn’t understand what the doctor said to me today, and I was afraid to ask the question.’ What the navigator can do is help them form those questions and actually write them down for them, and, if necessary, attend the next appointment with them to be able to talk to the provider, because a lot of times…the blood pressure goes sky high when they come in and they forget what they need to ask…so I feel like there is a link that helps [patients] to have the knowledge that they need to take care of themselves.” (Informant 13)
Likewise, interviews reflected how patient navigators were essential in integrating patient needs with knowledge of community resources. The ability to (1) extract specific patient needs, (2) interpret how those needs can be met based on the availability of existing community resources, and (3) know how to quickly link patients with those resources were identified as valuable skills of the navigator. One informant supported this by saying,
“I think [patient navigators] definitely, definitely, definitely have to have a knowledge of the resources in the area, more than just the doctors and hospitals available…I don’t believe the navigator has to know all the resources that are available, but how to link them. So there should be a history or experience in working in the community services, an awareness of the resources in the area, because something can come up that has nothing to do with a cancer diagnosis that could prevent the patient from proceeding to the next step…the [patient navigator] needs to be beyond the medical and know the resources for the area.” (Informant 18)
3. Scope and Stability of Contacts (Social Networking)
Interviews reflected the scope of contacts that patient navigators had within different industries or social groups (3.98 coded units per 10 pages of interview text). Those contacts most commonly referenced were patients with cancer themselves, along with healthcare organizations and professionals, community-based organizations, churches and other faith-based populations and organizations, and funding agencies. For example, one patient navigator noted how she developed a relationship with the members of local faith-based populations:
“…the last three years I have worked with the Amish and Mennonite population…it took us this long to make inroads, but…we worked very hard to establish a trust and we have been transparent, we have been completely open with them, we promised we would not cross any lines that they did not feel comfortable with us doing. So we pretty much have operated within the cultural realm and not overstepped our boundaries.” (Informant 27)
Many informants also emphasized the importance of strengthening their relationships (stability of contacts) with various agents in their network (2.75 coded units per 10 pages of interview text). Securing trusting relationships with physicians was a common theme in achieving PN program sustainability. To develop stronger ties with the hospital’s medical staff, one patient navigator noted,
“…at first no physician wanted to be involved…[the physician] felt like…if he recommended a mastectomy, that the navigator was going to say [to the patient], ‘Well, do you know about a lumpectomy or do you know you can go have plastic surgery out of town….’ So, we had a huge resistance….We didn’t overcome it until we got a surgeon…[who] had been part of other patient navigation programs, and he started just working with our navigator and building…a rapport. And now, he’s been here for…2 and a half to three years and has about 80% of the breast volume at our center. That’s why [the patient navigators] now see 80% of the breast patients.” (Informant 16)
Moreover, several informants also described how strengthening relationships with internal agents through participation in and validation of hospital committees was vital to the long-term success of PN programs (Informants 11, 15, 16, 19, 24, and 25). While this agent behavior falls under the category of social networking, it also crosses over to internal representation and communication, a key boundary-
spanning characteristic, and was described as being an important capability in fostering these relationships. For example, one informant noted,
“…I sit on a lot of committees here at the hospital that have nothing to do with my job, that are more like employee-based committees…—like our service-excellence team, our community-benefit team—just to kind of keep my face in front of other employees. Because when you are the second-largest employer in the county, your employees know other people who have cancer…so if I am sitting in a meeting across from someone in our accounting office and they say, ‘You know what? I should really have my aunt talk to you, because she was just diagnosed and what a great resource you’d be.’” (Informant 11)
4. External Representation and Communication (Boundary Spanning)
Interviews revealed external representation and communication as a primary mechanism for maintaining a presence in the community and disseminating information to and from external agents (3.40 coded units per 10 pages of interview text). Informants described how engagement and education of external agents, especially local physicians and their office staffs, are paramount to sustaining the PN program. One informant described his organization’s strategy in maintaining visibility in physician offices:
“One of the things we ask [our patient navigator] to do is visit at least five physician offices per month… And we want her to…talk to the office managers, practice managers, physicians—anyone who will listen about what she is trying to do and what she wants to do.” (Informant 16)
Similarly, another patient navigator discussed how physicians were initially reluctant to embrace the PN program at one hospital because they did not understand its purpose. However, after the navigator collaborated with the medical staff leadership, the community doctors changed their perspective. She described this scenario by acknowledging,
“[Community physicians] didn’t want me talking to patients…they didn’t understand what I would be doing, and that’s where the doctors who are the champions came in, and they really had to explain the process of patient navigation….I also did a presentation at a medical staff meeting for over a hundred physicians, and Dr. G. also presented with me so [community physicians] knew what patient navigation was all about and the value of it to their patients and their practices….I think when they saw me presenting with Dr. G., they knew that it was going to be okay.…” (Informant 24)
Several informants also discussed the importance of attending and/or participating in community and general education programs, such as health fairs and speaking engagements (Informants 8, 9, 15, 16, 19, and 25), community healthcare coalitions (Informants 12 and 16), local organization board membership (Informant 11), church groups (Informant 24), and local planning committees (Informant 12). Furthermore, participation in local and national PN network webinars, conferences, and other meeting groups was believed to be a growing avenue in communicating best practices across navigators (Informant 7).
