Examining Health System Stakeholder-Reported Barriers and Solutions to Foregut Cancer Care

Abstract

Purpose:

Many patients with foregut cancers fail to receive guideline-concordant treatment. The perspectives of health system stakeholders who care for these patients, and their insights into feasible solutions, remain underexplored.

Methods:

Using grounded theory methodology, 12 semi-structured interviews were conducted to identify multilevel barriers, and health system stakeholders proposed solutions to overcome these barriers.

Results:

Participants described barriers across four levels. Individual-level barriers included financial toxicity, transportation challenges, and psychosocial distress. Provider-level barriers included inadequate physician-patient communication and insufficient inter-provider communication. Institutional barriers included delays in accessing specialty care, scheduling inefficiencies, and institutional resource limitations. Policy-level barriers included care fragmentation, inadequate sub-specialty care in rural areas, and inadequate reimbursement. Proposed solutions included strengthening institutional psychiatric support programs, increasing inter-provider engagement, streamlining referral pathways, targeted rural physician recruitment, and reimbursement reform.

Conclusion:

Health system stakeholders proposed interventions emphasizing communication, navigation, and system-level integration that may improve the delivery of foregut cancer care in resource-constrained environments.

Introduction

Foregut cancers, including malignancies of the esophagus, stomach, pancreas and biliary tract, are associated with a high morbidity and mortality ^[1]. Optimal management typically requires systemic therapy, and surgery or other locoregional treatments delivered by coordinated, multidisciplinary teams. The complexity of systemic therapy regimens, technically challenging operations with lengthy post-operative recovery, and the need for specialized expertise challenge the delivery of care ^{[2, 3]}. As a result, 30–55% of patients with foregut cancer fail to receive guideline-concordant treatment ^[4–6].

Both the burden of disease and cancer-related mortality are higher in the Southeastern United States ^[7], where socioeconomic disadvantage, limited subspecialty availability, and substantial geographic barriers impede access to care ^{[8, 9]}. The increasing centralization of cancer care toward high-volume tertiary and quaternary centers has further widened the gap between academic and community settings, leaving many patients in underserved areas without accessible multidisciplinary expertise. Prior qualitative research has shown that patients treated at safety-net institutions in this region face significant barriers across the cancer care continuum ^[10]. Addressing these challenges requires coordinated multilevel solutions that extend beyond patient-level interventions.

Most studies thus far have focused on patient-reported barriers, or population-level administrative data ^[10–13], however, the perspectives of key health system stakeholders – community clinicians, ancillary staff, and administrators – who navigate these barriers daily to deliver care, remain underrepresented. Existing literature has also largely focused only on academic institutions, which may not reflect the operational and resource constraints of community and safety-net environments where many patients receive care.

Our study seeks to address this gap by exploring multilevel barriers and health system stakeholder-proposed solutions to foregut cancer care delivery within a large safety-net hospital and its surrounding rural catchment area in the Southeastern United States. Understanding these perspectives is critical, as these individuals advocate for patients, navigate system-level challenges, and frequently implement ad hoc solutions in the absence of formal infrastructure. By incorporating perspectives from health system stakeholders working in the community and safety-net setting, this study aims to provide actionable insights to guide the development of pragmatic, contextually grounded interventions to improve access, coordination and delivery of foregut cancer care.

Methods

Study Design and Setting

We conducted a qualitative study using grounded theory methodology to explore health system stakeholder perspectives on barriers to foregut cancer care. The study took place between May 2022 and May 2023 at a tertiary referral center in the Southeastern United States that serves as a safety-net hospital for central and southern Alabama, as well as adjacent areas of Mississippi, Louisiana and the Florida panhandle. The catchment area encompasses a racially and socioeconomically diverse population of approximately 5 to 6 million people living primarily in rural and medically underserved areas. The health system provides care to 200 to 300 patients with foregut cancers annually. The region has only two safety net tertiary hospitals that deliver foregut cancer care. Furthermore, the region is characterized by poor transport and healthcare infrastructure, with many hospitals experiencing chronic underfunding and resource and healthcare personnel shortages ^[8]. The Standards for Reporting Qualitative Research (SRQR) guidelines were used as a framework for study reporting. The study was approved by the University of South Alabama Institutional Review Board (IRB-2003085–1) ^[14].

