“You’re the person for me:” patient perceptions of cancer care team selection and experiences with care coordination within a rural hub-and-spoke health system Authors: Mia St Angelo, Erika L Moen, Laura B Beidler, Christine M Gunn Published: 2026 ABSTRACT AbstractPurposeCancer care requires coordination across several clinical and supportive care specialties. Patients often need multidisciplinary teams to receive guideline-supported care; however, little is known about patients’ perceptions of care team selection and coordination, particularly in the context of a rural hub-and-spoke health-care system.MethodsIn this qualitative study, we conducted semi-structured interviews with patients regarding breast, lung, or colorectal cancer who received treatment at a large, rural academic health system. The cancer multi-team systems (cMTS) framework guided interview coding and analysis.ResultsWe conducted interviews with 11 breast, 11 lung, and 12 colorectal cancer patients. The mean participant age was 66 (range: 34–85); 56% were female, and 41% lived in a rural area. Half (n = 17) received care only at the central hub location, with most remaining participants (n = 13) receiving care across both hub-and-spoke facilities. Results include two main themes: 1) patient preferences, including past experience, interpersonal factors, and practical issues, drove patient care team and facility selection; and 2) patient experiences with providers influenced perceptions of team-based care and teaming processes. We did not discern differences between participants whose care teams were colocated at the hub hospital compared with those spanning hub-and-spoke. We used the cMTS framework to illustrate the inputs and processes most relevant to this study population’s decisions and perceptions regarding care delivery.ConclusionFindings contribute to our understanding of patient preferences related to care team selection. These factors could be important in shaping future development of multidisciplinary cancer care teams.Supplementary InformationThe online version contains supplementary material available at 10.1007/s10552-026-02174-9.Keywords: Breast cancer, Colorectal cancer, Lung cancer, Care teams, Care coordination, Rural Introduction Cancer care is increasingly specialized, motivating the centralization of complex cancer treatments to high-volume regional centers to increase efficiency and care quality [1–3]. Prior work has shown that rural-residing populations are particularly impacted by the regionalization of cancer care [4–6], raising questions about the effectiveness and patient-centeredness of regionalization of care given patient preferences for care closer to home [7, 8]. Rural-residing patients frequently balance the potential benefits of higher care quality from a specialized regional center with the associated costs of travel burden [8–10]. Health systems that centralize complex or high-resource care to regional centers while maintaining a presence in smaller community health-care facilities [11–14] are referred to as “hub-and-spoke” models of care [15–17]. For patients navigating a hub-and-spoke health system, it is not uncommon for them to receive sequenced treatments from different facilities, leading to geographically dispersed, multidisciplinary care teams [18] [19, 20]. The Cancer Multiteam System (cMTS) framework (Appendix, Fig. 1) was developed as a resource for studying cancer care teams and care coordination [21]. This framework features relationships between structural inputs (e.g., features of the care team and health system), teaming processes within and across teams/facilities, clinical care, and the resulting multi-level clinical outcomes. A better understanding of the factors that influence care team selection among rural-residing patients is crucial for ensuring health system referral processes are patient-centered. Further, we need deeper insight into the relationship of geographic dispersion of care teams, a cMTS structural “input” that can impact care processes and outcomes, with patient experiences with care coordination. The objective of this study was to explore patient preferences for care team composition and locations of care and characterize patient experiences with cancer care coordination within a rural hub-and-spoke health-care system. We focused on patients with breast, colorectal, or lung cancer because these are common cancers that often require multidisciplinary teams [22]. These cancer sites include cancers where efforts to regionalize surgical care due to greater specialization and complexity have been described, particularly for lung and rectal resections [23, 24]. Using the cMTS framework to inform the development of interview guides, we conducted in-depth semi-structured interviews with patients who received cancer treatment within the health-care system to explore their preferences and perceptions of care team features, care locations, and care coordination. Materials & methods Recruitment for semi-structured interviews This study was conducted at Dartmouth Cancer Center, a National Cancer Institute-designated Comprehensive Cancer Center located in Lebanon, New Hampshire, USA. This study was approved by the Dartmouth Health Institutional Review Board (STUDY02001747). The Lebanon, New Hampshire facility serves as the “hub” of the health system, and numerous “spoke” facilities are located throughout the cancer center’s catchment area that spans New Hampshire and Vermont. This catchment area is largely rural [25, 26], with more urban areas in southern New Hampshire. We aimed to enroll a majority rural sample; however, we did not exclude participants based on rurality. Patients diagnosed with lung, breast, or colorectal cancer in 2022 or 2023 and who received care from a participating cancer clinician within the health system were eligible for this study. We first