Summary of Findings
These findings support the idea that patient navigators play a wide-ranging, integrating role, using their social networks to span boundaries in an effort to overcome the negative externalities in the cancer care system. We have argued that these negative externalities are the result of at least 4 key factors involving provider interdependence. This interdependence—which is determined by (1) the patient’s problem to be solved and (2) external influences, and interacts with (3) organizational configurations, and (4) individual provider behavior—helps characterize the extent to which boundaries and barriers exist in the system. Our interviews reflected many of these factors. We include a few examples here to emphasize the important role these factors play in defining the need for PN and the boundary-spanning role patient navigators play.
One informant described the nature of the patient problems their regional health system deals with as follows:
“…a lot of our people in this region put off healthcare until they can’t anymore, so they don’t access healthcare until they are symptomatic….After they have been diagnosed with cervical cancer…we also find out they have obesity and heart disease and diabetes.” (Informant 2)
Patient needs were also a common theme. One patient navigator described how patient needs can complicate things (even for navigators):
“So, [the radiation oncologist] brought [a rectal cancer patient] in to me and she left and I said [to the patient] ‘Okay, the doctor told me that she has given you prescriptions.’ He gets in his wallet and he pulls it out, and I bet you there were ten prescriptions in his wallet. [He] hands them to me and says he doesn’t know what to do with these….I typed up very detailed [instructions] on this paper how to get to [the pharmacy] in his area, and he lives right there….I get a call from the group that I was sending him to…and [they] said this man was found in the courthouse clutching your directions….She said everyone in the community knows this family and they are illiterate. Here I had taken all that time and didn’t ask the right questions and it never occurred to me….” (Informant 26)
Another informant indicated that low Medicaid reimbursement rates can affect access to care in their area:
“…one of our doctors, an oncologist here, does not take Medicaid, any Medicaid” (Informant 24). This reflects how both external influences and individual provider behavior can influence patient experiences. Finally, another informant revealed that fragmented organizational configurations still exist, despite the collective interdependence between the providers:
“Well, we are right now a little fragmented….We are located in some different locations….We are kind of the old structure; we have got medical oncology offices here, which are independent; and you have got radiation oncology, which is a hospital-based department; and we have a new medical oncologist who is coming in to help us build our program, which will also refer a secondary oncologist to work with that person.” (Informant 19)
Together, these examples support the idea that the need for patient navigation (ie, due to negative externalities) emerges from the interdependence in the system and how PN interacts with provider configurations and behavior.4
DISCUSSION and PRACTICE IMPLICATIONS
The theory and evidence we have presented suggest a strong link between the structure of systems and the use of social networks to span organizational boundaries. Our exploratory findings suggest that patient navigators serve an integrating role, building and using their social networks to span these boundaries and overcome barriers to care. Our theory suggests that such behavior results from negative externalities that emerge from system interdependencies. The Figure depicts these relationships, providing a framework that reflects both why PN emerges (negative externalities) and the roles we might expect patient navigators to play (boundary spanning within social networks).4,43 Specifically, our model suggests that a given health system (in this case, the cancer care system)4,43 may be characterized by a degree of provider interdependence that emerges both from the environment (external influences) and the object of the system’s existence (the patient’s problem to be solved). Organizational configurations (eg, differentiation and integration among providers) and individual provider behaviors (eg, their choices and interactions with others) interact with these interdependencies to produce externalities, both positive and negative. Negative externalities specifically result when organizational configurations and provider behaviors conflict with the underlying interdependencies (eg, the needs of the patient),44 thus giving rise to the need for boundary spanning. Based on our findings, our model further proposes that boundary spanners (ie, patient navigators) must work to build and capitalize on their social networks to overcome these externalities, theoretically generating positive outcomes for patients.
