Rural oncologists’ perceptions of specialty scarcity and repercussions for care delivery: a qualitative study

Abstract

Purpose:

To understand oncology physician perceptions of and experiences with specialist scarcity in their referral networks, strategies for delivering care following the departure of a colleague who they view as critical to their cancer care networks (i.e, a “linchpin” colleague), and impacts of shortages on patient care.

Methods:

We conducted semi-structured interviews with oncologists who practice in health systems that serve a predominantly rural patient catchment area. We used deductive and inductive approaches to pre-determine codes and then performed a thematic analysis.

Results:

We interviewed 20 oncology physicians from five sites. We identified three major themes related to specialist scarcity. The first theme described the effects of physician shortages on care team expertise, collaborative relationships, and patient volume. The second theme uncovered strategies oncologists use when facing physician shortages, including referrals to outside health systems or generalists, practicing outside their sub-specialization, and reallocating time from other responsibilities. The third theme identified unintended consequences of adaptive strategies, including greater patient travel burden, less optimal or delayed treatment, reduced access to clinical trials, and increased physician burnout and lower job satisfaction.

Conclusion:

Oncology physician shortages lead to myriad adaptive strategies and downstream consequences to patient and physicians. Mapping these cascades can help guide resources to mitigate the negative effects of departures and shortages.

Introduction

Geographic variation in oncology workforce shortages in the United States are well-documented, with one study reporting that 64% of counties had no oncologists with a primary practice location in that county.^1–3 Oncology workforce shortages have the potential to disrupt local and regional referral networks, ^4,5 and are particularly acute in rural areas.^6–8 Oncologists practicing in rural areas face unique challenges, including difficulty coordinating care with geographically distant providers and less redundancy in physician sub-specialization.⁶ As the referral networks for physicians in rural areas are often characterized by specialist scarcity,⁹ they can be particularly vulnerable to workforce shortages.

The importance of multidisciplinary relationships in cancer care has motivated the use of social network analysis to study the physician relationships that connect the oncology workforce and examine the associations between physician networks and patient outcomes. One quantitative approach for inferring relationships between physicians is to use patient-sharing networks assembled from Medicare claims data.^10,11 Prior work explored the associations between patient outcomes and a network-based measure of specialist scarcity called physician “linchpin score”.^12,13 A higher linchpin score for a medical oncologist, for instance, is generated when few of the physicians they share patients with are connected to another medical oncologist. These networks are hypothesized to be more disrupted if the medical oncologist with high linchpin score were to depart (e.g., through retirement or relocation), as the remaining physicians have fewer established ties to another medical oncologist. Using this linchpin score approach, studies found that patients with cancer who resided in areas with a high proportion of linchpin oncologists (i.e., oncologist scarcity) were more socioeconomically disadvantaged and less likely to receive multidisciplinary consultations following diagnosis.^12,14 Patients with cancer who were treated by linchpin oncologists have also been shown to have more delayed care and worse survival.^13,15

We have a limited understanding of physician perceptions of colleagues they consider to be “linchpins” in their own networks and referral behaviors in the context of actual or anticipated workforce shortages. Understanding the experiences and adaptive strategies of physicians in networks with specialist scarcity can inform interventions and policies that help create and support resilient referral networks. Given the greater risk of workforce shortages in rural areas and the unique challenges rural physicians face when coordinating care, the perspectives of these physicians are particularly important. The objective of this study was to understand the perceptions and experiences of oncology physicians with specialist scarcity in their referral networks, strategies for overcoming disruptions in the delivery of oncology care when a linchpin colleague departs, and potential or real impacts on delivery of cancer care to their patients.