Participant Recruitment

Health system stakeholders directly involved in the diagnosis, treatment or coordination of care for patients with foregut cancer were recruited using purposive and snowball sampling. Participants included both hospital-based and community-based providers to capture diverse perspectives from across the institution’s large, underserved catchment area. In addition to oncologists, social workers and administrators at the safety-net center, we actively sought participation from primary care providers (PCPs), gastroenterologists, and medical oncologists practicing in rural community settings outside the institution to ensure the inclusion of diverse perspectives from the broader regional network of referring providers. Participants were identified by the study team and through referrals from other participating health system stakeholders. Recruitment continued until thematic saturation of each health system stakeholder role was achieved. All contacted health system stakeholders agreed to participate in the interview.

Guide Development and Participant Interviews

A semi-structured interview guide was developed using grounded theory methodology. Open-ended questions and probing questions were used to facilitate dialogue and maximize participant discussion. The interview guide was pilot tested with a health system stakeholder to ensure relevance and clarity, with minor revisions made before formal data collection. The interview guide is attached as Supplementary File 1. One trained interviewer (a surgical oncologist with qualitative research expertise) conducted all interviews. Interview recordings were reviewed collaboratively throughout the study to minimize interview drift. Interviews were conducted in person or via secure videoconference, audio-recorded and professionally transcribed verbatim. Participants provided informed consent before the interview started and received no compensation for participation.

Stakeholder Variables

Simple demographic data were collected for each health system stakeholder through a brief demographic questionnaire capturing age, gender, race, occupation, institutional setting, and years in practice. Data were summarized descriptively to characterize the participating sample.

Qualitative analysis

All transcripts were verified for accuracy by the research team. Transcripts were then independently analyzed by two coders (JS and IL) using NVivo 14 software (Lumivero). Grounded theory methodology was used to guide inductive content analysis. Specifically, an objectivist grounded theory approach was chosen. This approach relies on minimizing the influence of a priori assumptions or theoretical frameworks, allowing codes and themes to emerge directly from the data ^[15]. Initial open coding identified key concepts, which were refined into a thematic coding structure through iterative discussion. The research team met regularly to compare codes, resolve discrepancies through discussion, and iteratively refine the coding framework. Coding continued until no new themes emerged. Interrater reliability exceeded 90%.

Reflexivity and Rigor

The interviewer’s dual role as a clinician within the health system was acknowledged and actively managed throughout data collection and analysis. Reflexivity was maintained through bracketing and reflective memoing. Bracketing is a technique where the researcher sets aside their own pre-conceived assumptions and knowledge prior to the study starting. Conversely, reflective memoing is a process where a researcher documents their own assumptions and biases that develop during data collection and analysis and reflects on whether these assumptions are introducing bias into the study. A pilot interview helped refine neutral questioning and ensured that the developed interview guide did not reflect potential clinician bias. The interviewer emphasized confidentiality and non-evaluative listening to mitigate potential dynamics that might influence participant responses. Analytical rigor was supported by maintaining an audit trail of coding decisions, ensuring transparency and reproducibility. Regarding data interpretation and coding

Results

Twelve health system stakeholders participated, including primary care physicians (PCPs; n=3), oncologists (n=6), a gastroenterologist, a social worker, and a cancer center administrator. Participant characteristics are summarized in Table 1.

Table 1:

Participant Demographics

Characteristics	N=12
Age, median (IQR)	55 (38 – 65)
Gender, n (%)
Female	9 (75)
Male	3 (25)
Race, n (%)
Black	1 (8)
White	8 (67)
Asian	2 (17)
Hispanic	1 (8)
Role, n (%)
Medical Oncologist	5 (42)
Surgical Oncologist	1 (8)
PCP	3 (25)
Social Worker	1 (8)
Administrator	1 (8)
Gastroenterologist	1 (8)
Practice Setting, n (%)
Community based	5 (42)
Academic/Safety-net hospital	7 (58)
Years in practice, n (%)
<15	4 (33)
≥15	8 (67)

Barriers to Foregut Cancer Care

Emerging barrier themes were clustered across four levels: (1) Individual, (2) Provider, (3) Institutional, and (4) Policy. Key themes and representative quotes are summarized in Table 2. The most frequently reported barriers included transportation challenges, financial toxicity, insurance status-related access barriers, and delays in diagnostic work-up.

Table 2.

