Achieving Coordinated Care for Patients With Complex Cases of Cancer: A Multiteam System Approach

Abstract

Patients with cancer with multiple chronic conditions pose a unique challenge to how primary care and specialty care teams provide well-coordinated, patient-centered care. Effectiveness of these care teams in providing optimal health care depends on the extent to which they coordinate their goals and knowledge as components of a multiteam system (MTS). This article outlines challenges of care coordination in the context of an MTS, illustrated through the care experience of “Mr Fuentes,” a patient in the Dallas County integrated safety-net system, Parkland. As a continuing patient with chronic illnesses, the patient being discussed is managed through one of the Parkland community-oriented primary care clinics. However, a cancer diagnosis triggered an additional need for augmented coordination between his different provider teams. Further research and practice should investigate the relationships of MTS coordination for shared care management, transfer to and from specialty care, treatment compliance, barriers to care, and health outcomes of chronic comorbid conditions, as well as cancer control and surveillance.

CASE SUMMARY

Mr Fuentes (a pseudonym) is a 63-year-old construction worker with type II diabetes mellitus, mild renal insufficiency, and mild congestive heart failure (CHF) who sees his primary care physician (PCP; Dr A) two to three times a year through a Parkland community-oriented primary care clinic. In 2014, he experienced bleeding per rectum, change in bowel habits, and some unexplained weight loss. Dr A suspected rectal cancer and initiated a referral for a GI evaluation. It took 4 weeks to obtain a colonoscopy, which identified an obvious rectal cancer. The patient was then referred to a medical oncologist (Dr C).

Dr C obtained staging studies suggesting the cancer was localized and potentially curable. Dr C outlined a multidisciplinary treatment approach to include preoperative chemotherapy concurrent with radiation therapy (neoadjuvant therapy), followed by surgical resection; then, the patient would complete a course of chemotherapy (adjuvant therapy). Input from colorectal surgery (Dr S) and radiation oncology (Dr R) was necessary to initiate this plan. More delays to the initiation of therapy were incurred as a result of sequential scheduling of surgery and radiation therapy visits. Eight weeks after seeking advice from Dr A, Mr Fuentes began chemotherapy with oral capecitabine administered concurrently with radiation therapy. He completed 5 weeks of chemoradiotherapy with minimal toxicities and then underwent surgery. Pathology from surgery found residual cancer and two of 12 lymph nodes with metastatic carcinoma. The postoperative stage was ypT3N1M0 stage III rectal cancer.

Four weeks after surgery, Mr Fuentes was seen by Dr C to initiate the planned additional chemotherapy but needed a mediport, which was then placed by surgery in the operating room because of scheduling delays with interventional radiology. Even with the port, delays occurred in getting the patient scheduled for chemotherapy. Eight weeks postsurgery, Mr Fuentes began his adjuvant FOLFOX (infusional fluorouracil, leucovorin, and oxaliplatin) chemotherapy. Initially, he tolerated his therapy well, receiving chemotherapy and scheduled antiemetics, including corticosteroids. Chemotherapy treatments were administered every 2 weeks. Mr Fuentes was not advised to continue to follow-up with Dr A.

Eight weeks after beginning adjuvant chemotherapy, Mr Fuentes presented to the Parkland urgent care center with moderately elevated blood sugars and mild bipedal edema. He was treated with short-acting insulin and furosemide and urged to visit his PCP. Three weeks later, Mr Fuentes saw Dr A. During the interim, his symptoms had worsened, and routine activities of life had become more difficult. Dr A noted his diabetes to be uncontrolled and his CHF uncompensated and admitted the patient to hospital. Notably, Dr A had received minimal information regarding the patient’s overall treatment plan from Dr C; she had received no information regarding the patient’s complicating problems and few blood sugar laboratory results over the prior 8 to 10 weeks of chemotherapy treatment. Mr Fuentes was seen 2 days postdischarge by Dr A. He resumed his adjuvant chemotherapy after a 1-week delay. Dr C, recognizing the role of treatments in aggravating his course, made significant adjustments to eliminate the routine use of corticosteroids and ensured that blood sugars were monitored. He encouraged Mr Fuentes to keep his routine appointments with Dr A.