Methods

Study Design

This qualitative study is part of a larger study that examined patient-sharing networks within adult oncology to evaluate access to cancer care. The research team included health services researchers, a practicing oncologist, a medical anthropologist, a biostatistician, and experts in qualitative research design and implementation. We used a phenomenological qualitative study design to explore the experiences of oncologists regarding specialist scarcity in cancer care through semi-structured interviews at five rural sites susceptible to oncologist shortages. This approach emphasizes individuals’ subjective experiences and meanings, complementing existing quantitative research by capturing the complexities of physicians’ decision-making processes and the impacts of those decisions when faced with shortages. The study was approved by the Institutional Review Board at Dartmouth College and adheres to the Consolidated Criteria for Reporting Qualitative Studies (COREQ).¹⁶

Recruitment

We aimed to recruit medical, surgical, and radiation oncologists practicing at rural cancer care facilities who had experience caring for rural-residing patients. Participants were recruited by both purposive and snowball sampling to reach our goal of five participants per site to facilitate an analysis of cancer care networks from multiple perspectives. We first invited oncologists by email to participate in the study through the existing professional networks of our research team. We purposively selected oncologists to ensure representation across oncology specialties. We then asked participating oncologists to refer additional oncologists in their network as needed to reach our goal of interviewing multiple oncologists in each network. Participants gave informed consent to take part in the study prior to the start of the interview.

Interview

We created a standard semi-structured interview guide consisting of four sections (Table S1). The first asked oncologists about physicians they work with while providing cancer care and who they would be concerned about if they became unavailable (e.g., a possible “linchpin” in their referral network). The second focused on how oncologists perceive their own linchpin status within their network. The third explored referral preferences and practices, and the fourth addressed general perceptions of clinician shortages in their networks and strategies of healthcare administration and leadership to address shortages. One qualitative expert conducted all the interviews via Zoom, which were recorded, and most interviews lasted 30 minutes.

Data Analysis

We transcribed interviews and imported them into Dedoose for analysis. CMT and KES led an iterative analysis and incorporated both deductive and inductive approaches. We pre-determined some codes based on the study’s defined focus and identified additional codes through in-depth review of the data. During the developmental stage of the codebook, CMT created case memos for each interview, which helped us identify the initial inductive codes . CMT coded two interviews which were reviewed by KES to assess code applications, definitions, and to jointly revise the codebook. CMT used the revised codebook to code three additional interviews which were reviewed by KES for additional revisions. Subsequent coding, which consisted of primary coding by CMT, and iterative review by KES, and a full research team meeting to review the codes, led to the final codebook, consisting of 42 codes. Periodic reviews by the entire research team supported the credibility and transferability of the coding and analysis, in line with best practices in qualitative research.¹⁷ Themes were developed through analysis of code counts and related excerpts, code co-occurrence, and analytic memos. Themes were reviewed by the whole research team for clarity and collaborative confirmation.

Results

We interviewed 20 oncologists with varying professional and personal backgrounds (Table 1). Study participants included 8 (40%) medical oncologists, 5 (25%) radiation oncologists, 4 (20%) surgical oncologists, and 3 (15%) other oncology specialties. Most participants were male (60%) and White (65%). Study participants came from five sites, but most worked at three NCI-designated comprehensive cancer centers with a rural patient catchment area (Table 2). The individual interviews from two other rural cancer care facilities provided insights into different site dynamics – though these insights are limited to individual perspectives.

Table 1.

Study participant characteristics

Physician characteristics	n, %
Specialty
Medical oncologist	8 (40)
Radiation oncologist	5 (25)
Surgical oncologist	4 (20)
Other oncologist	3 (15)
Years in Practice
Under 5	4 (20)
Between 5 and 9	6 (30)
Between 10 and 14	5 (25)
Over 15	5 (25)
NCI Cancer Center Affiliation
Yes	19 (95)
No	1 (5)
Age (years)
30-39	3 (15)
40-49	10 (50)
50-59	4 (20)
60+	3 (15)
Gender
Man/male	12 (60)
Woman/female	8 (40)
Race/ethnicity
White / Caucasian	13 (65)
Middle Eastern or North African	3 (15)
Asian	2 (10)
Black / African American	1 (5)
Hispanic, Latino, or Spanish origin	1 (5)

Table 2.