Stakeholder Reported Barriers to Foregut Cancer Care

LEVEL	THEME	QUOTE
Individual	Financial toxicity	(Q1) “The biggest thing I see is the inability to get the medication they need…. They can’t afford it.” (49F, community PCP) (Q2) “Then there’s the co-pay that patients are often blind-sided about. I mean, they know it will cost, but they often aren’t prepared for how much it will cost. Or even if they know, there’s nothing they can do to arrange for those amounts of money.” (36F, social worker) (Q3) “I’ve had patients who can’t afford the copay for an EGD or a CT scan. We know we order the appropriate tests, but they get denied or delayed because the patient can’t pay.” (54F, community gastroenterologist)
	Transportation challenges	(Q4) “Then transportation’s a big one still. I think sometimes they rely on rides from other people, so there are patients who miss treatment because the person they relied a ride on was not able to take them, or their car breaking down or just not being able to get to clinic for whatever reason.” (38F, academic medical oncologist) (Q5) “Some of my older patients don’t have family nearby, or anyone who can take them. If they don’t have someone to drive them to appointments, they just won’t go.” (57F, community PCP) (Q6) “Many of our patients who come locally rely on public transit, which can be unreliable. But the bus service is sporadic, some days it just doesn’t run.” (36F, social worker) (Q7) “Some patients travel three to four hours to get here, and when they’re juggling multiple appointments and coming on multiple different days, it can become overwhelming. (36F, social worker)
	Psychosocial distress	(Q8) “And then there’s just the sheer emotional burden. Cancer treatment is grueling, and some patients start to lose hope or feel it’s not worth the effort when every step—appointments, tests, side effects—feels like an uphill battle.” (66F, community PCP with academic affiliation) (Q9) “As they change from patient to survivor because that’s a big transition when they’re like I’m a survivor now I’m not a cancer patient anymore. Yes, you went through it, but you lived, and then they probably have seen friends that they met in the infusion suite that didn’t make it. It’s always that guilt…” (31F, academic medical oncologist)
	Patient denial	(Q10) “They feel like it’s going to go away…. So, some of it is denial, and some of it is fear, fear of what they may hear if they report it.” (65M, academic medical oncologist)
	Mistrust of the healthcare system	(Q11) “I think also there’s a distrust in some patients of going into an environment that they’re unfamiliar with. It felt like it was dangerous to travel outside of their little rural area. (66F, community medical oncologist) (Q12) “Many of my patients do not trust the healthcare system. But you can’t blame them. Some of that is historical, some of it is personal experience. They feel like doctors don’t listen to them.” (57F, community PCP)
	Competing responsibilities	(Q13) “Family responsibilities are another barrier. Patients have to juggle treatment with taking care of their kids or other relatives. I’ve had caregivers tell me they want to stop treatment because they can’t make it to appointments because they have to look after someone at home, like an elderly parent or spouse, or sometimes even grandchildren.” (66F, community PCP with academic affiliation)
	Employment constraints	(Q14) “Sometimes still finances do get in the way because sometimes they have to be on disability or leave their jobs to get treatment, even if they were working. (38F, academic medical oncologist) (Q15) “That’s the other big thing, them having to go to work because sometimes our work doesn’t excuse them from coming to treatment, even though we fill out the paperwork. (31F, academic medical oncologist)
	Low health and technology literacy	(Q16) “The other is that we have non-compliance from people who are under-educated, and so they don’t fully understand, and it’s difficult to get them to understand the importance of staying with the treatment regimen.” (67F, academic administrator) (Q17) “Some patients don’t know how to begin to use the internet and would never make it through the reams of paperwork necessary to apply for Medicaid or disability.” (66F, community PCP with academic affiliation)
Provider	Inadequate physician-patient communication	(Q18) “As far as providers… we don’t explain to them their situation and in language that they understand.” (49F, community PCP) (Q19) “You know I can’t count the number of patients who come back from their oncology appointment and haven’t understood anything of what they’ve been told. But they don’t say anything, they just ask me what to do…. And then I’m trying to contact the oncologist to get their notes, or to talk to them so that I can be the medical translator for the patient.” (57F, community PCP)
	Insufficient inter-provider communication	(Q20) “I don’t get notes. I didn’t know what was going on with the patient. I’ve had several patients who came to me after they had already completed chemotherapy. (49F community PCP) (Q21) “There was no doctor, nobody that I had a relationship with anywhere that I could get anybody into any hospital, even if I spent eight hours on the phone and end up in tears because of it, begging.” (66F, community PCP with academic affiliation)
	Limited time for comprehensive barrier assessment	(Q22) “I don’t think it’s that colleagues don’t want to do it [barrier assessment]. I don’t think they have time to do it.” (65M academic oncologist)