INTRODUCTION

Patients with cancer and chronic conditions (patients with complex cases of cancer) must navigate multiple health care teams, such as oncology, primary care, and other specialties, to receive care for all their conditions.¹ These teams vary in their ability to operate as a coherent system and in their effectiveness in achieving optimal health outcomes.^2,3 The patient often has limited familiarity with these teams and lacks the specific knowledge needed to traverse people and services involved in his or her clinical care. In turn, each team brings varying perspectives to the care of the same patient, following from their own (proximal) goals and normative processes for both critical and mundane tasks. The interface between primary and specialty care for patients contending with both ongoing chronic conditions and cancer offers a valuable opportunity to appreciate challenges of delivering well-coordinated care.

We examine the interface between oncology and primary care to explore the experience of the patient being discussed, a patient with multiple chronic conditions, after a cancer diagnosis at Parkland Health and Hospital System, the integrated safety-net system for Dallas County.^4,5 We view the health care teams encountered by Mr Fuentes as a multiteam system (MTS) with multiple actors, linked by complementary goals, knowledge, leadership, and communication needs. We describe these concepts first from the social science literature and then within the context of the patient’s experience.

KEY PRINCIPLE AND COMPONENT ELEMENTS

Among patients with complex cases of cancer, care is effectively coordinated in an MTS. To understand Mr Fuentes’s experience and the challenges that arise for effective care coordination, we consider the construct of an MTS and related coordinating mechanisms and processes. An MTS consists of two or more teams that, although independent with respect to their proximal goals, are interdependent in relation to a superordinate goal.^6,7 Some previous research has described teams across multiple disciplines within oncology;⁸ however, in this article, we move beyond oncology to specifically recognize care coordination within an MTS including primary care and oncology. This MTS achieves its superordinate goal (care for the patient with a complex cancer case) by coordinating components and resources, finding appropriate avenues of commonality while maintaining separate team identities and proximal processes.

Coordination involves alignment of components in a system, the effective management of dependencies among subtasks, resources, and people,⁹ to achieve collective performance. Coordination occurs at multiple levels: among members of an oncology team as well as among teams across clinical disciplines operating within an integrated health system like Parkland. Coordination processes may be explicit (eg, written instructions), tacit (eg, practiced procedures), or implicit (eg, shared understanding) and either relatively mechanistic (eg, rules) or organic (eg, emergent communication).¹⁰ Explicit coordination is often a product of planning, formalized and communicated through official documents and policies. Tacit coordination may occur through shared procedures that have been practiced and legitimized (eg, procedures in an operating theater). Implicit coordination refers to collective knowledge, or a shared mental model, that allows component units to “anticipate the actions and needs of their colleagues and task demands and dynamically adjust their own behavior accordingly, without having to communicate directly”^11(p164) (italics in original).

These types of coordination may be mechanistic or organic.¹⁰ Mechanistic coordination refers to stable or planned processes or rules. Technology systems, such as an electronic medical record (EMR), are increasingly used to support mechanistic coordination. Gawande¹² called for increasing use of technology in medical arenas, comparing surgeons to test pilots who use extensive explicit and mechanistic coordination tools such checklists and information-sharing technology for medical recordkeeping. In contrast, organic coordination emerges through interaction and communication, allowing adaptive responses to be formulated and shared.

USING THE MTS TO UNDERSTAND PRIMARY AND ONCOLOGY CARE

Viewing primary care and oncology care at Parkland as an MTS contributes to collective effectiveness by examining constituent team goals and coordination elements of knowledge, leadership, and communication. Here, we adopt a patient-centered view of Mr Fuentes as the focal point of Parkland care delivery, providing a common coordinating perspective.^13,14 Importantly, the patient is also a member of the MTS,¹⁵ who directly contributes to and experiences coordination effectiveness through different clinicians and support staff, as well as with nonmedical entities across the health system (eg, family, billing).

An MTS links primary care and oncology care teams; in addition to their own care delivery goals (eg, PCPs achieving good blood sugar control or oncologists managing long-term adjuvant therapy for cancer), both teams also have appreciation of interdependent goals (eg, chemotherapy can increase blood sugar levels, which need to be managed). Thus, an MTS approach tightly couples sets of activities within and between teams toward the superordinate objective of providing coordinated care.¹⁶ This goal system is supported through coordinating elements of knowledge, leadership, and communication, which bring together “specialized skills, capabilities, and functions”^7(p808) of the two constitutive teams, as well as other teams (eg, administrative) that the patient may encounter. These coordination elements may manifest as relatively explicit (or tacit or implicit) and organic (or mechanistic) and may relate to knowledge within and across the teams.¹⁷