This framework makes 2 important contributions to our understanding of health systems and the emergence of PN. First, drawing on systems theory, we provide formal insight into how organizational configurations and individual provider behaviors contribute to the problems plaguing the industry. Our framework specifies the broad sources of complexity (interdependence, externalities) observed in the healthcare system—(1) the patient’s problem to be solved, (2) external influences, (3) organizational configurations, and (4) individual provider behavior—and how exogenous factors (1 and 2) interact with endogenous organizational behavior (3 and 4) to produce the negative outcomes so often observed in the health delivery system. This aspect of our framework is consistent with recent empirical evidence suggesting that the efficiency of certain organizational configurations depends on the nature of the patient problems being solved.17
Second, our framework presents PN as an artifact of system dynamics. We specifically integrate theories on social networking and boundary spanning to describe the roles and activities of patient navigators, reflecting an understanding of system dynamics at both a macro level (external influences, organizational configurations) and a micro level (patient needs, individual provider behavior, provider interdependence). Systems thinking has been described as “the art of seeing the forest and the trees” and the ability to “act locally and think globally.”15,45 At a macrolevel, patient navigators engage in information selection, processing, translation, and transfer to make sense of and adapt to changes in the broader healthcare landscape, such as changes in payment systems, for example. In doing so, patient navigators often face microlevel challenges associated with the complex relationships between a variety of internal and external agents. Our interviews suggest that navigators approach these challenges through their awareness of agent behaviors and their use of external representation and communication, and that their ability to understand and manipulate their social networks is essential in helping their patients overcome the barriers inherent in the cancer care system.
Similar to other exploratory studies of this nature, our study is not without its limitations. First, our sample was small and focused on Appalachia, which may limit the generalizations that may be drawn from the study. Nonetheless, our sample included a broad range of regions within Appalachia and an array of individuals working in health systems with varying characteristics. Although Appalachia’s substantial rural population, documented high cancer mortality rates, and shortage of healthcare professionals indicate complex circumstances throughout the region, these complicated circumstances are not necessarily unique to this region.
While the specifics may differ across regions, we have no reason to believe that the general antecedents prompting the emergence of PN in Appalachia differ greatly from those of other regions of the United States.
Second, our study was not designed for causal inference, despite our framework specifying causal directions. Given that the intent of this research was to be exploratory and propositional in nature, future research is needed to further understand the relationships displayed in our framework. Finally, because our interview sample included nonnavigators, information bias may be present. It is possible that informants who are not actively participating in navigation duties and are only in observational roles may differ in their perceptions of what navigators do, leading to systematically different responses to similar interview questions. However, the results presented in Table 3 suggest that the magnitude of differences across the informant categories is small. For example, the university-affiliated informant group, which included no navigators, made slightly more references to the characteristics of boundary spanning and social networking than did the other informant groups in 5 of the 8 categories (per 10 pages of text). While these differences may reflect a PN knowledge base by individuals outside of those routinely performing PN, we believe that their inclusion may, in fact, provide a more objective perspective on these programs.
Despite these limitations, we believe that, by specifying the antecedents of PN, our study may offer valuable insights for managers and navigators regarding the role of cancer PN in health delivery systems. The important social-networking and boundary-spanning capabilities patient navigators need to traverse the barriers in the system reflect systems thinking and underscore the relevance of understanding “big picture” issues, aggressively managing social networks, and seeking to bridge gaps between interdependent providers, in addition to performing the tactical duties required of patient navigators.3 These capabilities encompass a broader view of the patient navigator role, a more relational role, which involves the mastery of both macro- and microlevel capabilities. These conclusions have clear implications for selecting and training patient navigators. Despite these implications, however, informants described how current hiring practices are based primarily on less sophisticated means (eg, training, education, and a candidate’s general interest level). One informant acknowledged:
“…the people that have been historically hired into these [patient navigator] positions were just put there because they have a general interest….Then the organizations reevaluate and see what problems they were having operationally and fix them from there….That’s how we’ve evolved to focus on the things we do….[Instead] organizations should think strategically and hire based on the capabilities of the person.” (Informant 1)
In moving toward a more focused, capabilities-based hiring process, delivery systems may experience a better fit between the nature of the task and the navigator’s capabilities.