Institutional	Delays in diagnostic work-up	(Q23) “The thing that I’ve seen the most would be the length of time to get appointments with either primary care doctors to start the evaluation for cancer-related symptoms…” (65M, academic oncologist) (Q24) “I think sometimes the problem with referrals is that it might be a referral of a mass that was seen, but then there was no workup done on it for a year and then the patient shows up.” (38F, academic oncologist) (Q25) I’ve had patients who can’t afford the copay for an EGD or a CT scan. We know we order the appropriate tests, but they get denied or delayed because the patient can’t pay.” (54F, community gastroenterologist)
	Delays in accessing specialist care	(Q26) But getting them there in a timely fashion is a huge challenge. The referral process can be a nightmare—appointments are booked out for weeks, sometimes months, and in the meantime, we’re sitting on a suspected cancer diagnosis with no way to expedite things.” (54F, community gastroenterologist)
	Insurance status-related access barriers	(Q27) “I’ve had patients who needed a PET scan for staging, only for insurance to deny it because they want a CT first. And when you try to get the CT, they want you to do an X-ray first. So much back-and-forth, you know. It can add weeks or even months.” (54F, community gastroenterologist) (Q28) “But I’ve had patients with insurance who delayed coming in because they knew they couldn’t afford the next steps, because of large co-pays and the subsequent bills that they would get because their insurance wouldn’t cover so many things.” (66F, community PCP with academic affiliation) (Q29) “If a patient has Medicaid, they may not be able to go to some places because it’s out of network. I’ve had to send patients two hours away just to find a specialist who takes their insurance.” (54F, community gastroenterologist) (Q30) “The underinsured group can be more challenging because they don’t qualify for full charity care but still can’t afford their out-of-pocket costs.” (54F, community gastroenterologist) (Q31) “They were waiting on their insurance to come through, and then they both ended up dying. The insurance network came through, but I mean they were just waiting, and that’s so frustrating because I think every patient, regardless of insurance, that is, should be afforded care at the same time.” (31F, academic medical oncologist)
	Institutional resource limitations	(Q32) “Affiliated services- physical therapy, occupational therapy, nutrition, psychology, all the supportive programs. We don’t have any of those in a reliable way at ___. They would all help.” (66F, community PCP with academic affiliation) (Q33) “The barriers on the physician group, or on our side as the providers, are available state-of-the-art testing and appointments for CT, MR, PET scan.” (62M, academic surgical oncologist)
	Overburdened Providers	(Q34) “ They overwork the providers, and then the level of care drops at that point when you have too many patients on your schedule.” (65M, academic oncologist) (Q35) “[Dr. Name] has so much value in practice, in knowledge, in enthusiasm, but he’s going to burn himself out. He’s got to slow down…. I was like [Name] at one point. Do everything, say yes to everybody who wants you to do everything, and next thing you know, I’m not doing well emotionally.” (65M, academic oncologist)
	Scheduling inefficiencies	(Q36) “A lot of times I see physicians saying, I’ll see them next week on so-and-so day, but they may have an appointment with another doc at the cancer center the day before, and they may drive a distance.” (65M, academic oncologist)
	Lack of transparency regarding system changes	(Q37) “I think transparency is what’s really lacking here. It’s like the troops on the ground need to know what the general pathway is, and I think we get little glimmers of it, but we’ll find that something’s changed, and we didn’t even know it was changed.” (65M, academic oncologist)
	Lack of clinician input into system planning	(Q38) “I think involving the appropriate clinicians in development of strategies for how to manage patients in general…this is a criticism of the system…the system tends to pick and choose what physician support they allow into the planning phase” (65M, academic oncologist)
	Frequent staff turnover	(Q39) “…but trying to maintain the turnover ratio—once you teach somebody how to do something, they become a pretty valuable employee.” (65M, academic oncologist)
	Inadequate health system infrastructure	(Q40) “That’s something that we deal with every day, even within this system, because we’re trying to grow, but we don’t have space to grow. So, how do you fix that? It’s hard.” (31F, academic medical oncologist)
Policy	Care fragmentation	(Q41) “In this community, fragmentation of care is worse than any community I have ever worked in. …If you can’t get access at one institution, you just go down the street to another institution where they have none of your records. Your doctors aren’t there. The workups start all over again. They may find conflicting information. When they do, the records don’t come across. Then the patient is confused as to who’s right and who’s not right.” (65M, academic oncologist)
	Inter-institutional competition	(Q42) “Get rid of the concept of competition… I think we all ought to learn to play in the sandbox, and out of that could come some wellness for.” (65M, academic oncologist)
	Inadequate sub-specialty care in rural areas	(Q43) “But you refer to GI in the community, and they don’t have that many advanced people doing all the fancy things you need for a biopsy of a pancreatic cancer. So, then they refer to another GI. The end result is also a delay.” (66F, community PCP with academic affiliation)
	Inadequate physician and health system reimbursement	(Q44) “There aren’t enough specialists like you, and the few we have are stretched thin. And reimbursements are bad in Alabama, much worse than some other states. So why would people like you come here when you could be elsewhere and make more money? It’s why hospitals are closing all across rural parts of Alabama.” (66F, community PCP with academic affiliation)