developed a list of cancer clinicians who had given the research team permission to recruit from their panels and then we received a list of potentially eligible patients from those clinician panels. This included 312 breast cancer patients, 114 colorectal cancer patients, and 172 lung cancer patients. We randomly selected 25 patients from each cancer type with the goal of recruiting 10–12 participants from each. Patients were mailed recruitment letters, with email and phone follow-up outreach, as needed. An additional round of recruitment letters was sent to another 25 colorectal and 25 lung cancer patients to achieve the desired sample size. Qualitative data collection A trained qualitative interviewer (L.B.) conducted all interviews. We followed a flexible, semi-structured interview guide that focused on the experience of receiving care at a large, rural health system and aligned with the domains of the cMTS framework. Interviews were conducted via telephone or Zoom and lasted between 30 and 60 min. After obtaining consent, interviews were recorded, professionally transcribed, and verified for accuracy by a member of the research team. Patients were compensated for their participation. Qualitative data analysis Data management and analysis were conducted using NVivo, a qualitative data analysis platform [27]. We used a codebook that was developed prior to coding and was aligned with the interview guide and concepts of the cMTS framework. Two coders (M.S. and L.B.) independently coded multiple transcripts and then met with the full coding team (C.G., L.B., M.S., E.M.) to review. Once we established consistency between coders, M.S. coded the remaining transcripts and L.B. conducted an unblinded coding review. The team met to discuss any discrepancies. The cMTS framework guided codebook development. We identified context, inputs, processes, and outputs that aligned with study concepts and assigned these elements to relevant codes. Codes were linked to components of the framework, providing theoretical context for analysis. This process contributed to the creation of an applied cMTS model (Fig. 1), which informed the organization of qualitative themes. Applied cMTS framework The complete codebook with definitions and associated cMTS framework concepts is in the Appendix (Table 3). A matrix approach was used for analysis [28]. Participants were listed along the y-axis and codes were listed along the x-axis. Within each matrix cell, qualitative descriptions of the participant’s representation of that code were documented. Theme development involved analyzing the nature of perceptions both within and across participants. We assigned each participant to a care location type based on their interview responses (hub-only, spoke-only, hub-and-spoke). This facilitated comparisons between participants by location of care and cancer type. We identified cMTS framework components and the conceptual relationships between them when developing themes and drawing conclusions. We assessed features of care teams both in alignment with the framework and those not included in the framework but identified as important during interviews. We explored the relationships between these elements and participants’ experiences with team selection and care coordination. To preserve the anonymity of participants and the study dataset, health-care facility names and health-care provider names have been redacted from participant quotes. For the purposes of this paper, “Health System” refers to the broader health network containing multiple affiliated health-care facilities, including cancer treatment sites and others such as primary care (Fig. 2). “Hub Facility” refers to the hub hospital within the Health System, and “Spoke Facility” refers to any other facility within the Health System where cancer care was received. “External Facility” refers to a facility not affiliated with the Health System. Health-care provider names are anonymized by specialty (e.g., “Medical Oncologist”). Visual representation of the Hub-and-Spoke health-care system Results Our study included 34 participants who were diagnosed with breast (n = 11), lung (n = 11), or colorectal (n = 12) cancer, with 29% in active treatment at the time of participation (Table 1). The mean participant age was 66 years (range: 34–85); 56% were female, 100% were non-Hispanic White, and 41% lived in a rural area. Participants were classified into one of four care location categories based on self-report: hub facility only (n = 17), hub-and-spoke facilities (n = 13), hub and external facilities (n = 2), or spoke facility only (n = 2) (Table 2). Table 1Participant demographicsCharacteristicTotal (N = 34)Breast (n = 11)Lung (n = 11) Colorectal (n = 12)Age (mean, [range])66.0 [34–85]67.5 [49–81]73.1 [53–85]58.0 [34–70]Race and Ethnicity White, non-Hispanic34 (100%)11 (100%)11 (100%)12 (100%)Gender (% female)19 (56%)11 (100%)4 (36%)4 (33%)Employment* Retired17 (50%)4 (36%)7 (64%)6 (%) Working full time12 (35%)5 (45%)3 (27%)4 (33%) Self employed4 (12%)2 (18%)1 (9%)1 (8%) Working part time3 (9%)1 (9%)1 (9%)1 (8%) Unemployed1 (3%)0%0%1 (8%) Other1 (3%)0%0%1 (8%) Active treatment (% yes)+10 (29%)3 (27%)6 (55%)1 (8%)Types of treatment received*+ Surgery23 (68%)11 (100%)3 (27%)9 (75%) Chemotherapy17 (50%)2 (18%)8 (73%)7 (58%) Radiation18 (53%)8 (73%)5 (46%)5 (42%) Immunotherapy8 (24%)1 (9%)6 (55%)1 (8%) Hormone therapy4 (12%)4 (36%)0%0% Health system – affiliated primary care provider (% yes)+ 12 (35%)4 (36%)2 (18%)6 (50%)Rurality Metropolitan2 (6%)0%1 (9%)1 (8%) Micropolitan 11 (32%)5 (46%)3 (27%)3 (25%) Small town7 (21%)3 (27%)1 (9%)3 (25%) Rural area14 (41%)3 (27%)6 (55%)5 (42%)*Participants were able to select more than 1 option, +Participant self-report Participant demographics *Participants were able to select more than 1 option, +Participant self-report Table 2Care location by cancer typeBreast (n = 11)Lung (n = 11)Colorectal (n = 12)Total (N = 34)Hub only7 (58%)7 (64%)3 (25%)17 (50%)Hub and spoke3 (25%)3 (27%)7 (58%)13 (38%)Spoke only002 (17%)2 (6%)Hub and external1 (8%)1 (9%)02 (6%) Care location by cancer type Theme #1: Patient preferences, including past experience, practical issues, and interpersonal factors contributed to patient care team and facility choices. History or tenure. The cMTS framework defines “history or tenure” as “previous experience working together or anticipated duration of the multi-team system.”