Shared Mental Model

Knowledge within an MTS serves as a coordinating element through its structure, outlining what should be shared in common by constituent teams (eg, primary care clinicians being knowledgeable about symptoms of cancer recurrence and oncologists being aware of the need for continued management of chronic conditions by primary care clinicians throughout treatment and survivorship) and what should remain unshared (eg, specialized knowledge that is necessary for proximal goal fulfillment but is not required knowledge for all components). This configuration of collective knowledge is referred to as a shared mental model.¹⁸ Coordination of respective proximal goals of team actors will be influenced by the actors’ shared mental model based on the extent of their mutual understanding of task or process. Essentially, mental models allow people to predict and explain behaviors, recognize and remember relationships among components of the clinical setting, draw inferences, and construct expectations for what is likely to occur next.^19,20 Although “multiple mental models [co-exist] among team members at a given point in time [e.g.] models of tasks/technology, of response routines, of teamwork, etc.,”^18(p432) an effective MTS is premised on the need to make mental models explicit and to collectively determine how the MTS shapes scope of practice, roles, and responsibilities.¹⁸

Leadership

Leadership and influence within an MTS are distributed among constituent teams and members. Importantly, these include the patient and his or her caregivers who share complementary knowledge, influence, and responsibility in choices about health services with clinicians. This contrasts with some current models of care where the patient is the recipient of medical decisions made by clinicians.²¹ The members of an MTS share a form of distributed leadership, with “lateral influence among peers”^22(p48) emerging as a consequence of the goal structure and distributed resources.²³ Therefore, MTS coordination may benefit from a facilitative mode of leadership, especially when new systems are needed to bridge gaps between care teams.²⁴ Leadership style, whether directive or more inclusive, can affect how bridging within an MTS is fostered and maintained.²⁵ Leadership can improve MTS coordination and performance through strategizing and then organizing actions that augment members’ functional roles and improve interteam relations.²⁶ In cancer care, transfer of authority and control at critical junctures of care is challenging. Facilitative leadership roles, such as a chief of oncology services, can help bridge the gap between the two teams.^27,28

Communication

Effective teams keep one another informed with timely and accurate information, using multiple and appropriate modes of information transfer that facilitate problem solving.^29,30 Working across teams, explicit acknowledgment of communication among MTS members and communication in planning next steps are associated with high-performing teams.³¹ Communication is usually explicit; it may be organic (emerging in conversation or other direct communication), planned (documented procedures), and supported by technology systems (as repository or conduit), enabling sharing across time and distance.³²

Closed-loop communication refers to the process of embedding feedback in a communication system to ensure fidelity of information transfer and application. Also referred to as verifications, the basic components of closed-loop communication include: message sent (including explicit and contextual information), receiver acknowledgment, affirmation (or clarification) of interpretation by the receiver, and acknowledgment as correct by the sender.³³ Furthermore, in a health care setting, it is critical that feedback loops ensure that communication results in the intended action (eg, a patient’s question submitted through an online portal must be answered). Closed-loop communication is particularly needed during transitions of care both within and between teams.^34,35

UNDERSTANDING THE MTS THROUGH THE CASE

Although oncology and primary care may operate well as separate functional teams, Mr Fuentes’s experience demonstrates how two care teams may fail to adequately recognize themselves as constituents of an MTS with joint goals and mutual interdependencies. Understanding how coordinated care is delivered through an MTS and its facilitating factors (shared mental model, facilitative leadership, and closed-loop communication) provides a coherent explanation of how a health system may succeed or fail in achieving optimal health outcomes.

Coordinating Mechanisms in the Patient’s MTS

In Mr Fuentes’s case, several coordinating mechanisms might have affected the care he received (Table 1). His care was compromised in key instances because of lack of optimal explicit coordination at the health system level. For example, establishing formalized, explicit policies around triaging high-risk patients for a timely colonoscopy and chemotherapy infusions and communicating these to all members of the MTS can help in reducing diagnostic and treatment delays.^36,37

Table 1.

Coordinating Challenges in the Patient’s Care

An EMR-derived care plan detailing Mr Fuentes’s chemotherapeutic drugs and regimens would have helped the PCP to better manage his chronic conditions.^38-40 This is an example of a mechanistic coordination process, whereby the health system can incorporate a care plan module into the EMR, which is accessed by the MTS involved in the patient’s care.