Individual Level Barriers

Barriers at the individual level included financial toxicity, transportation challenges, psychosocial distress, competing responsibilities, employment constraints, low health and technology literacy, patient denial, and mistrust of the healthcare system. Health system stakeholders reported that financial toxicity stemmed from the cost of essential medications, co-pays and deductibles, and diagnostic investigations (Q1- Q3). Transportation challenges were widespread due to the limited availability of reliable vehicles, limited social support, and poor public transport infrastructure (Q4- Q6). The frequency of medical appointments and travel distances to tertiary centers further magnified this burden (Q7). Health system stakeholders described substantial patient psychosocial distress from the cancer diagnosis and debilitating treatment regimens (Q8) and survivorship guilt (Q9). Patient denial of symptoms (Q10) and mistrust of the healthcare system that was driven by prior negative experiences or lack of personal connection (Q11- Q12), further impeded care. Competing family responsibilities (Q13), as well as employment constraints including loss of employment, income disruption, and inflexible work schedules (Q14- Q15), further limited treatment adherence. Low health and technology literacy also hindered patients’ understanding of their disease, treatment plans and insurance requirements (Q16- Q17).

Provider Level Barriers

At the provider level, health system stakeholders highlighted communication gaps and limited time for comprehensive barrier assessment as key barriers. PCPs emphasized that inadequate physician- patient communication, including the use of medical jargon, insufficient explanations and rushed consultations resulted in patient uncertainty about their diagnosis and treatment plans (Q18). Insufficient inter-provider communication and coordination between primary and specialty care was described as a persistent issue, with PCPs often forced to act as intermediaries to fill information gaps, despite rarely receiving feedback or follow-up communication from oncology specialists (Q19- Q20). Delays in patient referral completion were described as a persistent issue (Q21). Time constraints and heavy clinical workloads limited physicians’ ability to screen for social needs or assess patient-level barriers (Q22).

Institutional Level Barriers

At the institutional level, participants described barriers related to delays in diagnostic workup, insurance status-related access barriers, restricted specialist access, institutional resource limitations and organizational challenges. Delays in diagnostic work-up were attributed to lack of access to primary care (Q23), limited provider familiarity with work-up and diagnostic testing (Q24), insufficient follow-up systems, and the cost of diagnostic testing (Q25). Delays in accessing specialist care were attributed to weak referral pathways and limited availability of specialist appointments at tertiary centers (Q26). Insurance status-related access barriers were multifactorial, including prior authorization delays, denials, high out-of-pocket costs, and limited coverage of insurance plans including Medicaid (Q27- Q29). Lack of local institutional resources to help underinsured patients resulted in these patients needing to travel longer distances to access care (Q30), while uninsured patients also experienced substantial treatment delays (Q31).

Participants reported that institutional resource limitations hindered the delivery of effective foregut cancer care, citing shortages of social workers, navigators, and diagnostic imaging facilities (Q32- Q33). Overburdened providers reported high patient volumes prevented the delivery of appropriate care (Q34) and led to burnout (Q35). Scheduling inefficiencies resulted in patients having to attend multiple appointments on different days, compounding transportation and financial burdens (Q36). Finally, institutional culture and infrastructure posed additional challenges, including a lack of transparency regarding system changes (Q37), a lack of clinician input into system planning (Q38), frequent staff turnover (Q39), and inadequate health system infrastructure to support complex cancer care (Q40).

Policy Level Barriers

At the policy level, participants cited care fragmentation, inter-institutional competition, inadequate sub-specialty care in rural areas, and inadequate physician and health system reimbursement as barriers. Fragmentation of care, driven by a lack of access to services at institutions and incompatible electronic health record (EHR) systems, and compounded by a lack of inter-provider communication, created duplications and inefficiencies (Q41). Competition between regional healthcare systems discouraged collaborative care (Q42). Rural subspecialty shortages, particularly in gastroenterology and interventional radiology (Q43), were attributed to low physician reimbursement and underfunding of community hospitals (Q44).