Cancer Care Sites

Site and US Region	Number of participants	NCI Cancer Center
Site A: Northeast	7	Yes
Site B: Southwest	6	Yes
Site C: West	5	Yes
Site D: Southeast	1	Yes
Site E: Southeast	1	No

Theme Overview

Herein, we consider “physician shortages” in the context of a real or hypothetical departure of a colleague considered critical in the participant’s care delivery network. We identified three major themes related to linchpin oncologists and shortages (Table 3), including: 1) effects of physician shortages on care team expertise, collaborative relationships, and patient volume; 2) strategies oncologists adopt when facing physician shortages; and 3) unintended consequences of adaptive strategies on patients and remaining physicians. The inferred relationships between theme components are visualized in Figure 1.

Table 3.

Main Themes and Theme Components

Theme	Theme Components
Physician shortages impact care team expertise, collaborations, and patient volume	Physician shortages can limit access to specialized expertise, such as oncological sub-specialization (e.g., specific cancer types and treatment methods), and they disrupt collaborative relationships and interrupt existing workflows. They also increase patient volume for remaining physicians, who may already be at or near capacity.
Adaptive strategies in response to physician departures	Oncologists respond to physician losses in several ways. They may refer patients externally, modify triage decisions, take on additional workload, reallocate time from other duties to patient care, or refer to a generalist instead of a specialist.
Unintended consequences of adaptive strategies	Unintended consequences of adaptive strategies can disrupt both the quality and continuity of patient care, as patients encounter delays, changes in treatment plans, increased travel burdens, and limited access to clinical trials. Remaining physicians face the strain of increased patient volume, diminished multidisciplinary support, and logistical challenges.

Figure 1.

The inferred relationships between theme components.

Theme 1: Physician shortages affect care team expertise, collaborations, and patient volume

Participants’ perceptions of real or hypothetical departures of oncologists they considered most critical to their cancer care networks highlighted three key impacts that included losses/gaps in network expertise, losing established and collaborative relationships, and decreased capacity of networks to manage patient volume.

Physician Expertise Loss

Participants emphasized that caring for complex or rare cancer cases required expertise, intellectual support, and insights from those with extensive experience and sub-specialization. Even when an oncologist’s network contained multiple oncologists of the same type , sub-specialization in a specific cancer type or treatment method meant that the loss of a physician could create a significant gap of expertise. One participant described this challenge using an example of when a medical oncologist specializing in melanoma left:

“It was kind of a disaster. I mean, then what you had to have is different providers who weren’t per se, focused on a particular cancer type, then had to pick up that cancer type in those patients… I would have trouble as a breast oncologist, 11 years into my practice, treating melanoma.”

Participants also, though less frequently, highlighted the importance of the sub-specialization among physicians who support cancer treatment plans but are not oncologists themselves:

“So, there’s a particular cardiologist who specializes in cardio-oncology that I’ll send my heart failure patients to, because some of our drugs cause heart failure. And there’s a particular orthopedic surgeon who specializes in oncology… [if] I’m not sure if they need a hip replacement because of cancer in their hip, I’ll send to him.”

Loss of Collaborative Relationships

Participants identified four key ways a colleague’s departure would disrupt collaborative relationships. One centered on the value of established communication channels and collaboration, including being able to contact colleagues when needed and a shared willingness to collaborate. Second, participants noted that proximity to colleagues facilitates regular in-person discussions and can strengthen collaboration. The third reason is workflow efficiency, whereby established collaboration created shared processes, such as knowing preferred workups which enabled participants to work more effectively and reduce administrative time. Finally, participants emphasized the importance of trust and confidence in the clinical judgment of their colleagues. One participant spoke about the importance of consistency and predictability in referrals:

“I think it’s consistency. It’s knowing how someone practices and being able to predict what they’re going to do or having trust in their judgment. And so if you spread the referrals around, then you’re going to be, early on, referring to people who you don’t know their judgment.”