Lack of explicit or tacit coordination between the two teams and Mr Fuentes during his adjuvant chemotherapy resulted in a potentially avoidable hospitalization. As described in the optimized case (Appendix, online only), if the oncology care team had involved the PCP, Mr Fuentes’s diabetes and CHF would have been acknowledged and managed in a timely and expeditious way in the outpatient setting. If two teams have never worked together, a more explicit coordination mechanism may be required from system leadership to first formalize new workflows between the two teams. Over time, the two teams, through continuous quality improvement, can begin to more tacitly and implicitly coordinate care through shared workflows and the ability to adapt to each other as needed.¹¹

Factors Facilitating Coordination in an MTS

Facilitative leadership,^26,41 a shared mental model,^42,43 and closed-loop communication^27,44,45 are key foundational factors essential to achieving effective coordination in an MTS. Absent a shared mental model between the care teams and the patient, we might find Mr Fuentes believed his comorbid conditions to be less important in light of his cancer diagnosis. The lack of communication between team members further compromised care. The oncologist may have assumed that Mr Fuentes’s diabetes was being cared for by the PCP, while the PCP, left out of the communication loop, was unable to help serve as a patient advocate. Thus, lack of a shared mental model disrupted closed-loop communication and resulted in poor coordination of care.

Similarly, when Dr A suspected rectal cancer, it was unclear how or which provider explicitly stepped in to address any distress resulting from this information, particularly before confirmatory testing. Weeks elapsed before a colonoscopy was performed, with no communication with the patient to explain reasons for the delay. We might expect that Mr Fuentes’s level of anxiety intensified during this time.^46-48 Lack of a shared mental model was again evident in the failure to anticipate needs resulting from the pending diagnosis. Who was responsible for assessing Mr Fuentes's potential psychosocial needs, and what referrals should have been made? Who was responsible for following up on tasks? Who should Mr Fuentes have contacted and when? How might providers have shared accountability for patient needs with Mr Fuentes himself?

Poor clinician engagement can directly affect both patient trust and trust among other clinicians who should expect to be invoked and actively informed; this is an indicator of a faulty MTS.^49-51 As our optimized case describes (Appendix), a clinic navigator assisting Mr Fuentes could have helped coordinate his needs across different teams, reducing gaps in care that may arise as a result of teams operating independently. Similarly, a social worker could have aided Mr Fuentes by tracking his appointments, making reminder calls, and even arranging transportation. Such staff can serve as a boundary spanner^52-55 to facilitate a shared mental model of care across the MTS.⁵⁶

In Parkland, as in other settings, this MTS is virtual, with actors and actions separated by time and space.⁷ However, team identity and effectiveness are commonly fostered by colocation. Closed-loop communication becomes essential (specifically feedback, alerts, and checklists often operationalized within the shared EMR) to ensure that actions are made, received, acknowledged, and acted upon, with the results returning to the initial actor.^31,57 In Mr Fuentes’s case, mechanistic coordination opportunities for closed-loop communication through the EMR were not leveraged to trigger timely coordination prompts or reminders. Indeed, effectiveness of the EMR in improving care lies in its role as a vehicle for shared knowledge by members and units across the health system. Thus, coordination of communication is itself a process of collaboration that brings to attention what is necessary, expedient, and important.⁵⁸

Developing shared mental models and closed-loop communication within and across teams requires leadership to identify gaps in care delivery and engage stakeholders in developing a shared vision to improve care. A shared mental model codefines roles and responsibilities, while closed-loop communication ensures that assigned tasks are, in fact, performed and executed. In Mr Fuentes’s case, system leaders had multiple opportunities to engage primary care, gastroenterology, and oncology care teams and patient advocates in developing and implementing new models of coordinated care. For example, appropriate documentation of chronic conditions and cancer treatments were areas where system leaders could have leveraged the existing integrated EMR system at Parkland^59,60 to, for example, implement a survivorship care plan that was easily accessible to both Mr Fuentes and all the clinicians involved in his care.^61,62