Proposed solutions

Participants emphasized that effective solutions must address barriers at multiple levels, with a focus on building system capacity rather than relying on individual providers to improvise. Health system stakeholder proposed solutions are summarized in Table 3 and illustrated in Figure 1.

Table 3.

Stakeholder Proposed Solutions

LEVEL	BARRIER THEME	PROPOSED SOLUTIONS	QUOTE
Individual	Financial toxicity	Expanding ancillary staff	(Q45) “We desperately need social workers. There needs to be 5 more of me. And we need to have navigators who can guide patients through every step of their care.” (36F, social worker)
	Transportation challenges	Governmental grants	(Q46) “I mean that’s a step moving forward and trying to get those grants for patients that are far away to be able to stay somewhere local.” (31F academic oncologist)
		Temporary lodging near treatment centers	(Q47) “I think that the hospital could consider providing transportation.” (49F, community PCP)
		Improving patient education on existing services	(Q48) “And then give them information if they’re able and they can ride the bus, give them information about the wave if they’re on Medicare” (49F, community PCP)
	Psychosocial distress	Strengthening institutional psychiatric support programs	(Q49) “The cancer center doesn’t have a therapist. Emotional support programs are needed, and not that hard to establish. (66F, community PCP with academic affiliation)
Provider	Inadequate patient-physician communication	Improving physician- patient communication	(Q50) “The specialists at the big hospitals need to do a better job of making real connections with patients. Patients don’t just need expertise, they need trust.” (57F, community PCP)
	Insufficient inter-provider communication	Inter-provider communication & collaboration	(Q51) “It would also help if specialists worked more closely with primary care doctors. In small towns, we build relationships with our patients over years, so they trust us. If we had better communication between specialists and community doctors, I could help bridge that gap…. But if I had a direct line to their oncologist, I could reassure them and help them understand what’s going on.” (57F, community PCP)
	Limited time for comprehensive barrier assessment	Standardized barrier screening tools	(Q52) “At least there should be a screening process and then bring things to the attention of the doctor. [barrier assessment]”. (66F community oncologist)
Institutional	Delays in diagnostic work-up	Improving access to primary care and educating alternative providers	(Q53) “[nurse practitioners] But they need better education so that they can start some of these workups themselves. The training is not enough, but with apprenticeships, it can get there.” (66F, community PCP with academic affiliation)
	Delays in accessing specialist care	Increasing inter-provider engagement	(Q54) “Having an established relationship with specialists, especially surgeons, is really good. When I have a good working relationship with a surgeon like you, I can reach out directly and say, This patient really needs you, they to get in sooner—what can we do? That often speeds up the process.” (54F, community gastroenterologist)
		Streamlining referral pathways	(Q55) “So, what we ended up doing was I created a call center, and we automated. We’ve made it very user-friendly.… We get the very basic, then we call the office and get them to send over the information.” (67F, academic administrator)
	Insurance status-related access barriers	Leveraging technology platforms	(Q56) “It’s called Assist Point. It’s a digital platform…Assist Point will match that patient based on their disease and their insurance to potential co-pay programs or foundations that they can get assistance from.” (67F academic administrator)
	Scheduling inefficiencies	Centralized navigation	(Q57) “When you call, and you have a patient that needs to see the radiation oncologist, needs to see the medical oncologist, needs an MRI or a PET CT, or a bone scan, whatever, it all needs to happen on the same day…. So, I don’t know how, but it needs to be central. Involving radiology, cardiology, subspecialties of internal medicine, and oncology. [patient navigation].” (66F, community oncologist)
	Overburdened Providers	Physician wellness initiatives	(Q58) “I think physicians need to be healthy, and we’re not healthy as a group….one of the things I’ve proposed that we do is to try to develop a system where doctors don’t feel like they can’t take time off to refresh and to renew, because we’re telling patients how to be healthy, but we’re unhealthy ourselves. So, I think improving the health of the doctors, improving the ability to be able to say, I’m going to take a week off with my grandchildren and not feel guilt from it and know that it’s taken care of.” (65M, academic oncologist)
Policy	Care fragmentation	Standardized EHR	(Q59) “The idea of having a community electronic health system…. I think we should be able to interact across.” (65M, academic oncologist)
	Inadequate sub-specialty care in rural areas	Targeted physician recruitment and expansion of specialty outreach clinics in the community	(Q60) “Also work with the community to invest in recruiting community docs…. It’s hard to find doctors willing to practice in rural areas, maybe some sort of community practice by the doctors at the big hospital, that will also be a good idea.” (54F, community gastroenterologist)
	Inadequate physician and health system reimbursement.	Improved governmental re-imbursement	(Q61) “But we really need improved access. Medicaid expansion would be a start. More funding for primary care, more funding for rural health centers. Better reimbursement so that hospitals could stay afloat in rural areas, that would also help more specialists stay in the area. If they were paid better, comparably to other parts of the country, and if the hospitals had more infrastructure so that the specialists could do their jobs there.” (66F, community PCP with academic affiliation)

Figure 1.