Our framework also provides insight for system leaders about system dynamics and the sources of negative externalities observed in the system, providing leaders with a model for thinking about how to address the underlying causes. Indeed, patient navigation may be viewed as a transaction cost that arises because existing, fragmented organizational configurations leave to markets what may be done more efficiently within organizations.45 The same observation may be made about other emerging innovations aimed at improving coordination of care and overcoming barriers (eg, case management). From a system dynamics perspective, these innovations emerge because organizations fail to account for the interdependence between providers. Our framework suggests that greater attention to system dynamics on the part of leaders—and more specifically, greater attention to the nature of patient problems and patient needs46—may lead to improvements in organizational configurations and organizational behavior.
Acknowledgments: This research was funded by the Appalachia Community Cancer Network (ACCN) (3U01 CA114622-S1), and technical assistance was provided by the Community Sciences and Health Outcomes (CSHO) Core of the Penn State Hershey Cancer Institute. The authors wish to thank the project managers of ACCN and other key informants at our field sites who were interviewed for this study. We also wish to thank Diane Brannon, PhD, Marianne Hillemeier, PhD, and Barbara Gray, PhD, for their contributions on previous iterations of this manuscript. This study was approved by the Scientific Review Committee of the Penn State Hershey Cancer Institute and the Milton S. Hershey Medical Center (HMC) Institutional Review Board. All interviewees agreed to the terms of the informed consent prior to participating in this research.
Disclosures: All authors report nothing to disclose.
Corresponding Author: Christopher Louis, MHA, PhD(c), Department of Health Policy and Administration, The Pennsylvania State University, 504A Ford Building, University Park, PA 16802; firstname.lastname@example.org.
1. Wells KJ, Battaglia TA, Dudley DJ, et al. Patient navigation: state of the art or is it science? Cancer. 2008;113(8):1999-2010.
2. Parker VA, Clark JA, Leyson J, et al. Patient navigation: development of a protocol for describing what navigators do. Health Serv Res. 2010;45(2):514-531.
3. Parker VA, Lemak CH. Navigating patient navigation: crossing health services research and clinical boundaries. Adv Health Care Manage. 2011;
4. Simon HA. Applying information technology to organization design. Public Adm Rev. 1973;33(3):268-278.
5. Vargas RB, Ryan GW, Jackson CA, Rodriguez R, Freeman HP. Characteristics of the original patient navigation programs to reduce disparities in the diagnosis and treatment of breast cancer. Cancer. 2008;113(2):426-433.
6. Paskett ED, Harrop JP, Wells KJ. Patient navigation: an update on the state of the science. CA Cancer J Clin. 2011;61(4):237-249.
7. Tushman ML. Special boundary roles in the innovation process. Adm Sci Q. 1977;22(4):587-605.
8. Ancona DG, Caldwell DF. Bridging the boundary: external activity and performance in organizational teams. Adm Sci Q. 1992;37(4):634-665.
9. Tsai W. Social capital, strategic relatedness and the formation of intraorganizational linkages. Strategic Manage J. 2000;21(9):925-939.
10. Shortell SM, Rundall TG. Physician-organization relationships: social networks and strategic intent. In: Mick SS, Wyttenbach ME, eds. Advances of Health Care Organization Theory. San Francisco, CA: Jossey-Bass; 2003:141-174.
11. Burt RS. Structural holes and good ideas. Am J Sociol. 2004;110(2):349-399.
12. Freeman HP. A model patient navigation program. Oncol Issues. 2004; (19):44-47.
13. Freeman HP. Patient navigation: a community based strategy to reduce cancer disparities. J Urban Health. 2006;83(2):139-141.
14. Patient Protection and Affordable Care Act. HR 3590, 111th Cong, 2nd Sess (2009-2010). www.govtrack.us/congress/bills/111/hr3590/text. Accessed February 21, 2012.
15. Senge PM. The Fifth Discipline: The Art & Practice of the Learning Organization. New York, NY: Doubleday; 1990.
16. Weaver W. Science and complexity. Am Sci. 1948;36(4):536-544.
17. Clark JR. Comorbidity and the limitations of volume and focus as organizing principles. Med Care Res Rev. 2012;69(1):83-102.
18. Tushman ML, Scanlan TJ. Boundary spanning individuals: their role
in information-transfer and their antecedents. Acad Manage J. 1981;24 (2):289-305.
19. Horwitz JR. Making profits and providing care: comparing nonprofit, for-profit, and government hospitals. Health Aff (Millwood). 2005;24(3):790-801.