[21] While this usually refers to clinicians working together, participants also reported that history with both clinicians and the health system was impactful for decision-making. Many participants described familiarity and reputation as top factors in their choice of where to receive care. Breast participant #137 noted, “I'm a patient of [Health System], I've been here for 30 years–so, I would not consider going anywhere else.” Other participants trusted their referring provider’s advice, as described by Colon participant #107: “It’s a respected physician. So, I didn’t question it.” Geographic dispersion. “Geographic dispersion” is defined in the cMTS framework as the “degree to which component teams are colocated or dispersed”[21]. We found the rural setting and its unique effects on patient travel burden to be important factors in participant decision-making, which ultimately impacted geographic dispersion of their care team. When participants described travel burden, they discussed mileage, travel time, and ease of travel (i.e., highway versus rural backroads). Breast participant #140 weighed these factors as part of her choice of where to receive care: “Let me say it this way, most of the options, it probably is the same amount of time away. So, any option for me to get radiation care is going to be at least an hour or so away, but on the highway rather than going over the mountains.” This patient chose to receive care at a facility more easily accessed by highway travel versus mountain backroads. Travel burden was frequently weighed with perceived quality of care. This decision often centered on whether to receive care at the hub or a local community facility. Lung participant #136 commuted over an hour each way to receive the entirety of their care at [Hub Facility], despite accessibility to more local options: “I have really good friends that work in oncology right here in town, but I just felt better going to [Hub Facility].” Breast participant #224 chose to come to [Hub Facility] despite being offered more local care, citing confidence in [Hub Facility] facilities: “It was a choice. I mean, I've a bad experience at [Community Hospital], so I decided I needed some place else where I knew I could get care.” Others were more willing to receive care at a spoke facility affiliated with the Health System. Several participants referenced the hub-and-spoke model as a means of easing travel burden while retaining access to high-quality care. Lung participant #42 stated, “Driving out to [Hub Facility], getting the treatment, driving back, it was at least six hours. So, if I can receive the same care in [Spoke Facility], where that location is only 10 min from my office, of course, I'm gonna take that.” In some cases, as described by colorectal participant #107, patients referenced conversations with their care team about this choice: “I remember asking that clearly. [Spoke Facility] was more convenient. But if it was compromising my care, I would travel to [Hub Facility]. But I imagine that conversation was with [Physician’s Assistant], and she told me I would not be compromising my care. So that's why I chose [Spoke Facility].” Interpersonal rapport. Outside of the structural factors represented in the cMTS framework, feeling comfortable with their provider was an important deciding factor for many patients when assembling their care teams. As colorectal participant #43 described, a caring and credible demeanor was important in patients’ decisions to move forward with care: “I chose [Radiation Oncologist] because his demeanor was so comforting, and the same with [Medical Oncologist]. I have to say that they really made me feel like they cared for me.” A few patients compared their experiences with other providers when making their choice: “After interviewing the [Cancer Center] surgeon, I simply said, “You're the person for me.” I really felt much more comfortable in the interview and I'm smart enough that, I don't know, I just liked her qualifications better.” - Colorectal participant #41 Timeliness of scheduling. A few patients described delays at their preferred institution, which led them to seek care elsewhere. For some, timeliness superseded other factors, including location and clinician preference, as breast participant #310, who chose to receive surgical treatment at a local community hospital with a general surgeon despite an initial preference for more specialized cancer care, shared: “[Hub Facility was] like, ‘Yeah, we might be able to see you in October. But we only have two surgeons and one of them is out on leave.’ So, you know, it was clear that because of the lack of severity of my [case] -- even though to me it was very significant. I wasn't getting in there anytime soon and I really just wanted it out of my body. So, I opted to have my surgery at [Community Hospital].” Theme #2: Patient experiences with providers during cancer treatment influenced perceptions of team-based care and teaming processes. Most patients described team-based care in one of two ways: (1) teamwork within one large team, or (2) teamwork between multiple teams, usually aligned by specialty or phase of care (Fig. 3). No differences in team type were observed by care location (hub versus spoke, or single- versus multi-site) (Fig. 4). However, participants’ perceptions of care coordination differ across these two team types. Example representations of team structure Example representation of care location structure Teamwork within one large team. Participants who described one cohesive team experienced near-simultaneous care from multiple providers in different specialties. In some cases, this style of teamwork was observed directly, as described by Breast Participant #137, a hub-only participant: “When [Medical Oncologist 1] came on board, [Medical Oncologist 2] made it very helpful that he brought her into our sessions for the last three times he was there….and then, the last time I met [Medical Oncologist 1], [PA] was in the room as well.” In other cases, participants perceived cohesive teamwork to be happening outside of their interactions with their care team. Breast participant #224, a hub-and-spoke participant, stated: “[All care team members] knew everything…that one person had said, everybody else knew what was in my file. So, I felt that they were working as a team.” Participants in this group perceived high levels of “group decision-making” among their care teams, aligned with the cMTS definition as the “negotiation and creation of shared meaning among a group”[21]. Many participants perceived that all members of their team were informed about their case and made treatment decisions as a group, such as colorectal participant #46, a hub-only participant: “Every part of my body that had anything to do with that cancer, there were special teams that came in and saw me. The team for the vascular came and saw me, the team for the surgeon for my stomach, they all came and saw me, neurologist came and saw me. And they would stand right there and discuss it and talk to me about it and ask questions.” Information exchange was also prevalent among patients who described a cohesive, one-team structure. Breast participant #121, a hub and external participant, experienced information exchange, defined as closed-loop communication between parties, within their care team: “At the same time, I was having some hematology discrepancies and the hematologist at the VA worked with [Medical Oncologist] and [Surgeon] to make sure my care was comprehensive.” Teamwork between multiple specialty teams. Participants who perceived teamwork between multiple teams often described handoffs from one clinical group to another. Breast participant #181, a hub-only participant, described a linear flow from one specialty team to the next: “They all worked as a team but there was one group and then there was a different group afterwards.” These participants had more mixed perceptions of group decision-making. Some participants, such as breast participant #140, a hub-only participant, described positive information exchange and group decision-making among providers: “I feel like a team came together and said, ‘Let's look at this patient and decide what would happen.’ And I don't know that that's what happened, but that's what it felt like from my side of the screen.” Some perceived more siloed care, as described by lung participant #60, a hub-only participant: “The bronchoscopy surgeon didn't try to speak for anyone other than himself and his part and his role.” Other perspectives were somewhere in the middle, such as that of colorectal participant #92, a spoke-only participant: “It's a little tricky, because I think it was relatively segmented... I don't know if there was really ever a need for the surgeon to talk to the oncologist. Maybe they did and I don't know. But I don't really know if any of that happened.” Participants who described a multiple team structure witnessed shared leadership within their care team, which cMTS defines as the “transference of leadership functions among team members to take advantage of member strengths.” Strengths in this case refer to knowledge and skills of specialty teams related to their respective phases of treatment. Colorectal participant #41, a hub-and-spoke participant, described this process: “I think when they hand off to the oncologist, I think, it's definitely an oncology team at that point. But I never thought that the surgery team was not fully coordinated with the oncology team. I thought it was very good communication between various professionals.” Some participants described “boundary spanning” between teams providing sequential care, which cMTS refers to as the “actions of teams to establish linkages with one another:”[21] In these cases, although teams were not actively working together directly, participants did feel that they were aware of each other and communicated about the decisions they made. Lung participant #60 stated: “There was no discussion like, well, me and Bob were having coffee the other day, not that kind of thing, but it did not feel disjointed. Because I think the record keeping was good and getting moved around properly.” Other participants, such as breast participant #310, a hub-only participant, described taking on the role of boundary-spanning themselves: “I see my radiation oncologist next week for my final visit with her, but I know I have to have a mammogram set up, and I don't know if she's gonna make that referral or I have to go back to my PCP. Like there's just a lot of questions about like, I feel like if I don't stay on top of it, things will fall through the cracks.” Discussion The relationships between cancer care access, coordination, and quality are complex, and it is often challenging to optimize all three from the perspectives of patients, physicians, and health systems. This qualitative study adds to our understanding of how patient preferences influence the development of a patient’s multidisciplinary cancer care team, particularly when patients are presented with a choice of care locations. The two patient-centered themes that were identified and integrated with the cMTS framework were 1) patient