Multilevel Framework of Health System Stakeholder-Identified Barriers and Proposed Solutions to Foregut Cancer Care

To address financial toxicity, health system stakeholders proposed expanding ancillary staff, including social workers and navigators to assist patients in navigating financial aid programs (Q45). Potential transport solutions included governmental grants (Q46), temporary lodging near treatment centers (Q47), and improving patient education on existing services (Q48). To mitigate psychosocial distress, participants recommended expanding institutional psychiatric support programs, recruiting additional mental health professionals, and integrating psychosocial assessment and support into care pathways (Q49). PCP stakeholders highlighted the importance of improving physician-patient communication by dedicating time to relationship building and patient education, fostering trust, and increasing collaboration between specialty and primary care (Q50- Q51). Health system stakeholders also supported the development and integration of standardized barrier screening tools to facilitate early identification of needs (Q52).

Institution-level solutions emphasized leadership and structural reforms. Proposed strategies included educating non-physician providers on appropriate initiation of cancer workup (Q53), increasing inter-provider engagement (Q54) and streamlining specialist referral pathways (Q55). Leveraging technology platforms such as AssistPoint to connect eligible patients with financial aid (Q56) and establishing centralized navigation to reduce scheduling inefficiencies (Q57) were also recommended. Providers underscored the need for physician wellness initiatives to address burnout and sustain workforce capacity (Q58). At the system level, participants advocated for a standardized EHR across health systems or an interoperable system that facilitates data exchange, enhances information sharing, and reduces the impact of care fragmentation (Q59). To address rural subspecialty shortages, participants recommended targeted physician recruitment and expansion of specialty outreach clinics in the community (Q60). Finally, Medicaid expansion, as well as improved governmental reimbursement and funding for both physicians and health systems, were viewed as essential for equitable cancer care delivery (Q61).

Discussion

This qualitative study examined multilevel barriers to foregut cancer through the perspectives of clinicians, administrators, and ancillary staff at a safety-net institution and its surrounding catchment area in the Southeastern United States. Despite the importance of health-system stakeholder insights, no studies to date have explored them in the context of foregut management at large academic safety-net institutions. In this study, participants described how structural, organizational, and interpersonal factors converged to impede equitable access to timely guideline concordant treatment for patients with foregut cancers, but also identified pragmatic solutions grounded in their lived realities as they cared for patients.

An underlying theme consistently highlighted was the interdependence among domains: financial hardship and transportation challenges that patients experienced were compounded by institutional inefficiencies, limited communication and under-resourced systems. In the stretched systems where the participants delivered care, the responsibility for overcoming these barriers often fell to patients and their frontline care team, rather than the system itself. At the patient level, health system stakeholders emphasized the pervasiveness of financial hardship reflecting embedded socioeconomic deprivation, underinsurance and the cumulative burden of treatment-related costs. Healthcare expenditure is a leading cause of bankruptcy declared by individuals in the U.S., and patients with foregut cancer face disproportionately high procedure-related and intensive systemic therapy expenses ^[16–19]. Even insured patients faced high deductibles, copays, and insurance denials, creating cascading delays in diagnosis and treatment. While centralization of surgical care for foregut cancers is associated with improved outcomes, it has also increased travel distances for patients ^{[20, 21]}, particularly in the Southeastern U.S where public transportation infrastructure is sparse ^{[22, 23]}. These barriers contribute to missed appointments, treatment discontinuation, increased hospitalizations, and decreased survival ^{[22, 24–26]}. Participants emphasized the need for solutions that extend beyond individual assistance to broader, structural responses including expanded navigation and social work, as well as institutional support for transportation and temporary housing near treatment centers ^[27]. Psychosocial distress was frequently described as a barrier, consistent with estimates that up to half of gastrointestinal cancer patients experience depression or anxiety ^[28–31]. However, they reported limited access to psychiatric and support services locally and expressed a need for expanding institutional counselling support and integrating psychosocial assessment and support into care pathways.