IMPLICATIONS FOR CLINICAL CARE

To date, needs of patients with complex cases of cancer have not been well met, resulting in poor health outcomes.⁶³ A new cancer diagnosis often interrupts existing chronic disease management because a patient undergoes intensive oncology care for extended periods of time, during which attention to other conditions may wane.⁶⁴ Furthermore, patients with cancer often continue to be observed by oncologists after completion of initial treatment,^65-68 with little or no care coordination with PCPs. Thus, care is fragmented,⁶⁹ and providers are siloed,^70-72 resulting in suboptimal care quality.⁷³ Organizing care for a patient with cancer and other chronic conditions requires primary care and oncology care teams to acknowledge each other as part of an MTS with a superordinate care delivery goal for the whole patient. Then, effective care coordination mechanisms can organize care activities and provider information sharing to facilitate comanagement and appropriate service delivery.^68,70,74-76 Importantly, as facilitative leaders, clinicians should take the initiative to explicitly include patients as part of the MTS. Better use of patient-reported outcome measures^77,78 incorporated into the EMR^79,80 may further strengthen closed-loop communication, measurement, and accountability between patients and clinicians.^81,82

Approaches to shared care between primary and specialty clinicians are evolving, and effectiveness data vary.^83-86 For example, consensus guidelines for colorectal cancer survivorship address primary care management, including implications for care coordination.⁸⁷ However, such guidelines are rarely the products of joint deliberation and consensus between primary and specialty care. They fall short of engaging roles and responsibilities that follow from an explicit MTS approach, where component actors have articulated mutual expectations and delineated processes to achieve shared objectives through standard operating procedures (ie, a shared mental model).

Parkland has established acute response clinics to increase access for patients with complex cases at higher risk of hospitalization.⁸⁸ These transitional care clinics colocate multidisciplinary providers, fostering team-based care. Mr Fuentes’s urgent care visit and/or hospital discharge would have triggered referral to and an expedited appointment at the acute response clinic. Although scheduling delays are distinct from clinician communication, timely scheduling is a quality marker of care coordination^25,89,90 and a critical challenge for patient-centered care.^91,92

Even in integrated settings, care is disjointed because specialty training fosters organizational separation by practice discipline.^93,94 Patient-centered medical home initiatives seek to transcend clinical silos but remain nascent in oncology.⁹⁵ Optimizing the interface between specialty and primary care is a vital domain for quality improvement and practice change.⁴⁹ EMR registry functions could identify and monitor patients with complex cases whose multiple conditions create elevated risk for care transition challenges.^96-98

IMPLICATIONS FOR RESEARCH

Acknowledging an MTS is a key vehicle to improving care coordination. Quality improvement through both pragmatic trials and implementation research offers an opportunity to leverage evidence-based interventions, notably from primary care,^99-105 to understand how team-based approaches to coordination can be leveraged in oncology. Context matters^106-108; settings like federally qualified health centers or community clinics outside of vertically integrated systems like Parkland are important sites to understand how MTS operations can be sustained across clinic organizations and practices, especially absent a unifying, common EMR.^109,110 As we have argued, MTS care coordination operates at multiple levels, and a science of team-based care for patients with complex cases of cancer will require multilevel interventions and multilevel evaluations.^111-113

Systematic reviews of care coordination interventions have identified compelling evidence for improved outcomes,¹¹⁴ including: EMR-driven transitions using alerts and referral tracking,^96-98 intensive case management,^115-119 and team-based care.^120-122 These strategies improve care transitions, enhance continuity, reduce lack of referral follow-up, and ensure patients receive recommended care through routine information sharing. They also demonstrate improved patient experience of care and clinical outcomes.^{119,120,123,124} Despite evidence regarding best strategies to improve care for multimorbidity,^125,126 applications for patients with complex cases of cancer are needed.^127-130 Specifically, interventions have not been implemented within an explicit MTS framework approach to organizing and improving care of patients with complex cases of cancer.

In conclusion, MTS effectiveness is a significant determinant of patient experience and health system integrity. Without a shared mental model of that composite team, a culture of common goals, and appropriate patient engagement, it does not matter how much technology or how many tools an organization makes available to clinicians to improve outcomes.⁴³ Strategically, those individual patient-level outcomes are leading indicators of system-level performance, with direct effects on quality, mission fulfillment, and public funding. Further research and practice should investigate the relationships of MTS coordination with treatment compliance, barriers to care, and health outcomes.

Appendix

Full Case

Mr Fuentes is a 63-year-old construction worker with a history of type II diabetes mellitus, mild renal insufficiency, and mild congestive heart failure. He visits with his primary care physician (Dr A) two to three times a year for management of his chronic diseases. Approximately 1 year ago, he experienced bleeding per rectum (confirmed with a fecal immunochemical test), change in bowel habits, and unexplained weight loss. Dr A suspected rectal cancer and initiated a referral for a gastroenterology evaluation. The gastroenterology intake coordinator evaluated the request and asked that Dr A reformat her request to send the patient directly to the colonoscopy clinic. Back and forth communication between the gastroenterology clinic and Dr A took roughly 2 weeks to resolve. The colonoscopy clinic then reached out to the patient and scheduled the procedure 2 weeks hence. Four weeks after his visit to Dr A, Mr Fuentes underwent colonoscopy, with the identification of rectal cancer. On hearing the results of study, Dr A initiated a referral to Dr C, a medical oncologist, for further evaluation and management.