preferences, including past experience, interpersonal factors, and practical issues, played a key role in care team and facility selection; and 2) patient experiences with providers influenced perceptions of team-based care and teaming processes. We found that patients often considered their history with the health system, interpersonal rapport with physicians, timeliness of scheduling, travel burden and perceived quality of care when selecting their cancer care team. While each patient weighed these elements differently, most patients felt involved in determining which physicians and locations would provide their cancer care. Fong et al. identified many similar decisional factors that were of value to patients’ choice of facility, such as team characteristics, referrals and recommendations, and travel distance, including the availability of local satellite centers for some elements of treatment (e.g., chemotherapy) [29]. Ejaz et al. found hospital-related factors, such as perceived facility reputation and hospital caseload, as well as physician qualifications, such as surgeon case-specific experience, to be top contributors to patients’ choice of cancer surgeon [30]. For rural-residing patients with cancer, the tradeoff between travel distance and perceived hospital quality has been previously described [29]. A recent study found that some patients were more willing to receive care locally when their community hospital had a cancer center affiliation due to the perceived higher quality of care that comes with that affiliation [10]. Our results support this finding, as perceived care quality at local facilities factored into participant willingness to travel farther to the hub facility. Our findings provide new insight into specific factors weighed among rural-residing patients, including ease of travel (e.g., highway vs mountain backroad), longstanding relationships with the health system, and availability of timely scheduling. Theme two highlighted the different ways patients perceived cancer care teaming and coordination. Two main team structures described were one multidisciplinary team and sequential specialty teams for each treatment modality. Our findings suggest that geographic dispersion of teams within a hub-and-spoke system did not impact patient experiences with care coordination and teaming. Given that our study focused on a hub-and-spoke model within one integrated health system (i.e., with most sites using a shared electronic health record system), it is possible that within integrated health systems colocating care is less important to patient experience than other teaming and communication features. Group decision-making, care team hierarchy, and boundary spanning were described as enhancing patient experiences with care coordination. Limitations Our study had several limitations. Our participants were treated within one rural academic health system, and their experiences may not generalize to other settings, especially as we could not explore the extent to which insurance coverage and networks might have influenced choices between different health systems and networks. This was a small and entirely non-Hispanic White sample that only included three cancer types; findings and implications remain limited to these groups. These data capture the patient perspective only; insight on the perspectives of patient caregivers and health-care providers remains limited. Patients were interviewed at varying points in their cancer treatment, and perspectives may reflect current care experiences or be influenced by satisfaction with treatment outcomes at the time of interviews. Future studies could contribute to the literature by characterizing decision-making in real time to better capture patient experiences. Conclusion Patients are the central members of any cancer care team and integrating patient preferences as care teams are assembled following diagnosis is critical for delivery of patient-centered care. Our findings have identified several features of care teams and experiences with care coordination that can inform conversations with patients as they make decisions about their cancer care teams. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 177 KB) Author contributions Erika Moen and Christine Gunn contributed to the study conception and design. Material preparation and data collection were performed by Laura Beidler. Data analysis was performed by Laura Beidler and Mia St. Angelo. Manuscript drafting was led by Erika Moen and Mia St. Angelo. All authors read, provided feedback, and approved the final manuscript. Funding This study was funded by the Friends of the Dartmouth Cancer Center. Data availability The data that support the findings of this study are not openly available due to risks to participant privacy. De-identified participant data may be made available by the corresponding author upon reasonable request. Declarations Conflict of interest The authors declare no competing interests. Ethical approval This study was approved by the health system’s Institutional Review Board (STUDY02001747). Consent to participate All participants provided verbal consent prior to complete any study activities. Consent to publication No identifiable participant data in any forms are included in this manuscript; therefore, participant consent to publish is not necessary. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Mia St. Angelo and Erika L. Moen have contributed equally to the manuscript. Mia St. Angelo and Erika L. Moen are listed as joint first authors. References Associated Data Supplementary Materials Supplementary file1 (DOCX 177 KB) Data Availability Statement The data that support the findings of this study are not openly available due to risks to participant privacy. De-identified participant data may be made available by the corresponding author upon reasonable request.