At the provider level, inadequate physician-patient communication and insufficient inter-provider communication and coordination emerged as dominant barriers. PCPs described functioning as “medical translators” for their patients, who left oncology appointments uncertain about their diagnosis or treatment plan. Yet, the same PCPs reported receiving little communication from relevant specialists, limiting their ability to bridge these gaps effectively. These findings are consistent with our patient- and caregiver-focused research, which also identified inadequate specialist physician communication as a barrier ^{[10, 32]}. Strengthening bidirectional communication through standardized referral feedback and shared EHR notes could substantially improve coordination. Institution-wide communication training emphasizing plain language, use of shared decision-making tools^[33], and community outreach may further enhance patient understanding and trust.

Institutional barriers included delays in diagnostic workup and access to specialist care, often driven by unclear referral pathways, limited appointment availability and inadequate follow-up. Community PCPs reported difficulty referring patients to academic centers citing long waits and a lack of standardized referral pathways. Proposed solutions include streamlining of referral pathways, centralized navigation to coordinate appointments and reduce scheduling and transportation burdens ^[34–36]. Enhancing education for PCPs and advanced practice providers on early, appropriate diagnostic work-up was also suggested as a way to shorten delays and expedite referrals. Leveraging technology platforms and centralized navigation could help tackle gaps in care and address insurance status-related access barriers ^[37].

Participants emphasized that structural barriers at the policy level resulted in, and compounded local challenges, highlighting care fragmentation, inadequate sub-specialty care in rural areas, and inadequate physician and health system reimbursement. Non-standardized EHR systems were viewed as a major driver of fragmentation, limiting information exchange between institutions. Shortage of subspecialists in rural areas, compounded by poor reimbursement and rural hospital closures^{[38, 39]}, further limited patient access and strained existing systems. Proposed solutions included standardizing EHR systems to improve continuity of care and minimize duplication of investigations. Other practical interim measures could include providing community physicians with limited access or view-only entry into larger health systems’ EHR platforms for patient notes, laboratory results, and imaging. Although not a full substitute for interoperability, such systems may be an effective way to enhance communication and support continuity of care across institutions. Additional solutions include expanding specialty outreach clinics at rural and satellite community hospitals^{[40, 41]} and improving reimbursement policies to better support safety-net and rural institutions.

Across all levels, our findings reveal a system where patients and frontline clinicians absorb the consequences of structural disorganization. Financial and logistical burdens are compounded by institutional inefficiencies and policy constraints, creating cumulative disadvantage for patients already navigating socioeconomic hardship. Clinicians try to fill in the gaps for patients, reflecting both professional dedication and system fragility that relies on strained providers, individual improvisation, and ad hoc workarounds to provide day-to-day care.

This study contributes to the literature by capturing perspectives from a broad range of frontline health system stakeholders who are rarely represented in the literature. These participants, working in rural and safety-net contexts, provided a nuanced understanding of why patients fail to access timely, guideline-concordant care, and what practical strategies might help close these gaps. The inclusion of PCPs, who often have the closest longitudinal relationships with patients, added depth to the discussion of referral challenges and communication gaps that are often underappreciated in oncology-focused research. However, several limitations should also be considered. The sample size was modest and drawn from a single safety-net system and its catchment area, which may limit generalizability. However, this reflects the limited number of providers involved in foregut cancer care in the region, and allowed for in-depth, contextually rich insights. The surgical oncology expertise of the interviewer likely shaped the interviews in both positive and limiting ways: the interviewer’s clinical expertise may have facilitated openness as participants may have been more comfortable discussing operational issues with a peer who understood the context of cancer care delivery. On the other hand, this may also have influenced the narratives shared. Finally, as with all qualitative studies, findings represent perceived rather than measured barriers. Future mixed-methods studies should triangulate these data with patient and administrative sources to assess the prevalence and impact of identified barriers. Building on these findings, future studies should co-design and evaluate multilevel interventions identified here, such as centralized navigation, specialty outreach clinics in the community, and expanded psychosocial support, using implementation frameworks to determine the feasibility and scalability of health system stakeholder-proposed interventions, as well as their association with patient-level outcomes.

In conclusion, multilevel barriers continue to limit access to timely, high-quality foregut cancer care, particularly in underserved and rural communities. By incorporating the perspectives of community clinicians and health system stakeholders, this study identifies actionable solutions, including centralized patient navigation, expanded institutional psychiatric support programs, improving communication, streamlining referral pathways, and policy-level reforms, that could improve access and outcomes. Centering these frontline insights will be essential to developing interventions that not only are evidence-based but also practical and sustainable in the environments where they are most needed.

Data Statement:

Non-submitted data includes sensitive or confidential information.