After evaluating the patient and reviewing the colonoscopic procedure and histopathologic reports from the biopsy, Dr C completed a staging workup, requesting chest, abdominal, and pelvic computed tomography scans, carcinoembryonic antigen level, and liver and renal function tests. These tests confirmed a diagnosis of clinical stage II rectal cancer (T3 N0 M0). Dr C discussed with the patient the need for multidisciplinary care to accomplish the goals of care (curative intent). Dr C requested consultations with the colorectal surgeon (Dr S) and radiation oncologist (Dr R). Over a 2-week period, the patient was seen by all of the consultants, and his care was discussed at the gastroenterology tumor board. A multidisciplinary treatment approach was recommended, to include preoperative neoadjuvant chemoradiotherapy followed by reassessment and surgical resection, followed by the completion of adjuvant chemotherapy. Mr Fuentes was seen by Dr C, the treatment algorithm was explained, and arrangements were made to start first-line chemoradiotherapy with oral capecitabine concurrently with radiation therapy. He was simulated by radiation oncology and initiated on therapy four weeks from his initial visit with Dr C (8 weeks after presenting to Dr A with suspected rectal cancer).

After completing a 5-week course of concurrent adjuvant chemoradiotherapy with minimal toxicities, Mr Fuentes was re-evaluated by colorectal surgery and underwent surgery (transabdominal resection with a primary reanastomosis). Pathology identified residual cancer within the resected rectum (although treatment effect was present); lymphovascular invasion was present, and two of 12 lymph nodes showed metastatic carcinoma. The postoperative stage was ypT3N1M0 stage III rectal cancer.

Mr Fuentes saw Dr C 4 weeks postoperatively; she reviewed the pathology and outlined the ongoing plan with the patient. The next step recommended was to complete adjuvant chemotherapy with eight to 10 cycles of FOLFOX (fluorouracil, leucovorin, and oxaliplatin). To accomplish this, Mr Fuentes required a mediport for the infusional fluorouracil in the regimen. Dr C requested placement of the port by interventional radiology (IR). Although IR can typically accomplish this procedure with minimal sedation and on a single visit, the IR schedule allowed for an earliest appointment in 4 weeks. Dr C conferred with surgery, and they suggested taking the patient back to the operating room to place the port in the coming week. This required a preoperative visit by the patient. Although Dr C was concerned about the extra visits and use of such a high-level, high-cost center to accomplish the port, he agreed to the plan because he was anxious to get treatment started. After clearance by anesthesia, the port was placed in the operating room. Mr Fuentes returned to medical oncology the following week, and chemotherapy was planned; however, the next appointment slot available in the infusion room was 2 weeks hence. Eight weeks postsurgery, Mr Fuentes had his port and began adjuvant chemotherapy.

Initially, Mr Fuentes tolerated his therapy well. He had minimal nausea without emesis and no significant diarrhea or hematosuppression. Built into his chemotherapy treatment protocol was routine use of corticosteroids to prevent delayed chemotherapy-related emesis. Notably, blood sugars were not included with his routine laboratories. This occurred because, in the electronic medical record (EMR) ordering system, Mr Fuentes’s diabetes diagnosis was not entered in the treatment planning diagnosis for each visit. The EMR failed to record blood glucose because there was no corresponding diagnosis, and no acknowledgment of a noncovered notice was given to the patient.

Mr Fuentes was being observed by oncology every 2 weeks and was not explicitly advised to continue seeing Dr A for management of his chronic conditions. Mr Fuentes assumed that Dr C was taking care of all of his needs.

Eight weeks after beginning his adjuvant chemotherapy, Mr Fuentes developed breathlessness on exertion, fatigue, and leg pain, for which he sought care at the Parkland urgent care center. At that time, his blood sugars were moderately elevated, and he had mild bipedal edema. The urgent care physician treated his symptoms with novolog short-acting insulin and furosemide and urged him to visit with his primary care physician as soon as possible.