20. Ashkenas R. Simplicity-minded management: a practical guide to stripping complexity out of your organization. Harv Bus Rev. 2007;85(12):101-109, 146.
21. Herzlinger R. Market-Driven Health Care: Who Wins, Who Loses in the Transformation of America’s Largest Service Industry. New York, NY: Perseus Books; 1997.
22. Porter M, Teisberg E. Redefining Health Care: Creating Value-Based Competition on Results. Boston, MA: Harvard Business School Press; 2006.
23. Enthoven AC, Tollen LA. Competition in health care: it takes systems to pursue quality and efficiency. Health Aff (Millwood). (Suppl): 2005:W5-420–W5-433.
24. Enthoven AC, Crosson FJ, Shortell SM. ‘Redefining health care’: medical homes or archipelagos to navigate? Health Aff (Millwood). 2007; 26(5):1366-1372.
25. Dollinger MJ. Environmental boundary spanning and information processing effects on organizational performance. Acad Manage J. 1984;27(2):351-368.
26. Aldrich H, Herker D. Boundary spanning roles and organization structure. Acad Manage Rev. 1977;2(2):217-230.
27. Manev IM, Stevenson WB. Balancing ties: boundary spanning and information and influence in the organization’s extended network of communication. J Bus Commun. 2001;38(2):183-205.
28. Friedkin NE. Information flow through strong and weak ties in intraorganizational social networks. Social Networks. 1982;3:273-285.
29. Tichy NM, Tushman ML, Fombrun C. Social network analysis for organizations. Acad Manage Rev. 1979;4(4):507-519.
30. Reagans R, McEvily B. Network structure and knowledge transfer: the effects of cohesion and range. Administrative Sci Q. 2003;48(2):240-267.
31. Sparrowe RT, Liden RC, Wayne SJ, Kraimer ML. Social networks and the performance of individuals and groups. Acad Manag J. 2001;44(2):316-325.
32. Inkpen AC, Tsang EWK. Social capital, networks, and knowledge transfer. Acad Manage Rev. 2005;30(1):146-165.
33. Burt RS. Toward a Structural Theory of Action: Network Models of Social Structure, Perception, and Action. New York, NY: Academic Press; 1982.
34. Centers for Disease Control and Prevention (CDC). Cancer death rates—Appalachia, 1994-1998. MMWR Morb Mortal Wkly Rep. 2002; 51(24):527-529.
35. Yao N, Lengerich EJ, Hillemeier MM. Breast cancer mortality in Appalachia: reversing patterns of disparity over time. J Health Care Poor Underserved. 2012;23(2):715-725.
36. Strover S, Oden M, Inagaki N. Telecommunications and rural economies: findings from the Appalachian region. Paper presented at: 29th Research Conference on Communication, Information and Internet Policy (TPRC); October 27-29, 2001; Alexandria, VA. http://arxiv.org/ftp/cs/pa pers/ 0109/0109090.pdf. Accessed May 10, 2013.
37. Ward AJ, Coffey Kluhsman B, Lengerich EJ, Piccinin AM. The impact of cancer coalitions on the dissemination of colorectal cancer materials to community organizations in rural Appalachia. Prev Chronic Dis. 2006;3(2):1-13.
38. Behringer B, Friedell G, Dorgan K, et al. Understanding the challenges of reducing cancer in Appalachia: addressing a place-based health disparity population. Cal J Health Promot. 2007;5(special issue):40-49.
39. Puckett A. “Let the girls do the spelling and Dan will do the shooting”: literacy, the division of labor, and identity in a rural Appalachian community. Anthropol Q. 1992;65(3):137-147.
40. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277-1288.
41. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20:37-46.
42. Charmaz K. Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis. Thousand Oaks, CA: Sage Publications; 2006.
43. Bar-yam Y. Dynamics of Complex Systems. Boulder, CO: Westview Press; 1997.
44. Clark J, Singer S, Kane N, Valentine M. From striving to thriving: systems thinking, strategy, and the performance of safety net hospitals [published online ahead of print May 25, 2012]. Health Care Manage Rev. doi: 10.1097/HMR.0b013e31825ba9ab.
45. Williamson OE. Strategy research: governance and competence perspectives. Strategic Manage J. 1999;20(12):1087-1108.
46. Hernandez SE, Conrad DA, Marcus-Smith MS, Reed P, Watts C. Patient-centered innovation in health care organizations: a conceptual framework and case study application. Health Care Manage Rev. 2013;38(2):166-175.