Patient Volume Management

The third impact of linchpin oncologist departures was the capacity to manage patient volume at their sites, although this was the least prominent impact . Nonetheless, participants noted that if a physician departed their network, oncologists would face significant challenges managing the patient load. Participants described limited redundancy in oncologist specialties, with most participants describing their schedules and patient loads as already tight, with sometimes significant waitlists.

One participant highlighted potential challenges from losing a provider:

“I think it would be a serious problem. I think the volume of patients that I see and the fact that it’s at an outreach site and that there’s not a lot of extra bandwidth among my colleagues would I think induce a lot of strain… Right now, for example, I was away last week at a conference and…. I’m really, really behind, just because no one else has the bandwidth to work while I’m gone.”

Theme 2: Adaptive Strategies in Response to Physician Departures

When a colleague becomes unavailable, participants described a range of strategies for ensuring continuity of patient care. One commonly discussed approach was to refer patients to external physicians who may be outside their immediate network. Another approach was to adjust triage practices to better manage caseloads and focus limited resources where they are most needed:

“We would be understaffed, so that means we would have to triage what we do better. …We would have to prioritize what are those cases that our urologic oncology group would do and which ones of those we’re outsourcing to community urology.”

Finally, participants noted that a generalist might be used in place of a specialist when the latter was unavailable. While this was generally regarded as sub-optimal, it was sometimes necessary to ensure that patients received timely care.

Theme 3: Unintended consequences of adaptive strategies

Finally, participants discussed unintended consequences on both care quality and patient experiences. The need to refer patients to external providers can disrupt established patient-provider relationships. This requires new providers to familiarize themselves with sometimes complex medical histories, which can be complicated due to patient record-sharing challenges. Participants noted that the absence of a physician can lead to delays at various stages, from initial decision-making to treatment scheduling and implementation. One participant described how losing a provider increased wait times for appointments:

“Right now I can see a new patient in 48 hours, right this second. But when we didn’t have a partner and our volumes were higher…it was like a week and a half. And do I think a week and a half makes a huge difference in breast cancer? Probably not, but it’s not ideal… The problem is there are some groups, even within my wider medical club group, that are already at two weeks. So, if they lost a provider, they go to four weeks. Well, now actually that’s a problem.”

While some delays may not significantly impact care quality, participants emphasized that for aggressive or late-stage cancers, timely treatment is essential. Shortages can necessitate adjustments to the standard order of cancer treatments, which can potentially lead to sub-optimal patient outcomes. As one participant noted:

“Anytime that there’s a potential delay in treatment, the foreseeable consequence of that would be compromising the patient’s chance at cure. And that’s concerning. A lot of what we do in this kind of paradigm involves sequencing of treatments… we could try to adapt by altering what is our institutional standard practices by sequencing them in different ways to buy us time for the remaining provider to see that patient. But it would [likely] result…in less than ideal cancer outcomes.”

Participants acknowledged that while they could often “make it work” by triaging, increasing workloads, or substituting generalists for specialists, these adaptations do not guarantee the same level of care. The loss of intellectual support within a multidisciplinary team with established workflows, combined with the challenges of managing increased patient volumes, were discussed as factors that could weaken treatment plans and diminish care quality, acknowledged candidly by one participant:

“I could pretend that I can deliver the same amount of care whether I’m seeing 10 patients or 20 patients. That’s not accurate. So, there is a detriment of care at some point. When exactly that happens, I don’t pretend to have a crystal ball, but it does happen..”

Another potential impact for patients is reduced access to clinical trials, as they are often championed or managed by specific team members. If these physicians become unavailable, access to clinical trials can decrease as remaining physicians may not have the expertise, time, or involvement in research to continue.