Mr Fuentes did keep his chemotherapy appointment and, in light of the urgent care visit earlier in the week, was encouraged to call Dr A for follow-up. Mr Fuentes reached out to Dr A’s office and received an appointment for 3 weeks hence—all the while being maintained on his chemotherapy treatments.

During this time, his symptoms continued to worsen, and his activities of daily living were significantly compromised. When assessed by Dr A at the scheduled visit, she quickly noted his uncontrolled diabetes and worsening congestive heart failure (CHF). Dr A noted that she had received an initial note from Dr C, an end of radiation treatment summary from Dr R, and a summary from the urgent care center; however, she had not received ongoing information or messages regarding Mr Fuentes’s overall treatment plan from Dr C. She was able to view individual notes in the common EMR of the health system, noting that there were few notes regarding assessment of the patient’s complicating problems and few blood sugars over the prior 8 to 10 weeks of treatment. Also, Dr A did not have easy access in the EMR to information about the specific cancer drugs Mr Fuentes was receiving. She suspected that Mr Fuentes was receiving corticosteroids as part of his cancer therapy, which contributed to his high sugar levels and aggravated the fluid retention that worsened his pre-existing CHF. Dr A decided to hospitalize Mr Fuentes for rapid management of the CHF and diabetes. She messaged Dr C to outline her plan and sent Mr Fuentes to the emergency room.

Mr Fuentes was admitted to the hospital and seen by oncology and underwent an array of internal medicine consults to manage the complications of his care. After a 5-day hospital stay, with much improved control of his CHF and diabetes and with appropriate adjustments in his medications, he was discharged.

When seen 2 days postdischarge by Dr A, all seemed well. Mr Fuentes resumed his adjuvant chemotherapy after a 1-week delay. Dr C, recognizing the role of the cancer treatments in aggravating his course, made significant adjustments in Mr Fuentes’s treatments to eliminate the routine use of corticosteroids. He also updated the encounter problem list to ensure appropriate laboratory monitoring of Mr Fuentes’s diabetes was collected. Lastly, he encouraged the patient to keep his routine appointments with Dr A.

At the completion of his adjuvant chemotherapy, Mr Fuentes was asked to follow up with Dr C quarterly for 2 years and then every 6 months and subsequently referred to a nurse practitioner–led survivorship clinic that coordinated follow-up with Dr A.

Optimized Case

Mr Fuentes is a 63-year-old construction worker with a history of type II diabetes mellitus, mild renal insufficiency, and mild CHF. He visits with his primary care physician (Dr A) two to three times a year for management of his chronic diseases. Approximately 1 year ago, he experienced bleeding per rectum (confirmed with a fecal immunochemical test), change in bowel habits, and unexplained weight loss. Dr A suspected rectal cancer and initiated a referral for a gastroenterology evaluation. The gastroenterology intake coordinator evaluated the request, recognized that Mr Fuentes was at high risk for an underlying malignancy, and referred the endoscopy request to the oncology evaluation clinic. The oncology clinic navigator contacted Mr Fuentes and explained what had taken place so far. Then, working within the referral guidelines established in the clinic and with feedback to Dr A's office, the navigator arranged for a colonoscopy to be performed within the week by colorectal surgery. Mr Fuentes underwent colonoscopy with the identification of an obvious rectal cancer. The colorectal surgeon, Dr S, initiated the agreed-upon algorithm of the gastroenterology disease–oriented team to fast track staging and evaluation of rectal cancer and discussed with Mr Fuentes the need to get appropriate staging studies. Dr S also discussed how adjuvant therapy could increase the odds for cure, emphasizing the value of receiving care from both medical and radiation oncologists as the optimal treatment course.

The navigator arranged visits with Dr C, a medical oncologist, and Dr R, a radiation oncologist, for further evaluation and the development of a treatment plan. The navigator also introduced Mr Fuentes to one of the health system social workers, who helped him understand how the Parkland medical assistance program covered costs of treatment, because he was not yet eligible for Medicare. The social worker further assessed Mr Fuentes’s family dynamics and offered to have a Spanish-speaking oncology nurse available to help his adult sister understand Mr Fuentes’s treatment plan. The navigator also consulted with Dr C, who reviewed Mr Fuentes’s record and approved the scheduling of appropriate studies: chest, abdominal, and pelvic computed tomography scans, carcinoembryonic antigen level, and liver and renal function tests.

Over the next 10 to 14 days, the studies were completed, and Mr Fuentes was seen by Dr C and Dr R. He learned that the studies confirmed a diagnosis of clinical stage II rectal cancer (T3 N0 M0). Dr C discussed with Mr Fuentes the need for multidisciplinary oncology care to accomplish the goals of care (curative intent). Within 2 weeks of his presentation to Dr A, Mr Fuentes’s staging studies were complete, he had been seen by Dr S, Dr C, and Dr R, his case had been discussed by his multidisciplinary care team at the gastroenterology tumor board, and his treatment plan had been defined (preoperative neoadjuvant chemoradiotherapy followed by reassessment and surgical resection followed by the completion of adjuvant chemotherapy). The navigator compiled the recommendations in a written care plan that was shared with Mr Fuentes, and a copy was sent to Dr A.

The following week, Mr Fuentes was simulated by radiation oncology and initiated on neoadjuvant chemoradiotherapy with oral capecitabine concurrently with radiation therapy. During this treatment, his counts were monitored weekly with Dr C. Recognizing that he would need a mediport for future therapies, the navigator scheduled the procedure with interventional radiology to occur in the planned 6-week interim between the completion of Mr Fuentes’s concurrent chemoradiotherapy and his surgery.

After completing a 5-week course of concurrent adjuvant chemoradiotherapy with minimal toxicities and having his port placed by IR as an outpatient, Mr Fuentes was re-evaluated by the colorectal surgeon, Dr S, and underwent surgery (resection with a primary reanastomosis). Pathology identified residual cancer within the resected rectum (although treatment effect was present); lymphovascular invasion was present, and two of 12 lymph nodes showed metastatic carcinoma. The postoperative stage was ypT3N1M0 stage III rectal cancer.

Mr Fuentes saw Dr C 3 weeks postoperatively. The pathology report was reviewed, and Dr C outlined the ongoing plan with Mr Fuentes. The next step recommended was to complete adjuvant chemotherapy with eight to 10 cycles of FOLFOX. A chemotherapy appointment was scheduled, and Mr Fuentes began his adjuvant chemotherapy 4.5 weeks postsurgery.

During the several visits with Drs C, R, and S, Mr Fuentes was encouraged to maintain scheduled appointments with his primary care physician, Dr A. Interval laboratories and notes were shared with Dr A, and the visits were coordinated by the navigator in oncology. Attention was focused on Mr Fuentes’s nutritional status, his diabetic management, and his cardiac status.

Initially, the patient tolerated his therapy well; however, 6 weeks in—roughly during his fourth cycle of therapy—it was noted that Mr Fuentes's blood sugars were progressively rising. Built into his chemotherapy treatment protocol was routine use of corticosteroids to prevent delayed chemotherapy-related emesis. Dr C and Dr A discussed Mr Fuentes's care. Mr Fuentes consulted with Dr A, who reviewed the laboratories and also noted that the patient had bipedal edema and had gained 5 lbs since initiating therapy. Dr A added short-acting insulin and a small dose of furosemide. After talking to Dr A, Dr C recognized the complicating role of the prophylactic corticosteroids in the chemotherapy regimen and substituted aprepitant in the regimen.

Mr Fuentes completed his remaining chemotherapy cycles without event. At the conclusion of his therapy, he met with Dr C, who outlined future surveillance studies and a follow-up schedule. The oncology navigator created a treatment summary and a survivorship plan, again sharing the document with both Mr Fuentes and Dr A.

AUTHOR CONTRIBUTIONS

Conception and design: Simon J. Craddock Lee, Mark A. Clark, John V. Cox, Bijal A. Balasubramanian

Administrative support: Simon J. Craddock Lee

Provision of study materials or patients: John V. Cox

Collection and assembly of data: John V. Cox, Burton M. Needles, Carole Seigel

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Achieving Coordinated Care for Patients With Complex Cases of Cancer: A Multiteam System Approach

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO’s conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.

Simon J. Craddock Lee

No relationship to disclose

Mark A. Clark

No relationship to disclose

John V. Cox

Employment: Texas Oncology, University of Texas Southwestern Medical Center Simmons Cancer Center

Leadership: Texas Oncology, Parkland Health System

Stock or Other Ownership: Amgen, Medfusion, Merck, Pfizer

Research Funding: US Oncology

Other Relationship: Mary Crowley Research Center, American Society of Clinical Oncology, Methodist Health System

Burton M. Needles

No relationship to disclose

Carole Seigel

No relationship to disclose

Bijal A. Balasubramanian

No relationship to disclose