Geographical challenges in rural areas with limited redundancy in cancer providers can exacerbate the impact that shortages have on patients. Participants noted that patients in rural regions may face substantial travel burden and logistical difficulties when local providers are unavailable, further contributing to delays and disruptions in care.

Finally, participants acknowledged the additional strain on remaining oncologists. Reallocating effort to accommodate increased patient load, working longer hours, working outside their scope of practice, or losing trusted colleagues and collaborations can potentially lead to burnout, retention challenges, and less time for professional development, teaching or research. Reflecting on the challenges of a senior oncologist’s departure, one participant noted:

“Overnight, I lost the intellectual support of a way more senior person than me that I could discuss patients. It was very challenging. My clinic load went up by, I would say, 40% probably. I was constantly overbooked and double booked, and it wasn’t easy. I survived it, but I was exhausted….”

Discussion

Our study focused on oncologists’ perspectives on linchpin oncologists and shortages found that oncologists are well aware of and, in some cases, have experienced immediate impacts of cancer workforce shortages caused by the departure of a linchpin oncologist. Oncologist departures led to several adaptive strategies, each with the potential for negative unintended consequences to patients and remaining physicians, as illustrated in Figure 1.

Our study adds to prior qualitative studies that explore various aspects of the oncology workforce, including experiences, perceptions, and expectations regarding workplace dynamics. For instance, similar to our finding on the importance of collaborative relationships to facilitate efficiency, which found that general practitioners viewed their relationships with surgeons as critical in the decision about whom to refer for colorectal surgery and favored those relationships that facilitated faster referrals.¹⁸ Similarly, related to our finding on the importance of expertise, another study found that surgeon referrals for adjuvant treatment were influenced in part by informal consultations with medical and radiation oncologists who possess critical expertise.¹⁹ Building on this, we found that while increased patient volume and workload was a concern when considering oncologist colleagues leaving, participants talked more about the loss of specialized expertise or established relationships which are more difficult to replace quickly.

Our results can inform strategies to mitigate the possible negative consequences related to the adaptive strategies being used to cover the impacts of an oncologist departure. For example, areas that have less redundancy in their referral networks for specialists may benefit from virtual tumor boards with experienced physicians who can build trust and provide consistent expertise during workforce shortages. Or, if a departure leads to increased referrals to outside health systems, developing proactive processes for sharing clinical history can reduce the likelihood that delays will occur. Similarly, patients who may be at risk of experiencing treatment delays could benefit from additional navigation services during their course of care.^20,21 While some ramifications of workforce shortages are more challenging to address (e.g., increased workload), pairing adaptive strategies with additional tailored interventions or solutions can mitigate some of the negative consequences of oncologist departures. Future work examining the impacts of oncologist departures on patient care quality and outcomes are important next steps.

Limitations

Our study has some limitations. We were not able to recruit 5 participants from each site, and 2 of our sites were represented by individual perspectives. We also were limited in our ability to recruit oncologists practicing in community hospitals that were affiliated with the larger NCI cancer center health system. Having more of their perspective would bring nuance to how oncologist departures impact patients and remaining physicians in a hub-and-spoke model of care that is often observed in rural areas. Our results may not be generalizable to all rural settings; however, we achieved saturation in the three main themes.

Conclusions

Changes in referral and practice patterns following an oncologist departure may have implications for care quality. Targeted strategies that mitigate the negative consequences to patients and remaining physicians should be prioritized to alleviate the impacts of departures.

Context Summary.

Key objective:

What are the experiences with and adaptive strategies of oncology physicians practicing in networks with specialist scarcity when a colleague departs?

Knowledge generated:

Physician departures in networks with specialty scarcity can limit access to specialized expertise and disrupt collaborative relationships. In response, oncologists rely on myriad adaptive strategies that can have unintended downstream consequences such as delays, suboptimal treatment, or physician burnout.

Relevance:

Mapping the relationships between adaptive strategies following an oncologist departure with downstream consequences on clinical care can help guide resources to mitigate the negative effects of departures and shortages.